ForeThought

Dr. Vijay Bhatkar

2011-06-26T12:23:00.000-07:00

Apple and iCloud

2011-06-06T23:43:00.000-07:00

Apple and Steve jobs have been my favorite place and person. I have been following apple since the retun of Steve jobs and own 5 apple devices dating back to a decade. I am happy to see the iCloud announcement. This is a strategic move for apple to drive the cloud+appliance model that apple has so successfully built from the itunes/ipod days. The fact that it is harnessing appstore ecosystem and developers coupled with the "free" price tag will drive cloud adoption to the next level.

It is fundamentally changing definiton of Cloud from Application/Platforms/Infrastructure to a higher level abstraction of identity, information and will provide apple new insight into the content that is dear to people.

This will result in making apple smarter about identifying trends and designing features into the products they build making those products even more compelling.

The philosophy making it Simple, Secure and Scalable so that "it just works" is brilliant.

This is an inspitration to start something new to make it easier adopt cloud.

Lets me think of cloud solutions that are simple and effective to Indian consumers and businesses.

Balancing Regulation and Innovation for Cloud Computing in India

2011-05-18T02:36:00.001-07:00

Cloud computing is the most recent paradigm shift in Information technology. While it is not a result of any major technological breakthrough, Cloud Computing feeds on the advances in scalability and affordability of Power, Cooling, Security and Virtualization of Servers/Storage/Networking technologies.

The most comprehensive industry accepted definition of Cloud computing is from NIST “Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction “

The cornerstone of cloud computing is a new operational service model (SaaS, PaaS, IaaS) and new IT Infrastructure deployment model (Private, Community, Public and Hybrid Cloud)

The Prudence of Indian Government policies coupled with Industry enthusiasm have been fruitful to revolutionize information dissemination in past by encouraging massive adoption of
• Terrestrial, Cable and Satellite Television
• Wire line and Wireless Telecommunication
• Emerging Broadband and Internet adoption

Cloud Computing is a nascent Industry segment that needs nurturing and encouragement for India to emerge a global leader in this sunrise trend.
There massive need for cloud computing both for consumer and enterprise use.

Today 48% of Consumers of Mobile Phones are Internet users and use phones primary access devices. The data storage, application and security needs of these consumers are best addressed by cloud based information infrastructure.

The average spends on Information Technology as a percentage of Revenue is steadily increasing due to the pressures of global competition. The typical deployment model for IT infrastructure is capital intensive with long gestation. Cloud computing enables creation of burst able , on-demand infrastructure with a quicker provisioning and shorter time-to revenue. This allows the Industry leap frog from closet-held server directly to cloud and avoiding expensive captive datacenter buildouts.

There are handful data center and network providers that can deploy global scale coud computing platforms out of India.

The Power, HVAC and Security costs for creating massively scalable cloud computing farms are all but prohibitive in India.

There is need for government to create level playing on par with International Cloud deployments with appropriate regulatory measures for lower power tariffs, captive power plants and efficient cooling systems.

New Scale Acquisition by Cisco - Good For Tidal and LineSider

2011-03-29T15:56:00.000-07:00

Our Team(at Sify) has been fortunate to assocate ourselves with Tidal and newscale in building the first prototype of Cloud Services Delivery Framework Demo in Dubai

NewScale is definitely a well thought out service catalog and workflow environment with rich set of Webservices API's.

It is completely complementaty to TIDAL and LineSider as it solves a very real need to create and mange the product/service lifecycle in a Cloud computing environment.

It is a excellent choice for Integrating with VBLOCK/EMC IONIX for Baremetal Provisioing of Servers/Storage along with LineSider for Network Automation coupled Tidal Enterprise Orchestrater to realize most of the Public and Private Cloud Use cases.

It is a good move Cisco-Tidal, VCE/VBlock and the Cloud computing story from Cisco.

Blueprint for a new ITaaS Service Provider

2011-02-24T22:06:00.000-08:00

Cloud computing is raising the expectation on the agility, availability and access to sophisticated computing and communication platforms for small and medium business. Today most critical IT components for running a successful business are available as SaaS, PaaS or IaaS platforms. There is little value in recreating all these services from the scratch by a new age service provider.

It may be prudent to create a skeletal platform for essential services while taking a position of services broker to organize a rich portfolio by aggregating important services into tightly knit IT-as-a –Service offering.

The initial target market for such a service provider is Small and Medium Business. We will review aspirations and capabilities needed to serve the ever increasing expectations of small and medium customers from a modern service provider.

The Vision, Strategy and Execution plan to create a modern Information Technology as a Service (ITaaS) provider infrastructure is described on the following pages of this document.The design is driven by the needs of typical small and medium business.

Challenges of Small and Medium Customers

• Digitize, Organize and Analyze information flowing into customer organization with Cloud based Web 2.0, ERP, CRM, Content Management Applications and Technologies
• Design, Deploy and Manage Datacenter/Network/Cloud Infrastructure
• Train, Support and Enhance End user desktop/mobile computing experience

The goal of a new age service provider is to deliver world class services with highly talented Information Technology as a Service (ITaaS) Team.

Typical SME IT Mission:
The ITaaS Team supports Organization’s mission by providing a reliable, comprehensive information technology environment to enhance governance, transparency, and efficient business operations. The team encourages effective, innovative, and ethical uses of technology while assuring efficient use of university resources.

ITaaS Scope:
The ITaaS Team provides organization-wide leadership, management, and services in all areas of information technology. The team supports the use of technology for supporting:
• Software and Application services configuration and customization
• Data center, Network and cloud services administration
• End user Mobile and Desktop support

Typical IT Team Responsibilities:

Networking & Telecommunications
1. Responsible for design, engineering, installation, maintenance and repair of information transmission systems, including coordination with global networks and coordination of required licenses.
2. Responsible for management of the office network, including cable and port assignments, network addresses, phone numbers and user names; security, performance monitoring, traffic analysis and forecasting; and publishing the phone directory.
3. Responsible for providing telecommunications service and devices toclients, including consultation and service activation

Data Center
1. Responsible for management, administration and support of computing platforms, including file and print services, internal host-based computing systems and electronic mail services.
2. Responsible for capacity planning for existing and anticipated cloud systems infrastructure.
3. Responsible for system-level security on campus, through monitoring, threat avoidance and security countermeasures.

Application Support and management
1. Responsible for design, development and programming of custom applications, database applications, administrative reports and interfaces to enterprise applications.
2. Responsible for advanced statistical analysis, account generation and access privilege feeds to university systems.
3. Remotely Maintain and manage scanners, printers and allied equipment

Solution Strategy for ITaaS Provider will be followed...

Pari Networks - What does it mean for Cloud Automation

2011-01-26T15:06:00.000-08:00

There are three aspects to Cloud Automation:
1. Provisioning
2. Change Managment
3. Security/Compliance

Provisioning is typically across domains and is better enabled by strong service orchestration platforms that understand both the business(CRM/BSS etc) and the technology stack.
Tidal has a edge in this context with its strong background supporting Enterprise Software integration(SAP etc.) and Runbook automation use cases with its flagship product(Tidal Scheduler). Its recent focus on intelligent network automation using tidal enterprise orchestrator platform is paying off. However it's UI capability seems to be weaker than Linesider from a cloud worldview

LineSider addresses provisioning from access management and compliance perspective with decent user inteface and simple VM/appliance based delivery model for demonstrating technical provisioning capabilities.

Today Cisco announced the intention to buy Pari Networks. http://newsroom.cisco.com/dlls/2011/corp_012611.html

Pari has an appliance and software suite to collect and manage changes across Cisco device landscape. It seems to be a good managed services platform for network change managment.

The Angle that may or may not evident is the Cloud Automation capability that can be leveraged with the deep knowledgebase/metadata available in the Pari ecosystem. Pari with its capability to connect to and collect data from the most of Cisco devices has a potential to be a good network automation agent for Tidal. It is a very good start, as feeder for CMDB and as an equivalent to BMC - Bladelogic Network Automation(BBNA) suite. It looks like this platform can address the change managment aspect of cloud environments very well. It fill the key void of resource pools related to licensing and support with its proven smartnet integration.

There is a potential for this Pari appliance to be integrated with Tidal Orchestration for realizing quick and effective cloud provisioning platform.

LineSider is good approach for compliance with policy based networking(using LDAP) where as Pari approach to compliance is more on plumbing of configuration collection by replaying user actions. There is clearly a small overlap between Tidal, LineSider and Pari.

It may not be difficult to reconcile the interfaces and come up with stable systems integration approach that encompases all three cisco products based on the scale and complexity of the assignment. They solve different facets of cloud automation.

cool stuff....

2011-01-12T15:47:00.001-08:00

http://www.mckinseyquarterly.com/Media_Entertainment/Publishing/Clouds_big_data_and_smart_assets_Ten_tech-enabled_business_trends_to_watch_2647

LTE - Pervasive information instructure - Presence, Profile and Provisioning maturity.

2010-12-25T03:54:00.000-08:00

I have been working for a new age service provider for over 3 years. Since last 2 years, I am part of the team responding to RFP's for modernizing major service provider networks including staging, testing and deployment of 4G/LTE services.I have been following the 4G/LTE trends at TMForum and other related information sources since then.

History
========

In the mid-70's Telecom networks underwent a important change, they introduced the common channel signalling to break-away from physical network path to virtualizing the call setup. This was enabled by a relatively open standard called Signaling System No. 7 (SS7). According to Wikipedia - "SS7 is a set of telephony signaling protocols which are used to set up most of the world's public switched telephone network telephone calls. The main purpose is to set up and tear down telephone calls. Other uses include number translation, prepaid billing mechanisms, short message service (SMS), and a variety of other mass market services."(Wikipedia )

The SS7 standard allowed the cutilvation of third parties to develop open extensable frameworks for voice deployment. SS7 helped Telecom companies to innovate faster into SMS and later into 2G/3G data services. The current LTE/4G mobile network evolution is looking more and more like the early part of Internet revolution that the data networks went through in the early 90's. The emergence on authentication, authorization,secure connectivity, identity managment, naming and directory services(LDAP,SSL,Radius etc) in this period was primarily geared towards richer, more secure, personalized services to the consumers on the Internet.

In the late 90's, These innovations were leveraged by consumer internet services like yahoo, hotmail and others. Pioneering intranet platforms such as NetScape SuiteSpot and Microsoft Site Server/MCIS followed by carrier class email platforms from Openwave and MiraPoint leveraged these innovations(specially LDAP effectively).These products were quickly adopted by some service providers for public facing internet services.

In early 2000, Netscape and Openwave's Service provider success was closely followed later by Sun’s Java Advanced Intelligent Networks and Microsoft Active Operations Support Systems (OSS)/MCIS. These attempts that tried take advantage for earlier innovations in Telecom networks and marry them with the Internet technologies to influence the then emerging field of Advanced intelligent networks(AIN) failed.

From 2005 to 2010, Apple and Google have proven the relavance of cloud/application Services and thier positive impact on revenues for service providers. Apple with its secure, easy to use and elegant appliances(iPad, iPod Touch and iPhone) and consistent push towards content(music, tv and movies) coupled with rich applications created iTunes. iTunes has helped fortyifing customers for both itself and AT&T. Google has taken on the challenge to create rich and open platforms(Andriod and Chrome) for helping all other service provider in getting a byte out of the application enriched mobile world. Google is mildly successful.

