Skip to main content

Mission Critical iSCSI Storage Network

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.

Labels

Labels:

Comments

Post a Comment

Popular posts from this blog

ForeThought -- Global directory service for the Digital world

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 Da...
2 comments
Read more

Blueprint for a new ITaaS Service Provider

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...
2 comments
Read more

RAINing Appliances in the Clouds

Appliances always had a soft-corner in Data center and cloud strategy. My Brush with appliances started in early 2000. We were looking at soultions for traffic management and billing for small ISP. the solutions with software and hardware were quite complicated. We came across small company out of Israel called Allot Networks (www.allot.com). they did a good job in a simple way. Later at NVIDIA, I was exposed Netapp for NAS storage which quickly pivoted into block storage with iscsi. they literally gave away the iscsi protocol license for free with the box for a long time. On the network side, my Exposure was with F5 networks while managing nvidia.com. We were facing scalability issues during product launhces and driver release announcements. We tried newer players such as Array Networks and Netscaler while they were in Beta. We ended up assisting and supporting Netscaler with our feedback to evolve the product to have in-memory caching. In VMworld 2013, I have seen a emergence ...
8 comments
Read more