SANRAD Dives Into The SSD Cache Pool

iSCSI pioneer SANRAD’s foray into solid-state disk (SSD) caching should help the company redefine itself from just another iSCSI bridge provider to a player in the much bigger storage virtualization and acceleration market. SANRAD's new V-Switch XL adds application-optimized SSD caching to its hardware iSCSI virtualization appliance.

Howard Marks

October 6, 2011

3 Min Read
NetworkComputing logo in a gray background | NetworkComputing

iSCSI pioneer SANRAD’s foray into solid-state disk (SSD) caching should help the company redefine itself from just another iSCSI bridge provider to a player in the much bigger storage virtualization and acceleration market. SANRAD's new V-Switch XL adds application-optimized SSD caching to its hardware iSCSI virtualization appliance. Even more interestingly, the company tells me it will be releasing its technology as a virtual storage appliance complete with SSD caching.

I must admit I was never a big fan of SANRAD’s original iSCSI to Fibre Channel and SCSI bridges. When working with several clients in 2005, 2006, a reseller would come in pitching a SCSI or, even worse, Fibre Channel array with a SANRAD V-Switch claiming, "iSCSI belongs in the network, not in the storage array." While this made some sense, considering that SANRAD is part of the RAD Group, a phalanx of Israeli companies each working its own niche of the network business, its solution was inevitably more complex and more expensive than a Lefthand or Equallogic system with iSCSI baked in.

As it moved more storage virtualization features--like volume management, snapshots and, most significantly, replication--into the V-Switch, it became more of a storage virtualization platform that could compete with FalconStor’s NSS or Datacore’s SAN Symphony (which I still refuse to call a storage hypervisor). In fact, I saw several applications where a pair of V-Switches would be a good solution.

In one case, we used a pair of V-Switches to create a completely redundant storage system across a campus environment. Servers connected to the primary V-Switch, which mirrored the data to arrays in data centers at opposite ends of the campus. Should that V-Switch fail, or the whole primary data center go off line, the V-Switch in the backup data center would assume the virtual IP address and continue processing data from its local array.

The V-Switch XL is available in 1U and 2U rack-mount versions, both of which use SANRAD’s front-plug PCIe modules. In addition to the usual PCIe slots for Ethernet and/or Fibre Channel adapters, the front-pluggable modules can accept PCIe flash, up to 2 Tbytes per module, or additional network cards. The 1U appliance has three module slots, and the 2U has another two for a total of five.

SANRAD has tweaked its caching algorithms for common applications, including SQL Server, Exchange and virtual desktop infrastructure (VDI) environments. That allows them to be smarter at predictive read-ahead and cache warming. SANRAD’s briefing deck included benchmark data showing up to seven-times performance improvements on TPC-H queries against SQL Server.

By default, the V-Switch XL sets up a write-through cache that will accelerate reads. Users can change this to a write-back cache that will also speed up writes at the expense of data consistency on the back-end Fibre Channel or iSCSI storage system. Users selecting write-back caching should use the V-Switch, rather than their backend storage, for snapshots andreplication.

I’m most interested in the SANRAD VXL, which implements the V-Switch XL as a virtual machine. Like other VSAs (virtual storage appliances), the VXL can take a server’s local disk resources and present them as iSCSI volumes for other virtual servers to use. Throw a local SSD in the server, and now you’re cooking with gas as the five 1-Tbyte drives attached to a PERC or SmartArray controller become a flash-accelerated iSCSI array that can synchronously replicate to another server with a similar setup.

About the Author

Howard Marks

Network Computing Blogger

Howard Marks</strong>&nbsp;is founder and chief scientist at Deepstorage LLC, a storage consultancy and independent test lab based in Santa Fe, N.M. and concentrating on storage and data center networking. In more than 25 years of consulting, Marks has designed and implemented storage systems, networks, management systems and Internet strategies at organizations including American Express, J.P. Morgan, Borden Foods, U.S. Tobacco, BBDO Worldwide, Foxwoods Resort Casino and the State University of New York at Purchase. The testing at DeepStorage Labs is informed by that real world experience.</p><p>He has been a frequent contributor to <em>Network Computing</em>&nbsp;and&nbsp;<em>InformationWeek</em>&nbsp;since 1999 and a speaker at industry conferences including Comnet, PC Expo, Interop and Microsoft's TechEd since 1990. He is the author of&nbsp;<em>Networking Windows</em>&nbsp;and co-author of&nbsp;<em>Windows NT Unleashed</em>&nbsp;(Sams).</p><p>He is co-host, with Ray Lucchesi of the monthly Greybeards on Storage podcast where the voices of experience discuss the latest issues in the storage world with industry leaders.&nbsp; You can find the podcast at: http://www.deepstorage.net/NEW/GBoS

SUBSCRIBE TO OUR NEWSLETTER
Stay informed! Sign up to get expert advice and insight delivered direct to your inbox

You May Also Like


More Insights