Road ahead
==========

In Persuing Quest for learning more about Pervasive Information Infrastructure, I had accidental oppurtunity to visit LTE Lab of a major service provider.

The public domain use cases being pursued on 4G-LTE were very interesting:
- Digital Signage - immersive customer service
- IPTV
- Triple Play (Quadraple play - ??? TV, Internet, mobile and cloud based applications)

These 4G-LTE use cases relied on and en-riched by geo-Location, advanced identity management and personalization. These features are provided by default on most Apple, Google and Microsoft operating system based mobile devices. The future is driven by richer experiences on thinner devices powered by the cloud services.

The Mantra of profile, presence and provisioning is gaining momentum as strategy for service providers. The un-paralleled success of iTunes as a unified profile management platform to provide commercial value by brokering application and content delivery using both internet and telecom networks. The Software-as-a-Service(SaaS) leaders such as SalesForce are persuing platform as a service(Force.com - PaaS) route to reach the consumers. Google, Microsoft and Oracle are attempting to provide integrated knowledge worker applications for one flat fee per month leveraging a common profile.This has prompted service providers such as AT&T to create cloud service brokerage portals(AppStores). All these trends reinforce importance of strong, open, federated profile. Profile and identity management based on open standards and/or published API's (LDAP, SAML,OAuth etc)ky to success of service providers in attracting software development community to embrace thier platforms/networks.

VoIP on LTE is emerging as a default approach to voice services followed closely by increasing use of Telepresence(Webex, Facetime and Skype).this trend is increasing the importance presence services. The presence enabling protocols such as SIP, SIMPLE and XMPP more relavant today. It is time to build presence aware applications for simplifying provisioing,metering and billing for application/network usage.

Service providers need to acquire customers Fast. This has to be done in a Correct manner to reduce/eliminate errors and to keep the customers Happy . The most important aspects of process efficiency is to identify the system dependencies and limitations in provisioning. This situation can be eased by implementing centralized profile management approch for devices such as CMDB(configuration managment database as defined in IT service Managment) and Configuration/inventory management (as defined in TMN). The mutiplicity of individual systems and connected networks is increasing the complexity of provisioning. The is complexity on user side can be reduced by leveraging common profile and active use of presence services.The active utilization of master data(profile and CMDB) and state information(presence and montoring) increases provision efficiency.

Call for Action
===============

Profile Management - The actions items for LTE equipment/application providers is - where ever possible, keep the option open to use protocols such as Diameter/LDAP/SAML for authentication, profile lookup, PCRF(Policy and Charging Rules Function) and PCEF(Policy and Charging Enforcement Function) of LTE(S6a and S7). Ensure there is a profile managment strategies(such as CMDB/inventory managment etc)for hardware and software resources in use, it is important expose the device/service profile information in a open standard.

Presence - Align the presence functions between a common set in protocols such as SIP, SIMPLE and XMPP to the LTE_Attached, LTE_Active and LTE_Idle states. Encourage the use of these functions and related dependencies for provisioning, Call Setup and Collaboration. There is a need to asses the impact and add it as key factor in the use cases for LTE applications.

Provisioning - The success of LTE is based on quicker adoption richer cloud services. Open protocols and standards for provisioning such SPML and use of Dynamic orchestration platforms that can interact with Profile and Presence Services are essential to Cloud Services Delivery Framework (CSDF) implementation.

Innovative use of profile,presence and provisioning enables fast and correct process to acquire happy customers for service providers.

vBlock and FlexPod - L3 capable Access network and Non-IT Automation is need of the hour.

2010-12-22T22:10:00.000-08:00

Our team had the good fortune of finishing the first vBlock deployment today as a Lab environment of a major Service Provider.

Without Prior knowledge of Flexpod, we suggested adding dedicated Nexus Nexus 5500 based Access layer with L3 capability to vBlock for completeness of their Private Cloud Requirement.

Incidentally the Launch of FlexPod today with integrated Nexus 5500/1010 along with UCS and Netapp 3000 Series came as a good surprise and Qualified our approach in terms of Bill of Material for Private cloud.

We still donot see any L3 capabilities, Security and Application networking built-in.

The Value of Private cloud deployments is the power of simplicity through stack standardization and while maitainting the performance, reliability and security of a vertically integrated infrastructure.

Ideally intelligent non-it components such as switched pdu's, mini-VESDA and sensor networks need to be integrated in the hardware layer.

The orchestration and management stack needs leverage the abilities of non-it and IT infrastructure elements in creatively reorganizing Data center energy, security and workload distribution footprint.

The objective is to focus gains in productivity across the both the system/network administration and the user community.

It needs to enable meeting business goals quicker, at a predictable cost, in a replicatable manner.

Foundation for Pervasive Information Infrastructure

2010-12-07T20:50:00.000-08:00

The Traditional Definition of information in the Enterprise is changing from simple unstructured and Structured ASCII Data to more broader binary formats.

In the era where almost every business class mobile device is sensor-enabled, connected and Video capable. This popularity of embedding complex sensors, displays and end-effectors in to mobile devices is creating new and interesting Data collection and anlaysis use cases.

The proliferation of sensor-enabled data-rich, video and voice capabilities into the business workflows is only a matter of time.

Before too long we will see emergence of voice and video streaming functions becoming main stream information source. Transcribing and annotating audio/video streams enable easy indexing and search capabilities. The Data corellated from embedded sensors and sensor network will add a dimension to voice and video data to make it context and location aware.

The Broadcast(like youTube), multi-cast(like netflix) and conference(webex) capabilities are becoming standard features in Enterprise class collaborative applications.

In a world where more rich-information is produced, consumed and discarded, the Design of the pervasive Information Infrastructure needs to be more real-time, flexible, scalable and secure.

Presence(instant messaging) capabilities provide immediacy and context , Profile capabilities capture end-user meta data and provisioning capabilities simplify service adoption.

The Foundation for creating a pervasive instructure relies on Presence, Profile and Provisioning maturity.

Policy Based Network Automation

2010-12-01T15:21:00.000-08:00

Cisco's Announcement to acquire Line sider is interesting to say least.

The Approach of dynamically creating VLAN based on User Roles and tearing them down is a very interesting use case for Network Automation, this idea of very powerful when you look at the big picture of cloud workloads from virtual machine perspective.

A better use case for Policy based Networks is to identify Virtual machine with a Certificate and Mac address on wakeup and by doing do the policy lookup/LDAP magic, the associated Nexus 1000v port can be configured to belong to the correct vlan.

All in all it is nice story to follow...

LineSider - Cisco Acquisition

Emerging Cloud Market places.

2010-11-10T19:12:00.000-08:00

The Cloud market places are emerging for buying and Selling Capacity of IaaS platforms.
http://www.spotcloud.com/Technology.5.0.html
I am impressed by AT&T Renewed focus on promoting vertical third party SaaS apps from its portal.

Salesforce.com has strong ISV community on Force.com - PaaS and has started pitching for migration of Lotus, Sharepoint and other custom Apps into Force.com platform.

This Cloud Marketplace trend is a welcome change in accelarating the adoption of cloud services.

Private Cloud Journey...

2010-09-24T02:04:00.001-07:00

Private Cloud is a secure, stable, scalable and simple shared services model deployed within enterprise network boundary adhering to cloud computing principles.

Primary Goal of Migration to Private Cloud is Seamless transfer of Application Services from Current Data Center to the new facilities built using Shared Services Architecture for Private cloud.

AppsCloud – A Methodology for migrating to Private Cloud

The Emphasis of this methodology is on:

Understanding application workload Behavior

Ensure Security

Leverage technology innovations

Adherence to Cloud Computing Principles

The Goal is to de-risk the datacenter migration, reduce disruption due to change and simplify the path to Private cloud adoption.

Understanding Application Workloads

Application Infrastructure in a large organization is inherently difficult to comprehend as a single entity due to diversity of Application workloads (Email, Database, Data warehousing, ERP, Analytics etc). Deployment complexity coupled with primary and multi-level interdependencies, the resulting emergent behavior presents an even greater challenge in understanding.

Data Center Migration Methodology

The Conventional Approach of Datacenter Migration

Rebuild the Physical, Infrastructure

Deploy applications,

Perform Data migration

Test Application level integrity and Scalability

Cut-over / Go live

This Approach is ridden with risks, uncertainties and a fairly long re-deployment cycle inhabiting the move to Private cloud.

Most of the Unix environment and some mission critical Windows servers (Oracle Database other business critical apps) are required to be migrated in the above approach.

The Windows workloads are more amenable to Virtualization/Private Cloud and Unix workloads are being rapidly converted to various supported variants of Linux for being virtualization friendly.

The several core technology innovations driving this transformation increasing adoption of Private Cloud by enabling features such as redundancy, multi-tenancy and high availability. These features help to realize the shared services architecture and to de-risk the transformation

Core technologies enabling the Private Cloud

The Popular tools and Core technologies utilized in realizing the goals of the Private architecture are:

Service Orchestration

Self-Service Portals

Monitoring/Management and Reporting Platforms

Master Data Management (CMDB/Asset/Inventory management)

VMware Fault Tolerance

VMware HA

vCenter Heartbeat

vMotion

Storage vMotion

vShield Manager built-in backup

EtherChannel

vPC

Device/Link Redundancy

UCS Dual Fabric Redundancy

Migrating to Private Cloud

The Most important drivers to Private Cloud migration is a need for optimizing the cost, improving manageability, increasing scalability and ensuring security. Moving from discrete server infrastructure to a virtualized share services model seems to meet most of the Cloud compute objectives of pay-per-use/pay-as-you-grow, on-demand scalability and easier management.

The key to developing a robust Private Cloud design is clearly defining the requirements and applying a proven methodology and design principles.

Introducing the key requirements for secure, stable, scalable and simple shared services Private Cloud architecture...

The security requirements for Private Cloud defined in terms of Confidentiality, Integrity and Availability pillars for the Shared services Architecture:

Isolation

Isolation or Secure Separation ensures meeting Confidentiality objective where one application/tenant does not have access to another application/tenant's resources, such as virtual machine (VM), network bandwidth, and storage. Each tenant must be securely separated using techniques such as access control, VLAN segmentation, and (If possible) virtual storage controllers. Also, each layer has its own means of enforcing policies that help reinforce the policies of the adjacent layers.

Service Assurance

Integrity of the environment is maintained by Service Assurance techniques. Service Quality is ensured with isolated compute, network, and storage in concert with SLA driven service delivery, performance monitoring and incident management during both steady state and non-steady state. For example, the network can provide each tenant with a certain bandwidth guarantee using Quality of Service (QoS), resource pools within VMware help balance and guarantee CPU and memory resources, while Hitach with its active-active architecture can balance resource contention across storage volumes.

High Availability

The focus on High Availability is reflected in the design of infrastructure to meet the expectation of compute, network, and storage to always be available even in the event of failure. Like the Isolation/Secure Separation pillar, each layer has its own manner of providing a high availability configuration that works seamlessly with adjacent layers.

