LSI’s Take on Data Center Flash
LSI Corp. has launched a new blog that covers (among other things) flash storage. It’s only natural – the company’s SandForce subsidiary is riding high on the SSD wave and LSI’s HBAs are finding widespread use, both internally and externally, in the production of two-hop PCIe SSDs.
A recent post called “What are the Driving Forces Behind Going Diskless” by LSI Fellow Rob Ober outlines the leading arguments that data center managers use against using solid state storage, debunking them one by one:
- Flash is a commodity
- $8K PCIe card in a $4K server – really?
- The stuff wears out and disks last forever
- Crazy expensive $/bit
Mr. Ober says he has been spending a lot of time with the biggest brand-name datacenters in the world, ones that have 200k+ servers, and from 1.5-7 million disks. He observes that they are very cost-conscious, focusing on both CapEx and OpEx, and know that anything that increases production or reduces cost is interesting, and it’s best if you can achieve both. One of those datacenters is diskless, and others are moving in that direction.
It’s a good post that is well worth reading.
Objective Analysis clients are already familiar with these arguments, since we have been explaining the same reasoning for the past six years. SSDs bring compelling improvements to performance and cost control. One element that Ober missed that our clients are aware of is the trade-off of SSDs against DRAM. It’s another way to improve performance while lowering system capital cost and power consumption that is already gaining widespread acceptance in the data center community, and will eventually be adopted in the PC.