FalconStor and Violin get cache in hand
SAN accelerator is go
SaaS data loss: The problem you didn’t know you had
FalconStor and Violin have come up with a flash-based SAN accelerator featuring a dedicated cache area to speed up writes.
Take one Violin 1010 flash memory appliance (a PCI-connected solid state drive) and have it fitted to a server running FalconStor's Network Storage Server Gateway and storage management software - the resulting box is called the NSS SAN Accelerator and is sited inline between app servers and the Storage Area Network (SAN) they access.
The FalconStor software moves hot data from the SAN into the Violin SSD cache and that accelerates reads. But it also sets aside a specific area of the cache, the SafeCache, to store incoming writes.
The cunning thing is that the app doing the write is told it's completed before it's actually written to disk, so it doesn't have to wait for the disk write to complete before moving on to do other work.
The combination of read and write caching provides the overall SAN performance boost. FalconStor says its lab tests show that a cache size of one per cent of the SAN's total capacity will deliver a better than 100 per cent increase in performance.
This type of SSD caching is similar to that carried out by IBM's SAN Volume Controller using STEC flash SSDs.
FalconStor's flashy NSS SAN Accelerator will be available in Europe from May from its channel partners, with a list price of $32,000. That includes a 500GB Violin 1010 solid state drive. ®
COMMENTS
yes
Yes, most storage arrays use DRAM for write caching, but are limited to a few gigabytes due to the cost of DRAM. SAN Accelerator augments this with 500 GB to 4TB of flash memory.
flash as cache?
"app doing the write is told it's completed before it's actually written to disk"
Uhmmm... don't most storage arrays worth their salt do exactly that?

IT infrastructure monitoring strategies
Agentless Backup is Not a Myth
Top 10 SIEM implementer’s checklist
Steps to Take Before Choosing a Business Continuity Partner
Enabling efficient data center monitoring