Stable and Consistent Management

Technology innovations at the platform level are required to consistently manage, rapidly provision resources with clear visibility into resource availability. In some of our smaller deployments in its current form, each layer is managed by vCenter, UCS Manager, Nagios/Munin, Cisco Data Center Network Manager and Hitachi Management tools, respectively. We are collecting analyzing information from these tools to making capacity planning and provisioning decisions.

Need for Data center Intelligence for Private Cloud

As the environment expands in thousands of workloads and spanning multiple data centers capacity planning tools such as VMware capacity planner fall short in assessing the big picture requirements such as Network Topology, Non-IT resources (Power/HVAC) optimization, Business Continuity and Workload distribution.

There is a need for Data Center Intelligence tools such as Tideway(BMC) and Cirba to assist in decision making process. These data center intelligence products gather information by automated discovery processes and integration into the management /monitoring tools. They connect to systems of Record such a Asset management Database/ CMDB , ERP and CRM systems and correlate the northbound business data with the southbound intelligence on infrastructure to assist in quick and informed decision making

Data Center Migration techniques for Private Cloud

A Data Center Migration can leverage industry standard Datacenter intelligence and virtualization tools and methodologies for Capacity planning, Server Consolidation and Physical to Virtual conversion. This approach helps to de-risk the application redeployment complexity and quicken migration process to Private Cloud

Simple Datacenter Migration Approach for Private Cloud

1. Capacity Planning using Cirba, Tideway or VMware capacity planner

2. Physical to Virtual conversion using VMWARE converter

3. Multi-site Cluster with Live Migration - using Symantec/VCS and/or VMware SRM

Path to Cloud – Road Ahead

Virtualization ensures portability of the application infrastructure, the easy migration across datacenters.

The benefits can be realized with the advances in Server, Access and Virtualization Technologies.

Cisco is leading the pack by creating a Transparent Virtualized Shared Services Environment across multiple datacenters(Cisco UCS/Nexus), Other Server vendors(IBM/HP/Dell etc) are building Cloud-in-a-box solution with Clustered Servers(infiniband/10g Ethernet) and Storage, Storage Vendors (EMC/Netapp/Hitachi/HP) provide multi-tenant storage frameworks with remote replication capability, Boutique appliances are focusing on the other critical pieces of the puzzle such as Security(Hytrust/RSA), WAN optimization(Riverbed/Cisco WaaS) and Scalability(F5/Citrix)

These developments are paving way for easier and more affordable path to Private Cloud.

Building a Solid Cloud Team

2010-07-12T23:02:00.000-07:00

Cloud computing is loosely used to describe the convergence of servers, storage, network unified with a common automation framework. In order to create a sustainable and successful team, there is a need to invest into a SOLID cloud team with clear objectives of creating a SIMPLE, STABLE, NIMBLE and CAPABLE Engineering Leadership.
Service Orchestration, Logistics, Infrastructure Design & Deployment (SOLID²) Cloud Team

Server Technology Automation & Build Lead Engineer (STABLE)
Storage Infrastructure Management Practice Lead Engineer (SIMPLE)
Network Infrastructure Management & Build Lead Engineer (NIMBLE)
Capacity Availability Performance Automation & Build Lead Engineer (CAPABLE)

This could be winning team.... on lighter vien...

Data Center 3.0 - Unified Information Infrastructure

2010-06-30T20:46:00.000-07:00

The Management challenges, Eco-Consciousness and Escalating costs are driving the Technology providers to move from discrete solutions for Compute, Storage and networking to Unified Information Infrastructure.

The Unified Information Infrastructure as a framework builds on Unified Compute, Unified Network, and Unified Storage. There is a need to incorporate the management and security services to complete the Data Center 3.0 requirements.

The Advancements in network technologies and Standards (10 G Ethernet, FCoE, IPMI etc) are fuelling the Vision of a Unified Network. It is practical to unify the management traffic, inter-process (cluster)-communication- IPC, Storage area network (SAN) and Local area network-LAN traffic onto a single unified fabric. Some solutions are driving this to the realm of reality today.

With the advent of in Blade server technology and Virtualization at the access network layer for these solutions, there is a clear shift toward Unifieds Compute. Unified Compute consists of Servers, first level i/o virtualization for IPC, LAN and SAN along with a integrated management stack
Unified Storage is emerging from just providing LUN’s and dedicated Front end ports to providing Virtual Storage contexts (Netapp vFilers), Storage gateways( for front-ending desperate storage solutions) and Software storage arrays spanning across virtualized platforms.

There is a unification of NAS, SAN and Backup (Virtual Tape) Devices at Storage controller and Storage network level.

This is a logical framework for management of physical devices and advanced data center support services along with a set of key considerations that one has to be aware of in making informed choices in during various stages of maturity of the Unified Information Infrastructure.

Building a Cloud Computing Lab

2010-06-22T06:33:00.000-07:00

Cloud adopters should consider the following key elements when implementing Cloud service delivery frameworks:

Leverage existing infrastructure within the prospective Cloud providers.
Use a graduated adoption from semi-automated to fully automated processes.
Balance between customization and standardization while making an effort to have differentiated services.

In order to help cloud adopters realize the value of this approach, It is better to walk the talk....A Cloud Lab can be created and evolved in the same manner as a Production Cloud deployment.

As a Physical Manifestion of a cloud, there are three critical components

Core Network and Shared Data Center
Workload Execution Engines
Resource Pool Management, Securty, SLA Analytics and Automation Infrastructure (on-going development/enhancement roadmap)

A Practical Deployment plan for

Core Data Center Network Services is as follows:

Core Network Services such as CRS, Nexus 7k
Cat 6k Services complex with Service blades for Firewall/Loadbalancers/SSL Accelaration/Wide Area Accelaration etc),
Storage-as-a-service(MDS with EMC VMAX-FC, Clariion-iSCSI, EMC-DataDomain(VTL/Dedup), may be even NetApp-NAS)

The end result is a creation and demonstration of Multi-Tenant PaaS SKU's for Routing/VRF, Firewall Contexts, Storage, Backup and ACE/WaaS contexts to supplement the core IaaS SKU's

Workload Execution Engines:

A VCE - vBlock 1(VMWARE, UCS B-Series + EMC Clarrion)
Secure Multi Tenancy(Microsoft, UCS-C series + 5k +Netapp) - Servers Storage and Access Network
This can complemented with Sample Workloads such as Unified Messaging/Collaboration(MS-Exchange/Cisco Unity/Call Manager/Sharepoint/Webex), ERP (SAP/Oracle) and VDI (each one taken as mini-deployment to simulate a customer on the Cloud)

The end goal is Creation and Demonstration of Core IaaS SKU's for Compute, Storage and Application Bundles(Basic SaaS SKU's)

Resource Pool Management, Security, SLA Analytics and Automation Infrastructure:

There is a need to create both in-band patform with Web/ReST based tools and out-of-band platforms with SNMP/IPMI and Native Command Line Interfaces for enhanced security and control of Below-the-line Activities.
Profile and Identity Services - LDAP-Active Directory/Raduis/Profile Server(SPML) are the key to tie all applications/infrastructure elements together with Access Control and Authentication.
The Components for Automation Platform are Tidal Enteprise Orchastrator for Service Orchestration and System-wide Semi-Automatic Job Execution and Tracking using Business Support Portal Front-end
XMPP based Presense service for Assessing Capacity/Performance and Availability of to existing Resource Pools and managment of Resource Pools using native tools like VMWARE Vcenter, EMC Control Center and UCS manager
Master Data Managment of the Entire Environment is Maintained in BMC CMDB with Business Support portal to Enable Semi-Automatic Tasks (Moves/Adds/Changes)
The Service Catalog Management and simpler BSS functions are handled using Open Source/Sify developed tools jBilling/NMS/Helpdesk
SLA Analytics and Reporting Platform using Crystal Reports(now from SAP)
Log Consolidation using SPLUNK and Profile Services

This environment needs to drive home the Completeness of Vison, Strategy and Execution to meet the Customer objectives of:

Security
Compliance
On-demand services / increased customer stickiness
Reduced TCO
Time-to-provision

VMWARE Certification

2010-06-21T17:10:00.000-07:00

It is nice go through the grind once again for this VCP certification...
Now I am VSP, VTSP, VCP... Don't mean to be bragging here... :-)
Quoting from Frost "Miles to go before I sleep"...
Cisco UCS certification is on the horizon...
Watch out...

UCS - installation fun...

2010-06-09T18:24:00.000-07:00

We are installing UCS...It is probably the easiest blade system to bring it into the network

In a UCS system - Upstream ---> UCS Manager/Fabric Interconnect ---> Fabric Extender in each Blade Chasis ---> Mez card on Server Blade
Connect to the Interconnect through serial cable and configure Managment IP address by completing the setup steps...
http://www.cisco.com/en/US/docs/unified_computing/ucs/sw/cli/config/guide/b_CLI_Config_Guide.html

We need to connect to Mgmt0 port - of the Fabric interconnect to Mgmgt net and access

it throught the IP set to the Management port

Once you are able to launch the gui... it is familiar territory
http://www.cisco.com/en/US/docs/unified_computing/ucs/sw/gui/config/guide/GUI_Config_Guide.pdf
The Rest of the setup can be very specific to your special needs...

The simplicity of UCS is refreshing...

Innovation for the bottom of the pyramid...

2010-05-27T06:21:00.000-07:00

Ideas are are like channels on Radio/TV. They are on air 24/7. All you need is the sensitivity to tune in to them. Once in tune encourage the torch bearers. Walk with them for couple of rounds of execution.

This is a story of innovation that impacts the bottom of the pyramid. Our team is proud to be part of simple yet critical innovation that continues to improve healthcare for millions of below poverty line patients.

About One and half year ago...We had opportunity to work with a organization called aarogysri (www. aarogyasri.org). We had won a contract to provide a email platform for this initiative.

Providing a email platform was nothing more than a product sale, however we started engaging with different stake holders and over casual conversations.

We understood the Business Model where Aaroagyasri trust(funded by government) pays a contracted amount to third party insurance company on behalf of below poverty line family, each hospital provides services to patients with the Aarogyssri card, aarogyasri provides pre-auth, procedure is done and hospital gets paid by insurance company.

The Chief of Aarogaysri expressed the inordinate delays and lack of control on quality of care provided by the hospitals. He had already established a track record of being IT savvy by introducing manned kiosks in each of the hospitals and sophisticated application to automate various steps of pre-auth, treatment and payment procedures.

The current mode of communicating with the hospitals was to send paper circular, memo or notification and wait for the hospital to respond with in statutory amount of time and then act on the corrective or punitive measures. This mean that if there was a erring hospital there was no way to stop them continuing the errors in a reasonable timeframe.

We tuned into the chief’s idea and customized the Web front-end of the email system to incorporate the circular, memo or notification functionality with mail system itself.

We had created:

1. Single-Sign-on(SAML based) with the current aarogysri portal

2. Created ReST based Web Services for Enabling and Disabling specific functions such as pre-auth based on responses to memo and notification.

3. Incorporated neccecary changes to the web front-end to execute the workflow with proper time-stamps and auditable history.

With the introduction of the new system, the erring hospitals were bound respond to the email based memo’s arrogyasri immediately otherwise the workflow would disable them from admitting new patients.

In the end this resulted closed loop communication the health care providers and aarogyasri resulting in better faster processes.

To email system aarogyasri expanded from few officers (200) to entire eco-system healthcare providers, supporting insurance providers and all employees of aarogyssri (5000+ in total).

We are proud to be apart Aarogyasri’s innovation story…

This taught us a few things on technology front which were eventually rolled in to the cloud strategy...

We believe that the Software effort of Aarogysri can be converted into a Cloud platform offered to other states in the country....

Cloud Services Marketplace

2010-05-25T04:42:00.000-07:00

One of my friends Karthik and I were talking about Cloud Marketplaces and his boss/my friend Raj Buyya (www.buyya.com) has written academic view on Cloud Market place.

I had worldview that was grounded in reality of today's state of the art in cloud technology. In

Today's Cloud Providers are very weak in providing B2b interfaces, there is no single window for:

Uniform API's or Standards for Provisioning for SaaS, PaaS and IaaS

Monitoring and Management frameworks

Delegated Administration portals or mature Reseller Relationships

Model for ISV relationships for Pay-per-use deployment of Applications

Model for Engaging Consulting and SME' resources

Model for Management platforms and Managed Service Providers

Promotion and Adoption models -

Monitization of Assets in Muliple Business Models(Free, Fremium, Ad-Supported, Premium, Concierage)

Training and Certification

Federated Identity Management

Comprehensive Services Catalog

A Practical approach is to Create a Cloud services delivery intermediary that does the Service Orchestration for Cloud Services and enable realization of a vibrant marketplace where customers can do comparision shopping, enjoy ad-supported goodies, select providers for whatever type of level of service they need.

more later...

Sify Reaching for the Clouds...

2010-05-23T02:29:00.000-07:00

The Cloud Announcement of Sify did not come as a big bang one shot deal...It has lot to do with the Sound strategy, Business acumen and Persistent Execution of our Business Team.

At the the outset there is a lot of hardwork and confidence of various technical teams to pull it off...There are several things that were done on the technology camp(Network, Storage, Servers and Data Center teams) for over two years that culminated into facing this deployment with conviction, courage and grace...

The Major emphasis in cloud computing environments on how to execute the key processes such as Fulfilment(order processing and Provisioning), Assurance(monitoring, managment & SLO/SLA's) and Billing (usage reporting and Billing) with as as much automation as possible. These process famework has been established with our early SaaS initiatives.

The First forray in to building a SaaS platform came about somewhere around March 2007. This was just before I joined sify... the leadership saw the value in going after creating SaaS platform for multiple key applications such as E-Mail, Online Testing, Supply chain Managment and Security operation center. Messaging/Email was one of the first one...Our virtual team was lucky ot be assigned this job...There were several platforms evaluated and decision was zeroed down to Homegrown, Zimbra and Microsoft Exchange. I joined Sify in July 2007.

Hosted Messaging and Collaboration (Microsoft Exchange) was decided to be first SaaS platform. The decsion was to go and create Standardized hardware platform for internal and external deployment of Messaging for both Sify.com and SaaS platforms.

The initial deployment was around shared services architecture. This was little ahead of time with its emphasis
Servers: HP BL460 (50+ blades
Network: Cisco Catalyst 37xx - Gigabit Copper as a Unified Fabric for IP an iSCSI
Storage: EMC ClariioonCX-3-20, 3-40 and AX150 - All iSCSI deployment - HP-EVA(100+ TB)
Security Platform: Fortigate, Cisco Ironport
Application Network Services: Cisco CSS load balancer
Automation Platform : Ensim

After initial hiccups with deployment partner. We got our key people trained on Microsoft HMC platform in Singapore. This helped us deploy the HMC platform in-house expertise. The Study of HMC architecture and our understanding/ability to leverage Virtualization helped go against the grain and reduced the total server count by a double digit number. This architecture was validated by the Microsoft Professional Servcices team. We have seen that this approach had direct bearing on getting the platform profitable sooner due to lower platform costs.

The Bold deployment of Physical infrastructure in couple of datacentres helped us hone our remote management capabilites

We have realized the importance of launching smaller disaggregated services before bundling them together. the first disaggreated service that was launched was Anti-Spam Service based on Cisco Ironport, We built a Provisioning and Delegated administration portal for this service. later There were few more services the complemented the SaaS platform such as Mail-on-Mobile based on Synchronica and Blackberry, SharePoint etc. The lesson was to be judicious about build vs buy decisions.

In our case a fully built up server farms enabled faster execution on other customer requirements/projects . The live testament to this success is in deployment of projects such as Complaince email solution for close to a million users, sify.com portal, Storage as a Service PoC.

One of the lessons learned was to add more services (such as hosted CRM, OCS etc). We initialially implemeted CRM with very little off-take.

It is however difficult drive market adoption high value services-- as the complexity increases, it keeps getting better proposition to buy the automation platforms. However It is important to build the glue code/ components/configuration expertise that can emerge as unique differentors.

We are invested heavily into creating the body of knowledge for automating the cloud platform. Most Vendors have limited world view of the cloud computing - generaly restricted to virtual maching provisioning and rudimentery vLAN setup at distributed switch level.

Our Roadmap is a go beyond the current state-of-the art interms of cloud computing into the future of with virtual data ceneters. Our network team has thought through the processes and hand offs neccecary to realize the vision of virtual Data centers with not just a VM and some storage.

The Virtual Private Data Centers are technically a extention of customer datacenter over MPLS VPN. This is a post cloud scenario of appliance + cloud where CPE in MPLS cutomer premises as well as the muti-tenency at every level (access, aggregation and core).

The Class of Vendors for such broad automation objective are from Job Sceheduling and run-book automation camp. The key players in the VDC automation are Tidel(Cisco), Opalis(Microsoft), IBM and Control-M(BMC)...

The Sify iONi product lime is getting a makeover to handle portal interaction between users and workflows...

Master data Managment and Aggregation using standards based approcah such as RSS feeds

The Assurance activity needs to rechristened home grown monitoring and managment along judicious usage of opensource/vendor provided tools.

The Billing integration is a medium term automation candidate based on scale numbers.

Our goal is to enhance the current toolset to reflect our relavance as cloud platform developer by building upon the capabilities provided by OEMs/Management platform vendors. It is important to develop, deploy, provision and explore the ability to provide umbrella Cloud Services Management Framework(CSDF) to accomplish the objectives set out by the business leaders for increasing the yield per square feet of datacenter space and value of the per MB of Bandwidth sold.

Cloud API's

2010-05-17T14:10:00.000-07:00

Amozon EC2 is probably the most widely adopted User Centric API for cloud services. Google, SalesForce and other Providers expose api for transfer users and information into their own cloud platform. There are approaches such as simple cloud api from Zend to focus on helping the development of cloud native (php) applications using resources available in the public cloud.

VMware created a comprehensive vCloud API to cover both user centric and provider centric functions of the cloud.

vCloud API provides the necessary abstraction for automation of subscription/management activities by medium to large enterprises and forming inter-cloud relationships. The User centric API elements available in vCloud API include:

Inventory Listing
Catalog Management
Provisioning
Application Configuration

In addition to the above, The API includes Service Provider activities for managing Organizations, Users, Roles and Product Catalogs.

There is a need for open standards to evolve and cover SaaS, PaaS and IaaS Cloud deployments with active contributions from Technology Providers, Service Providers and Cloud Consumers.

S.A.M.E (Simple, Agile, Mobile, Efficient) CLOUD for Better IT Services

2010-02-27T19:18:00.000-08:00

There are active Cloud computing initiatives by several service providers. This is article is an attempt to address the pains of jumpstarting, Ramping up up such initiatives along with providing a long term roadmap for sustainable management of large muti-site cloud computing infrastructures.

Most of the Service Provider Environments have a well defined Business Support Systems-BSS(Customer Portal, Service Delivery and Rating/Billing Systems. The Operation Support systems-OSS for Service Resource Management/inventory/CMDB, Customer Account/Profile/Identity management and Equipment Provisioning Systems. These systems are generally optimized to handle Network services and Hosted Services.

Radius and Cloud Computing

2010-02-05T18:48:00.000-08:00

The Service Provider Economics have been driven by RADIUS from the early days of Dialup internet services.

Over the years, RADIUS gained prominence as vital standard supported almost every networking device for Authentication, Authorization and Audit.

There are Many Successful deployments of RAIDUS as enabler of Billing and Reconciliation.

Good RADIUS article

VDI for Video delivery

2010-02-02T21:30:00.000-08:00

I had a chance to meet Senior Telcom Executives. There was a healthy discussion about the first/last mile challenges.

The conversation lead to challenges faced by ISP's due to growth Video streaming and download traffic.

The Problem statement:
- Bandwidth consumption of on Conumer network due to video traffic both due to streaming and download

Detailed Description:
There was a discussion about how consumer business is negetively impacted due to the popularity of Online Video. Added to this, The Content providers are not willing to colocate the content with the SP. there are consumer who consume gigabytes of bandwidth from residential customers due video download services such as netflix.

Technical background:

HD Video takes anywhere between 1 to 20 Mbps bandwidth for streaming the video(MP4,MPEG2, SD, HD) . the current experience is jerky with wait times to accomadate for the download capability and congetion. The video rental sites have a different behevior due to the need to downbload to rights managment appliance(roku/PS3/WII) to be played on tv.

Proposed Solution:

The Advent of Virtual Desktop infrastrucutre is gaining popularity in the enterprise. the is huge body of knowledge being built to address Bandwidth Video quality and audio quality issues.
There Several Remote Display Protocols such as:
1. RDP - 7 from Microsoft
2. ICA - Citrix
3. PC-over-IP - TeraDici

Promising but lesser known products:
- Wyse TCX extension for RDP
- Quest RDP optimization Pack
- HP RGS standard.
Most of these protocols take under 100+KB to 1 MB.
Due to the low bandwidth requirements we are proposing a Cloud based VDI instance for Broadband subscribers to allow them download movies or stram movies using a VDI instance.
This in turn means the service provider is able to terminate HD on a Free Cloud based VDI and deliver it using VDI stream protocol.

This approach can be using anyone of the above methods.
-As the adoption increases, the Downstream last mile network bandwidth pressures are reduced due to lower predictable traffic profile which can further be optimized by traffic optimization solutions

- Centralizing and containing Video traffic of the Content Provider traffic to single cloud environment compells content provider co-locate to Service network to control cost and complexity.

- Influencing Road map thin client vendors to incorporate newer optimized protocols to provision them as client access nodes in consumer networks.

After some testing we came to a realization that VDI approach better for downloads(like iTunes - Movies/podcasts or P2P downloads) than the streaming(Youtube-Flash or Netflix-silverlight)

Conclusion:
Although there are newer video standards, VDI based Video Standards are simpler to impement, easier to scale and secure to operate the time has not yet come to deploy them in scale.

The key inhabitors of this approach are the video quality and sound synchonization over VDI.

The Answer may be in creating a simpler network with high bandwidth between Home and CO with distributed VDI residing at the edge along with protocol and codec optimization for High quality interactive Remote Audio/Video Experience.

References
Bandwidth - http://www.icf.at/en/6000/how_much_bandwidth.html

VDI Protocols -http://searchvirtualdesktop.techtarget.com/news/article/0,289142,sid194_gci1353748,00.html#

Location for a Secure, Powerful, Well Connected, Cool and Green Data Center

2009-12-05T15:18:00.000-08:00

The Choice of a Data Center location is largely influenced by emotional need for security, Business Sense and Technical Prudence.

As one of the largest data center provider of India, our team is constantly juggling between these factors.

It is important to understand the emotional needs of user community. Even in this age of borderless commerce and cloud based virtual infrastructure; there is still a premium for secure and accessible data center facilities. Users continue to feel safer if their servers are close to the work location. They are willing to pay a premium for accessible and secure locations. In addition to emotional needs, there may be regulatory and political factors for locality of the data. The recent drive by government of India to fund State Data Centers(SDC) in many of the Indian states under the National e-Governance Program is a visible example to fulfill these needs.

In my Experience, It was easy to convince the datacenter users/customers to move-in the computing assets after a brief tour of the data center. I am assuming that this experience gives them a sense of security and accessibility to trust their assets to the datacenter providers.

It makes perfect business sense to evaluate initial investments, Input costs and consumption patterns before deciding the location of the data center.

The real estate cost, construction viability, un-interrupted availability and consistent price for electrical power play a vital role in choosing the location of the data center. Some Projects are being conceived in locations close to power sources such as hydro-electric, thermal, and other power plants to reduce the power distribution losses leading to lower power tariffs and increase in viability/profitability.

An important factor for Internet Data Centers is the proximity to Internet Exchanges (IX). Multiple access paths, reduced latency and high bandwidth internet connections allow building of reliable high performance VPN based overly networks for the datacenter users.

The climate of the location plays a crucial role in determining the HVAC needs and drives the associated costs, Dry and moderately cooler locations have reduced HVAC costs.

In addition to HVAC, virtualized multi-tenanted design of compute, storage and network infrastructure determine the level of commitment to meet the Green IT objectives.

Green IT may not only be statement of corporate social responsibility but also allows to leverage local government incentives tied to meeting the Green/Environmental objectives.

Picking The Right Spot for a Data Center helps in successfully building a Secure, Powerful, Well Connected, Cool and Green data center of the future.

http://www.sifycorp.com/scripts/datacenterservices.asp (our team)

Post Recession New World of SAME and Greener IT

2009-09-10T23:13:00.000-07:00

This is the age of IT savvy business executives. they have a keen insight into the benefits and pitfalls of IT.

This crop of Business executives constantly challenge the IT departments/CIO's on both merit and value of implemented technologies.

The recent downturn has further increased the scrutiny in to the exact spend and benefits of IT investments.

This resulted in a significant reduction of spend on Big ticket/Big Toys(Large SMP servers, Storage and Network Overhauls) and Growth oriented Buzzword deployments(BI/CRM etc).

The New Mantra for CIO's is to plan and execute growth oriented projects with S.A.M.E (Simple, Agile, Mobile, Efficient) and Greener IT :

Simple – CIO's are looking for simpler more transparent closed ended deployments of software and hardware infrastructure at a predictable cost. This driving the adoption XaaS - (X as a Service -SaaS-Software, PaaS-Platform, IaaS - Infrastructure etc) in the Enterprise.

Agile- The Projects generating unique Intellectual Property are increasingly being orchestrated from a mashup of cloud based services, Proprietary and open-source software .

The Projects are typically of short time frame preferably using SOA (Services Oriented Architecture) With Agile development techniques

Mobile- Mobile Devices like Iphone, Blackberry and Android and Netbooks have grown into full fledged compute/access devices with ability to view and edit data from standard office applications. The new class of Rich Internet applications are increasingly mobile aware.

CIO's are compelled to provide not just email but richer workflow functionality along with related documents on the Mobile devices.

Efficient- Efficiency is the new buzzword for both technology providers and the CIO's. Apart from Classic VMWARE based Efficiency mantras, There are efficiency bounties from storage technology providers like Netapp and Symantec using concepts like thin provisioning, de-duplication and archiving techniques.

The Network Virtualization from Cisco (Nexus 1000v and VN-link) and New players like Xsigo are banking 10 GigE and Converged ethernet standards to tout the benefits against classic Fibre channel and Gigabit Ethernet networks

Greener IT – The early adoption of power, cooling and space efficiency were driven by eco conscious CIO's.

The Technology trends due to the physical barriers have ended the Ghz race for processor design and resulted in power efficient scale out multi-core processor designs.

Small Form factor SAS/SATA Hard drives and Flash based storage are helping the power efficiency of storage subsystems

Server vendors are touting Blade technologies and skinless servers to drive extreme scalability and Private cloud/grid like deployments in the enterprise space.

Network and Security Technology Providers like Juniper, Fortinet and Cisco are Providing Multi-tenancy for Both Core, Edge and Application Network services. this helps to drive the power, space and cooling efficiencies of networking devices.

The introduction newer more efficient remote desktop access protocols with multimedia capabilities, security benefits and power efficiency are increasing the relevance of Virtual Desktop Infrastructure from Task work environments into mainstream enterprise workplace.

The Innovations in software like Application Servers and Federated/Grid based database technologies based on Multi-threading and Paralellization approaches are reducing the need for powerful scale-up requirements for the servers. This in turn contributes to greener IT.

In conclusion, In the post recession new world, The Greener IT has become mainstream requirement and the Focus on growth oriented projects for the CIO's is the SAME.

http://www.linkedin.com/in/ramkaran

Bicycles and Virtual Desktops

2009-07-30T09:31:00.000-07:00

When I was kid...
We used to go on bicycle trips...
one, two some time three people on one bicycle...
the bicycle trip was more fun with more people sitting on them but painful for the driver.

Sharing a Single user appliance with multiple users is generally painful for the primary owner of the resource.

Let's take the case of Desktop computers. If you carefully analyse a desktop computer, there are three main functions:

1. Access - Keyboard, Video, Mouse, Sound, USB etc.
2. Storage - Harddisk and CDROM
3. Compute - Processor and Memory

Let's take the bicycle analogy again of sharing access - this can be realized by software like VNC Server and VNC Viewer - here One Driver(VNC Server) and Many passengers (VNC viewers) can share the Access experience. In this case the passengers(VNC viewers) are the sufferers. So Sharing the Access layer may not be such a good idea.

The same is not true for both storage and compute. The Capabilities and capacities of desktop storage and compute engines are far beyond the utilization levels of a single user. It may make perfect sense to slice these resources and share them with Access only(thin client) devices.

The concept of conventional VDI solutions has focussed on replacing the Desktop environments with Data center based compute and storage architecture. This mainframe like approach has mainframe like limitations on scalability and affordability.

We(sify team) are are proposing a hybrid Distributed VDI approach by Sharing only the Storage and Compute parts of the standard desktops for Achieving affordable and Sustainable VDI deployments.

The Concept Reduce Compute power, Reuse existing hardware and recycle the access devices is powerful yet simple approach to affordable computing.

The Practical Realization of this approach hinges on the fact that each desktop acts as access device in the environment and is configured to share its:

1. Compute resources
2. Storage Resources

There are equal or more number of access only devices like thin clients that are brokered by a middle ware layer to access the share resources.

We are open to comments and critical review of this idea to perfect it...

Please feel free and flame...

-4thought

Top 10 Priorities for Data Center Optimization

2009-06-15T13:32:00.000-07:00

A friend of Mine shared these question to help his friend to face the media...

All the answers are straight from my heart...

1. Is cutting costs in the data center a priority for most of your customers? What are the key factors pertaining to cost that pose a problem in data centers?

Security enforcement, Power Density and Cooling efficiency

2. How do you propose to go about helping clients cut costs without new investments?

Reuse servers, Reclaim Storage and Consolidate Database/application workloads – this will help in cutting costs.

3. What are the key criteria to improve performance of the data center?

- Balanced architecture with mixture of Dense(blade-like) and standard Workloads to balance power and cooling requirements.
- Well planned storage foot print with balance of performance and capacity storage.
- Plan a Flat low latency network core switching and routing with properly configured structured cable plant.
- Simpler non-intrusive security infrastructure so that server cycles are not used extensively by security applications.

4. What are the architectural changes / enhancements that could improve efficiency?

Influence Applications to utilize a scale-out architecture. Simplify the Database setup into Scale-up clusters for license optimization and easier management. Centralize security for improving the efficiency in access, authorization and audit functions.

5. Is server virtualization the best way to optimize a virtualized environment?

Holistic view of servers, storage network needs to assessed. Measure cost/benefit of virtualization. Specially due to significant licensing costs and performance requirements, some workloads may not be viable as virtual instances.

6. What advise do you give your customers who want to optimize their infrastructure at the same time plan for the future growth?

Focus on periodic de-staging of critical assets into lower tiers. Build capacity with commodity hardware. It is cheaper to acquire and run more standard commodity servers than fewer high performance servers.

A Well architected Standard server farm with 1 U/2U servers can help in maintaining a balance in power and cooling in the data center.

Invest in server virtualization and storage management software for monitoring usage and intelligent workload distribution.

7. While data centers are regaining a lot of traction, what are the common mistakes to be avoided?

- Plan your cooling capacity in a scalable manner ( wall the unused areas of datacenters to reduce power and cooling)
- Carefully plan the purchase of highly dense servers/blade, ensure the per rack cooling and power requirements are not violated
- Avoid costly upgrades and maintenance contracts by moving servers into lower tiers of usage (dev/stage/poc)

8. How can companies design contribute much towards cost saving?
Simpler well architected application design and deployment strategy can take advantage of commodity hardware virtualization to result in dramatic power saving and reduced cooling complications.

9. With Cisco entering the server market with unified computing, what impact will it have on the ecosystem dynamics?

The initial investment in UCS from cisco may pay for itself. With its tight integration to VMWARE and Concepts like VN-Link in Nexus class switches, it is a compelling environment to reduce cost of running Highly available server environments.

10. Companies may tend to neglect security enhancements with limited budgets. What is your opinion?

Security in Data center environments is not only about tools, it about right type of processes and their enforcement. Open source tools allow low or no-cost security enhance while right processes are required to make it happen.

Managed Services for Small Business

2009-04-15T03:55:00.000-07:00

The Perspective in this blog is a first hand account from working for a network service provider under going a transformation. We are moving from simple bandwidth and data center services provider to a complete managed services provider(for IT Infrastructure) by partnering with several hardware, software and tools providers.

Let me do jargon bursting for easier comprehension...

SaaS - Software as a Service - Providing a in-house/third party software in a shared Multi-tenanted (apartment leasing type) model with automatic upgrades and technology refreshes in a Pay as you go model.

Platform as a Service (PaaS) - Providing basic building blocks of IT like Servers, Storage and Networks along with some tools that enable hosting/running compatible applications on a pay as you go model.

Cloud computing - Similar to PaaS in a pay per use model with more generic/abstracted platform services.

What are Managed Services ?

Managed services in general is the practice of transferring day-to-day related management responsibility as a strategic method for improved effective and efficient operations. Managed Services for IT infrastructure allows enterprises to focus on the core competence and pass the responsibility of day to day IT management to the experts.

Why are Managed Services relavant to Small Business ?

We are living in trying times; the economic turmoil coupled with global competition and a deep awareness of climate change has a lasting impact on the business strategy for Small and Medium Business Owners.

Managed Services free up the business owner/management from day-to-day sustenance services to help him focus on strategic objectives and practical measures relevant to his/her core business.

There is a new social networking driven consciousness in corporations and individuals on the necessity to create innovative products/solutions, identify frugal manufacturing/execution options and safeguard this world for a better tomorrow.

This consciousness is driving adoption managed services to create new technologies/processes for harnessing human knowledge, technology sustainability and economic viability of enterprises.

How do managed services help harness human knowledge Small and Medium Businesses?

Communication, Collaboration and Knowledge management are key components of an effective social network to harness tacit and explicit human knowledge. The Businesses of all sizes are realizing this value of social networks and are investing heavily into communication/collaboration infrastructure.

We intend to reduce the inherent complexity of operating Multi-Media communication infrastructure by driving the adoption Voice Services, Contact Centers, Messaging and Video Conferencing as Managed services or on a SaaS platform.Our capability to deliver these services is largely due to symbiotic relationships with our partners who provide technology, tools and expertise to quickly ramp up the SaaS/PaaS platforms to serve the customer. We have challenged the conventional wisdom by collaborating with two competing 600 pound gorillas(google/gmail and Microsoft/Exchange) in the collaboration technology to serve different classes of customers.

How can managed services provider help Small and Medium Businesses in Technology Sustainability?

Managed Service Providers bring the technology innovation in Virtualization (on Storage, Network and Server fronts) to enabling new and unique business models for Remote Infrastructure Management for enterprises of all sizes.Virtualization not only allows easy realization of compute on demand infrastructure but also creates compelling business case for outsourcing system builds and software deployments to Remote Managed Services providers.

There are new trends in Desktop virtualization that remind us of déjà-vu of mainframe days with data center based user computing environments.

This creates a clear opportunity for standardizing, simplifying and remote management of desktop infrastructure directly translates into faster ROI for these services.

There is a rapid convergence is happening in storage technology, the innovation in storage arrays (like Thin provisioning and de-duplication) and the advent 10 gb Loss less Ethernet along with support for emerging standards(FCoE) is reducing the cost and complexity in network storage.

This convergence is creating opportunities for Data Center providers like us to morph in to Managed service Providers by offering Storage on Demand both for within Data center use and for Remote replication/Business continuity.

Last but not the least, Original equipment providers are focusing on Green IT technologies by reducing their carbon foot print and measuring themselves on metrics like Performance Per Watt and power consumption ratios(Idle to full capacity).It is easier for us(Managed Service providers) with our deep partnerships with technology providers like Cisco, EMC, HP and IBM to rapidly adopt the technology upgrades.

There may be a time when organization may stop buying Hardware, software or security appliances and instead opt for specific business functions as a service from managed service providers to realize the true benefits of sustainable green IT Infrastructure.

How do we improve Economic Viability for Small and Medium Business ?

It may be challenging to allocate and spend capital expenditure on IT assets for Small and Medium business in this economy.

PaaS/cloud computing offerings from managed service providers make it easier for small businesses to consume IT assets in a pay as you go model.Managed Service providers are able to leverage the economies of scale, relationship with technology providers and financial institutions to acquire the Building blocks for SaaS and PaaS services on revenue share/annuity model.

Small and Medium businesses can move from CapEx to OpEx model for IT investments to conserve cash and improve economic viability by subscribing to Managed services.

It makes sense to partnering with your Managed Service Provider...as an early adopter.. Typically SaaS and PaaS services are available for trials, beta testing and early adopter programs.

The Small and Medium Business can take advantage of early adopter programs, these are generally in the spirit of partnership. The providers have a deeper engagement during these engagements and may not charge for other wise billed professional services. The reward is significant enough to warrant the risk.

In a Nutshell, As the technology stacks are become more complex and expensive. The economic reality of managing the complexity may compel us to seek and nurture symbiotic relationships managed service providers for long term success.

Automating Storage as a Service

2009-02-22T12:29:00.000-08:00

There are hardly any tools to truely automate a Multivendor Storage Services Environment.
Storage Admin's are stuck with Veriety of Independent Web interfaces or commandline tools with little or no common semantics for common configuration elements or a provisioning activities. The procedures are generally convoluted with myriad manual settings...

This is the most realistic dipiction of Multi-vendor Storage Administrators Plight...
I was a ECC(http://www.emc.com/products/family/controlcenter-family.htm) user and large-size env.. EMC Storage admin.. ECC got very close... but it was pig... to slow. Good for Symmetrix..Weak for for Clariion and third party arrays..

There are tools like NetAPP SanScreen(http://www.netapp.com/us/products/management-software/sanscreen/sanscreen-service-insight.html), Symantec Command Central( and HP AppIQ(http://h18006.www1.hp.com/products/software/bto/srmgt/index.html), itheon's SRA http://www.itheon.com/solutions/storage_resource_analysis.html, look interesting, They are mostly limited data collection and do very fractured provisioning/monitoring/Host-Side SRM Functions...

I am still on the lookout for a simple tool that is intellegent enough to understand a SAN/iSCSI Network. Automate Zoning/VLAN creation, Mapping of LUNS, help in LUN masking and collect host-side Usage information.

This is the first step to storage automation, consequently integrating CMDB, Incident and Change management logic into the storage environment with Explicit awareness of Snapshots, Replica's and Remote replication processes(like EMC Replication manager).

The Second Step is understanding the complexity/flexibility offered by Storage arrays for features like SANcopy and other host based Bit-Wise LUNcopy software and integrating into the hierarchical storage management lifecycle.

The Third Step is to enable Physical(Boot form SAN) and Virtual Machine Management for Optimised storage usage. For example: VMware's Storage DeDuplication for VDI solution Allows one 10 GB LUN being used for 100's of desktops. Understanding Application level storage usage (Storage Resource Management Functionality) can be benefecial for Not only for Remote side Storage allocation using Thin Provivisioning techniques but also for Backup Verification

The Landscape for Storage services both on the cloud and in house environments is changing rapidly. Cloud Storage here to stay, however storage resource management is even more critical in cloud storage since there is a direct revenue impact due to over provisioning. The In-house environments are converting to more centralised service provider like architecture.
Using Centralised tool is extreme valuable, however if the cost is prohibitive or if the products donot meet safety and usability standards. we need to go back to the vendors to get commandline API's to start automating the key processes(Provisioning, Chargeback/Billing and Capacity planning) as soon as possible.

Virtualization 2.0 - Benefits and Challenges

2008-12-04T00:39:00.000-08:00

Virtualization as a technology existed from the mainframe era in large glasshouse data centers.
Transfer of virtualization technology on to commodity x86 based hardware has increased the adoption of virtualization.

The first generation of x86 based Virtualization technologies had limited use in testing, development and re-hosting legacy applications.

We are seeing rapid evolution of Virtualization technologies by the vendor community into what we know as Virtualization 2.0.

What is Virtualization 2.0 ?

In my view, Virtualization 2.0 is about taking the virtualization from development, staging and production servers to Desktop (Virtual Desktop Infrastructure - VDI), Mobile and affordable Cloud based Disaster Recovery/business continuity applications.

Virtualization 2.0 signifies free/inexpensive access to advanced hypervisor based Virtualization technologies from various vendors.

This is allowing the change in deployment models for virtualization from typical scale-up approach(SMP with large Memory servers) to scale-out model on low-cost cost off-the-shelf
hardware.

There is a lot of emphasis on hardware Assistance (Intel VT etc), Provisioning, Metering and Management tools.

Trends and Technologies propelling Virtualization 2.0

Green IT - The proponents of Green IT have raised the consousness on energy efficiency and resource utilization.

Cloud Computing - Creating large uniform Compute clusters to enable cloud computing is a common practice. Amazon has taken lead on providing on-demand compute and storage resources to scale up your operation. Portability of Server state provided virtualization is fundamental requirement for the success of cloud computing.

Blade Servers - Blade servers are improving manageability and creating opportunities for labor and material savings in the datacenter. Their elegant design, easy operations and reliable performance are helped the adoption of virtualization

Networked Storage - Networked storage both SAN/NAS have helped in consolidating servers and services in Virtual machines by removing scalability constraints imposed by direct attached storage. Today VM's are directly being mounted from networked storage.

Networking - the average speeds of internet connectivity are growing. this growth is fuelling need for better utilization cloud based virtualization.

Impact of Virtualization 2.0 on Storage and Networking

Virtualization 2.0 would not only feed on Networked storage but also generates a need for larger network capacity.

· In case of VDI the Virtual Machines of Non Logged on users remain dormant and when accessed consume a lot of bandwidth.

· Mobile Device are the next target for virtualization due to falling cost of flash storage and ease of application downloads due to better connectivity.

· In Disaster Recovery Scenarios…Storage Network based replication helps Virtualization to create easy recovery options.

What are the benefits of Virtualization 2.0...

Virtualization demands better managed data center environment due to power and cooling density due to higher compute/storage resource utilization.

The effect of better utilization of compute, network and storage resources translates into the hard dollar savings on IT infrastructure while promoting the Green IT agenda.
The cloud computing or Server in the cloud is relavent to internet centric businesses, academic and one-off projects.

A more compelling story is emerging due to virtualization in the Enterprise Cloud computing.
There is Rapid development, evolution and availability of provisioning, metering and management tools for virtual servers and cloud computing.

This trend is creating opportunities for in-expensive capacity building for enterprise cloud computing farms using low-cost servers, network elements and storage(iscsi).

Virtualization helps typical stateless workloads like Internet Security gateways, Web servers and application servers to be moved into enterprise clouds to reduce overall cost of the infrastructure.

The savings realized from redeploying expensive servers and network elements to new data oriented state based applications.

Disaster Recovery and Business continuity have become compelling in terms of cost and reduction of deployment complexity due to advances in synergy of storage and virtualization technology.

The Availability of ample bandwidth and on-demand cloud computing is enabling even the small and medium enterprises to implement working Disaster recovery environments using virtualization.

Challenges of Virtualization 2.0...

There is a significant challenge in the management large scale virtual infrastructures. There are no clear boundaries and responsibilities in terms network, storage and datacenter management teams.

The impact of faults and incidents is felt across the environment. it hard to troubleshoot the performance and stability issues as there is significantly small body of knowledge on Virtual server implementations of software products.

The support teams for most of the Enterprise software vendors are not particularly geared to support virtual environments.

The open-source product deployment is little bit more complicated as there is lack of community participation on deployment issues.

There is a significant learning curve to realize the synergies of Virtualization and networked storage. Most of the Virtualization still limited development, test and isolated production environments.

Security of Virtual Server environment is a hard nut to crack.. Wider availability of virtualization technology lends more discovery of vulnerability and exploits. This creates resistance in deploying mission critical workloads on virtualization.

Virtualization 2.0 is here to stay...

In spite of the learning curve, cost and security issues, Virtualization is here to stay due the flexibility, energy and cost savings.

Virtualization helps us by improving the time-to-innovate, time-to-test and time-to-deploy.
It improves the efficiency of IT staff and allows them to focus on innovation in servicing customer requirements.

2008-05-16T11:36:00.000-07:00

Ubiquitous Storage - data everywhere

How many times we have heard someone saying:

"Send me the Quarterly numbers….I will check on my Blackberry…"

"It’s a big file….gmail it…or put it on the thumb drive…."

"Lets look up every thing on xyz product – designs, manufacturing records, marketing materials and support cases"

When you start connecting the dots of data management. It is no longer limited to the data center or desktop...

If you follow the bit trail....There is data everywhere...

The Data management revolves around a simple information lifecycle and associated tools:
Data Creation – MS Office, ERP
Distribution – Email, Web, FTP..
Action – Document management systems & other applications
Disposition – Backup and archival application

On further study, You will realize there are three distinct classes of storage:

Personal Storage

iPod generation
Smart Phone, Blackberry, PDA
Thumb drive

Internet Storage

iDisk, SkyDrive, Gdrive
Flickr, youTube, gmail
Online backup

Enterprise Storage

Gigabyte mailboxes
Terabyte Data warehouse
Petabyte File Archives

The History of Personal Storage can be traced back to punch cards, microfilms, large floppy disks(5 1/4 or larger) and tape. The adoption of solid state drives into consumer electronics has driven down the cost of personal storage devices. The evolution from Kilobyte to terabyte size personal storage devices creates compelling need for data management.

There is notion that Internet Storage is always available and accessible. It does not eliminate the need for data management. There is a sense of loss of control and lack of privacy with internet storage. This in turn creates new opportunities for data management for Internet Storage.

There is no need to emphasize the need for data management in the enterprise storage space....

Innovative companies like Fabrik (http://www.fabrik.com/) tacking these data management challenges by marrying personal storage with internet storage and topping it up with online backup services with its recently acquired a personal Storage Vendor (http://www.simpletech.com/).

There are others like Apple who have integrated simplified backup capability into a Wireless Access Point with deep integration into the operating systems and tying it with .MAC services based internet backup.

There are players like Synchronica offering solutions to backup mobile phone based personal storage.

We cannot ignore the Google Gears approach to always available data for Google Apps. Microsoft SkyDrive and Live Integration with its Cloud OS(Internet Cloud) Strategy is a step in the same direction.

There are hardly any vendors seriously pitching Enterprise storage in the Cloud. The new trends in virtualization like Virtual Desktop Infrastructure, Cloud Computing, Grids, Managed Applications and Software-as-a-Service are first steps to moving enterprise data into the cloud

Ubiquitous Storage can be viewed as a convergence of personal, internet and enterprise that’s always in sync. This convergence is inevitable, Its happening on mobile devices like laptops and smart phones today. It is taking shape in the data access layer in the enterprise with Mash-ups and enterprise portals.

There future challenges of data management revolve around providing consistent means to manage personal, internet and enterprise storage.

I firmly believe that next generation operating environments and web computing platforms will realize the vision of seamless always available data everywhere...A. K. A. ubiquitos storage

Mission Critical iSCSI Storage Network

2008-04-15T12:48:00.000-07:00

Storage Networking is a tricky animal...My brush with networked storage platforms started from the time we needed few hundred megabytes of shared storage for building a cluster to enable database and email consolidation in the late 90's.

The essential character of Block based network storage continues, the goal is to protect and consolidate mission critical workloads.

Networked Storage using traditional FC-SAN's are getting more complicated in the quest for speed and functionality.

In my view, there is a need to simplify networked storage to reduce risk, decrease mean-time-to-repair and reduce costs.

Drawing from my personal experience, More complicated SAN sub-systems and network elements are harder to understand, harder to troubleshoot and expensive to deploy....

Personally, I have moved away from SAN implementations with FC-Front end networks after attempting to use use FC-to-ISCSI routers and put up with the complexity in the nework layout and provisioning challenges.

The Approach...I took was to select medium-to-high performance native iscsi storage arrays with FC-Disk based Backend networks. Deployed the front-end network(for connecting to servers) with Redundant Gigabit Ethernet(layer 3 capable) Switches to mimic a FC network for Multi-Pathing and fault tolerence.

This allowed me to maintain the essential performance posture(with FC disk backends) while maintaining the front-end simplicity for iscsi networking.

There is still a question performace in activities like Synchronous replication, where FC-SAN technologies are superior but these needs are becoming less acute.

Applications Like MS-Exchange supporting varied data replication topologies( Cluster Continous Replication-CCR, Local Continous Replication-LCR, Standby Cluster Replication-SCR).

Databases(oracle,MS-SQL, MySQL, Postgres) are supporting high availability using Grid/federation, Mirorring and Master/Slave models.

Advances in Storage awareness in the modern operating systems(Windows, Linux and Solaris) allow support for native multi-pathing for iscsi.

Support for Features like VSS and VDS on Windows, GFS on Linux and ZFS on Solaris allow for less stringent application aware iscsi friendly asynchronous replication between storage systems.

Another Major Challenge is the emergence of Virtual Machines and thier impact on performance and availability of Networked Storage.

The FC-Storage have inherent disadvantage in VM based environment due the multiple layers of device drivers and in-memory indirection of I/O calls.
Practical impact of on processor usage in hypervisor based VM setups using FC-SAN are evident in due to lack availability and optimization of transparent ToE type solutions for FC.

The Ability of a Virtual Machine to use raw ethernet adapters with ToE capable drivers results in little or no loss of Storage I/O performance. there is practically zero impact processor performace due to the use of networked storage.

In conclusion, With the emergence of relatively inexpensive 10 gigE switch and HBA Solutions, continued sophistication of Operating systems and applications, the time is right to start adopting iscsi for your mission critical storage networking needs.

2008-03-28T07:22:00.000-07:00

SaaS - Microsoft way -- H M C 4.0

2007-12-21T09:30:00.000-08:00

Microsoft offers its Crown Jewel MS-Exchange 2007 and Windows Sharepoint Services as Hosted Messaging and Collaboration Solution.

This is a incrimental development of products for Service providers like MS - Internet Services, HMC 3.5 with Exchange 2003....

This solution requires eloborate Hardware and Software Infrastructure covering almost all server products from Microsoft Stable.

MS-SQL Server
MoM -SCoM
Active Directory - CDI
ISA
SharePoint
Provisioning System

I fail to understand, why do we need such diverse set of software products to run an already complex mail platform like Exchange in a multi-tenent mode...

It not only increases the exposure in terms of complexity but also results in skills shortage to run such a diverse environment.

A Seemingly simple deployment tool has so much depth to it that, If we end up making any mistake...it will lead to a long and unanticipated delay....

The some portions of supported software platform do not run on the Latest 64bit windows servers.

The setup forces us deploy critical infrastucture like MS-SQL on a 32 bit platform due to nuan ces of Microsoft Provisioning Framework and .NET version limitations.

Our team has implemented HMC about 4 times till now, There are new uncertinities that are cropping up in the 5th install cycle...

In the name of making the deployment and multi-tenency easy...We are seeing that the HMC framework is ending up to be brittle...

Inspite of all this, I should hail to HMC as a very spirited effort to simplify the Service Provider operations.

Simplifying Web Security

2007-11-11T10:13:00.000-08:00

There are a bunch of Blog posts on 11/11 Cyberattack.... Threats like these help in understanding the pains and gains of planning/deploying effective web security measures....

The fundamental principles of Security - confidentiality, integrity and availability need to drive the decision making process....for securing Web Properties(Portals, ecommerce and business websites)

In my opinion, the defences need to be created at every level of the infrastructure....the key layers being:

Hardware
=======
Network
Storage
Servers
Appliances

Software
=======
Operating System
Webserver
Application server
Database server
Special appliance firmware/os

The typical threats are centered around increase in load (DoS/DDoS), compromising data (sql injection or Web server exploits) and compromising Authentication/Authorization methods.

A simple approach to deal with most of the issues is by using a 3 step approach:

1. Improve application security - Application security is turning from art form to more exact science with thier relavent calculations. this is slow and painful process of securing, reauditing and assessing the impact on the overall security posture

2. Increase Server capacity -- Methods like vmware and/Blade server environment help us to define the scale that is needed to handle sever capacity... biggest challage is to augment server capacity as soon as we see a threat. the simple Robinhood approach- to steal from peter to give it to paul... The way this works:

1: To have a Virtualization software(vmware/Xen/Virtual server) is used as a part of every server build.

2: Convert your physical servers in virtual servers and create the set of files that makes up the know good image. Keep there VM images are all of the webservers that are serving low volume content. activate the VM's behind the same load balancer...inorder to handle any suddent traffic bursts...

3. Tighten the Perimeter -- The approach here is to protect each layer of security so that some of the illegite traffic should not be able to enter....

Snort-inline - IDS/IPS
ArmorLogic - Application firewall
Fortigate - Unified threat managment

Are we home safe with all these methods... not yet...the key to responding any crisis is calmly listening/listings all the possible threats, map it to various measures undertaken to reduce the attack surface.

ForeThought - DeMystifying SAP Security

2006-11-13T13:19:00.000-08:00

ForeThought - DeMystifying SAP Security

I attended SAP Tech-Ed in Bangalore... I am now Officially SAP Netweaver Security Certified. Let me share my understanding about SAP Security.

In my opinion, Classical SAP Security model with Users and Roles is simple yet a well thought out mechanism for giving access to Data and Functionality inside SAP.

The Areas that need clearer understanding/implementation knowhow are Security model for Portals and Business Intelligence.

One of the Areas which is left to abmiguity is security in the Presentation, Session and Transport Layers. Evidently SAP is moving from SNC to HTTPS protocol for Server to server Communications.

With the new emphasis on WebDynpros(Portals) and Exchange Infrastructure(XI) the Diag and Rfc Protocols are giving way to HTTP. This transformation is not very evident as most SAP implementations are centered around SAP GUI. The Core Single-Sign On Central User Admin(CUA) is still based old model.

A Promise from from SAP to deliver the functionality using Standard MS-Office platform is still fairly limited from a security perpective.

If you are typical IT guy...Dont be surprised to see your fellow SAP Basis person speak your same language of LDAP authentication, Kerberos, SAML and SSL Accelaration.

Infact It was heart warming to see the emphasis on Standards based security implementation approach for designing Internet facing SAP Solutions. This includes emphasis on Application gateways, Load balancers, Intrusion Detection, Identity Management and Change Management.

Future of SAP activites
================
User Provisioning - Manual(within SAP), Automated(SPML based) or Managed through LDAP
Role Assignment - Manual(within SAP -PFCG), Automated(SPML based) or Managed through LDAP
Role Creation - Manual Inside SAP - PFCG
Transport Security - Standards based/handled in Appliances (e.g App. Gateways/ Load Balancers etc)
Operating System/Database Security -- Not in Scope

At the pace at which SAP is delivering the new products and functionility this new approach allow them keep a close tab security.

Out-of-the-Box - Avoding Custom Side-effects

2006-10-23T12:06:00.000-07:00

In my experience.. Most large software infrastructure product deployments run into the path of becoming non-standard implementations.

IT decision makers are faced with a barrage of requests for customisation even after initial deployment of the product. The pressure of these requests drive even the very seasoned managers in to a trap of massive customisation. These changes end up in the installation without objective evaluation of real and percieved business value and their impact.

This is worse..if the changes are executed during the roll-out or right after....

The resulting deployment is vulnerable and may result in:

1. Lack of Stability
2. Complication in Upgrade path
3. Less security

The Support issues around such implementation cause undue stress on the staff.

The other side of the coin is an arguement about the importance of functionality for improving efficiency and user experience.

A Tight control on the finalization of requirements during pilot phase, A Disciplined approach towards function testing and User Acceptance testing, Feature Benchmarking of Similar sized Application Deployments are key to maintain a healthy balance.

In my opinion, When deploying new products...It is better to err on the side of caution and work work towards implementing a standard out-of-the-box product with minimum customization to ensure a simple, secure and scalable installation.

This Approach can be coupled with stacking up the incoming feature requests and enhance ments requested after the initial roll-out into a planned next revision of the deployment.

In this way.. The change requests can be carefully evaluated by the vendor and the implementation team. They can proiritised based on their impact on simplicity, security and scalability of the deployment.

Speaking of Business.....

2005-09-28T00:27:00.000-07:00

AVDL V1.0

CAP V1.0

DITA V1.0

DocBook V4.1

DSML V2.0

ebXML CPPA V2.0

ebXML MSG V2.0

ebXML RIM V2.0

ebXML RIM V3.0

ebXML RS V2.0

ebXML RS V3.0

OpenDocument V1.0

SAML V1.0
SAML V1.1 and
SAML V2.0

SPML V1.0

UBL V1.0

UBL NDR V1.0

UDDI V2 and
UDDI V3.0.2

WSDM-MOWS V1.0

WSDM-MUWS V1.0

WS-Reliability V1.1

WSRP V1.0

WSS SAML and
REL Profiles

WSS V1.0

XACML V1.0 and
XACML V2.0

XCBF V1.1

Little did you know we had so many dielects in speaking of business....the technology around this dielects is rapidly maturing....Although basic transactions remain same...novel business processes are being woven around the data created using these dielects...

Frameworks are being built to leverage the intelligence that can be derived from these processes....Buzzwords like Business Process Monitoring(BPM), Corporate Process Monitoring(CPM) are used to describe this convergence of Historical data and real-time data for effective decision making...

This is just the tip of the iceberg....As standards get matured....We can expect third party data sources bieng used for correllation and benchmarking process metrics across the across the industry verticals in real-time...Resulting in constant process innovation to stay ahead in the industry....

It is important to build a simple, secure and scalable infrastructure and flexible process configuration capabilty to leverage these emerging languages of business into innovative business processes.

Road to Simplicity

2005-09-09T22:29:00.000-07:00

Hurricane Katrina Demonstrated the how complex - reasonably prepared organizations... who modelled the the scenarios and designed reponses to the impending disasters failed to execute their DR plans effectively.

The Systems and underlying Infrastructure failed miserably... the preparations did seem to have taken into consideration that the Infrastructure like Roads,Transportation, Power and Communication Systems would themselves be vulnerable for catastrophic failure...

There was little or no forethought in relocating and concentrating people in locations like the Dome and the convention center.. straining the already fragile and vulnerable infrastructure...

The countermeasures were less-than-perfect...It was impossible to reach most parts of disaster stricken areas by road or communicate over the phone.

It is said that FEMA/Department of Homeland security did not inform the public about a Simple emergency hurricane readiness kit...like they did after September 11....

There was little or no governance right after the disaster... chaos previaled leading Looting and Arson...

At the time when all of the large scale infrastructure elements failed...people improvised with bicycles, kayaks, battery-torches, White Flags and learned to survive...

Experts believe a well co-ordinated deployment of smaller simpler rescue approaches could have saved more lives and provided quicker relief.

In the hind sight Katrina required a nimble, flexible and well-co-ordinated systems on behalf of the government instead Multiple big bang poorly organized efforts by multiple High powered organizations.

What can we learn from this event...How can we apply it to the IT infrastructure... and DR Plans....

Security and Disaster Recovery is genarally a after thought in most IT infrastructures.

We need to start thinking of architecting solutions with security and DR right from inception.

As IT industry progresses to create feature rich solutions to solve myriad enterprise challenges....The complexity of each of the solution is exponentially increasing...

This complexity is responsible for multiplying the effects of IT Infrastructure Diasasters.

The Complete Security Triad of Confidentiality, Integrity and Availability also favours smaller and simpler implementations.

Let'stake confidentiality...It is more challenging to assure confidentiality with Large scale systems with multiple connected components. since each of them(servers, storage and networks) can present more formiidable challenge of implementing security controls. Adding technical solutions like inband appliaces and applications increase the complexity will make the entire system look like a Goldberg's Machine...

Integrity is much more simpler to explain... It is harder to recover data corruption from larger storage systems and than a standard disk drive... since the formats layout and tools required are also complex....There is a false sense f security in storing data in large scale systems...it can easily exceed designed capacity...resulting major data protection proles like backups restores and archiving. It makes it harder to recover from disasters.

Larger More complex servers with Multiple levels of Redundancy...
Networks with high availability and Storage subsystems with six 9's availability still cannnot protect you against a misconfiguration or bug in the software.

Most of us fail to realise that software is seldom written to fail gracefully...Specially if contains custom code written with a assumption of system being up all the time.

Even the Most seasoned software solutions are still vulnerable to this effect due to constant enhancements and the lack effective regression testing in High availability Scenarios

High avilability is hard to realize...In such cases...The basic Risk eaquation of SLE(Single Loss Expectancy) x ARO(Annual Rate of Occurance)...favours smaller and simpler implementations.

We need responsive, nimble and flexible systems infrastructure..instead of large, sophisticated, hard-to-understand systems.

Architecting such a IT solutions with these principle allows us to be nimble and flexible to respond disasters more effectively.

Co-ordination, communication, Flexibility and nimbleness are key to effective diasaster response.

Implementing IT solutions on Intelligent, Simpler, Smaller platforms(Grids, farms, blades etc) with strong emphasis on people, processes and communications is the Road to simplicity.

ForeThought -- Global directory service for the Digital world

2005-04-10T23:35:00.000-07:00

ForeThought

Real world is full of people. We have phone books,yellow pages, maps and postal addresses to get to them.

Real world is governed by laws, rules and norms.

Digital world is full of devices and devices only(severs, desktops, appliances, gadgets etc..)

In the digital world, We have DNS, IP Addresses/SMTP mail address, Routers/gateways.

All of these methods are still rudimentery to lookup and locate attributes, features and functions of devices.

Most devices serve a useful purpose. some of them are misused to inflict pain and damage in the digital world.

Clear knowledge attributes, features and functions of devices can hel in reducing the risk and help us in responding to misuse of these devices.

An idea of a Global Directory Service -- for locating Devices worldwide....These devices are attributed to People, Resources and Services....

This type of service can help in Asset tracking, Monitoring, Communications, Incident response, forensics and Data-minig.

One of the attribute of the device is owner of the device.

It is virtually impossible to have this kind directory with current network topologies and IP addressing mechanisms. Most networks rely on Private IP address ranges and isolated using firewalls and NAT appliances.

As world turns more IP based techniques to locate and document devices on a IP network need to be refined and perfected.

Tracing communications and mapping topology of IP networks.

Geographic co-relation of IP networks of IP resources is a interesting idea.

The idea is to tap in the information at all layers of the application stack to automatiatically update the Global directory service for the devices.

There is ray of hope with the new IPv6 Protocol...There a possibility that all these devices an be represented by public IP address.

This may allow us realise the vision of creating Global directory service for devices.

ForeThought-Consolidate for the right reasons

2005-04-01T03:51:00.000-08:00

ForeThought

Consolidatation is one of the Buzzword used to rationalize spending money on new Infrastructure products and technologies.

The Arguments almmost always centers around reducing the management costs(both in terms of people and infrastructure).

While I agree on this premise for consolidation, I have my reservations on the effectiveness of the exercise If it is not performed with a holistic approach.

There are often independent consolidation projects for:

Datacenter Consolidation
Server Consolidation
Storage Consolidation
Network Redesign/Consolidation
Database consolidation
Application Server consolidation

There are lot of cycles spent on justification, discovery, redesign, implementation and testing of each of the above initiatives.

Some times the conclusions from each of the exercise is contrary to each other. A Storage consolidation effort does not gel well with Database consolidation etc. etc....

India - Dreams And Aspirations

2005-01-24T13:54:00.000-08:00

I have come to India after 3 years...

The India I knew is fast disappearing...

Lazy Irany Hotel Chai breaks...

Street corner hangouts on a stationary bike...

Those spontenous galli cricket sessions...

The little twinkles of happiness are bieng repalced by....

Fast paced Cafe Coffee Day's....

Bowling Alleys...and Bars

Life has suddenly found a fast lane in my home town hyderabad....

Looks like most kids out of college end up earning more than my father earned after 20 years of Government service...

Good Dinner outing with the family runs into half the person's salary...

Its easy to buy car or a home(with a loan)... and there is lot of peer pressure to do so....

We have yet to realize and embrace the magic mantra to a fulfilling life...

Simple Living and High Thinking...

Its never too late for introspection...analysis and eventual change...

Lets look at Changing from Consumerist Habits ( more soaps, more tooth pastes/tooth brushes, more crocin, more refrigirators, more TV's...). Every one of these habits result in waste, pollution and unsustainable apitite for either things that are not nessecary or things that are not affordable in the long run...

It is better to inculcate clean, economical and sustainable habits in order to positively influence the conservation of precious resources...

The Mantra is consume less, waste less and apply thought to produce just enough to meet the requirement.

2004-11-10T12:36:00.000-08:00

skyline

2004-11-10T12:33:00.000-08:00

Storage in the Enterprise

2004-07-26T00:14:00.000-07:00

Buzzwords of Yesterday:

RAID, NAS, SAN, DASD, HSM, Networked Storage

Buzzwords of Tommorow:

iSCSI, SATA, SAS, ILM, Utility Storage, Storage Grids, Storage Security

Enterprise Architecture - Then and Now

2004-07-19T22:13:00.000-07:00

Enterprise Achitecture(EA) has been ever evolving discipline.

The evolution was almost always tied to the then prevelent management concepts and technology trends.

The correllation of management ideas, technologies and enterprise architecture is not new.

However understanding these topics in the current context and applying the resultant ideas to emerging enterprise-wide challenges can be a valuable exercise.

Early days of Automation were inspired by Scientic Management concepts of F.W. Taylor.

They were focussed on the following:

1. Clear Delienation fo Responsibility
2. Separation of planning from operations
3. Incentive schemes for workers
4. Management by execption
5. Task Specialization

These concepts took over 60 years(F.W.Taylor -- 1881) to be fully incoporated in automated systems.

Eary Mainframe machines were purpose built -- Accounting or Scientific/Engineering  Applications.

There was clear separation of planning systems and operational systems. the Feedback loop
of applying results from planning to operation was not automated.

Enterprise Architechure was based on the foundation of task specialization and operational efficiency with a task.

Advances in Electrical engineering, Mainframes and Wincester Disks were prime movers on the technology front. They rapidly repaced more mechanical devices in major industries both on accounting and manufacturing front.

simple

2004-07-18T01:08:00.000-07:00

Ramkaran

SIMPLE SECURE STABLE SCALABLE

2004-07-17T08:34:00.000-07:00

I work for a highly successful emerging global corporation.
I am intrigued by the challenges architecting and managing global scale Information Infrastructure.

In my view, long term viability such an infrastructure is based on the folllowing factors:

SIMPLE ideas
SECURE implementation
STABLE maintanance
SCALABLE Approach