Sun debuts FlashFire, calls record books
2TB array boasts record-beating bandwidth, apparently
Sun has announced its solid state FlashFire products - an apparently record-breaking 2TB array and a 96GB card.
The F5100 array, some details of which were revealed in September, has up to 1.92TB of single-level cell NAND flash, originally thought to be supplied by STEC but now from Marvell, and comes as a 1U rackmount shelf. Host bus adapter firmware comes from LSI. The product has 64 SAS lanes (16 x 4-wide ports), 4 domains and SAS zoning, It can perform 1.6m read IOPS and 1.2M write IOPS, with a bandwidth of 12.8GB/sec which, Sun says, is a world record.
It also says the read IOPS figure is equivalent to 3,000 hard drives in 14 rack cabinets. The F5100 uses 1/100th of the space and power, at 300 watts, of such a collection of hard drives.
The individual 24GB flash modules are rated at a mean time before failure (MTBF) of 2m hours. Eighty of them are used to make the 1.92TB model. A 480GB product has 20 and an 960GB variant 40. The 480GB product does 397K read IOPS and the 960GB product 795K read IOPS; performance, both of IOPS and bandwidth, increasing as flash modules are added. Latency stays the same though, at 0.41ms read latency and 0.28ms write latency. Endurance is given out as six years, assuming 50 per cent reads and 50 per cent writes.
Sun positions this as a database accelerator for Oracle and MySQL. The unit can be zoned into 16 partitions, one for each of up to 16 hosts. The device can form part of a Sun ZFS hybrid storage pool, embracing solid state and hard disk drives. It is managed through StorageTek Common Array Manager Software, and protected by a super capacitor instead of by a battery backup system. If power fails then a capacitative energy system flushes the DRAM to flash and prevents data loss.
F20 PCIe card
Sun has used the same flash technology to build a 94GB flash card, the F20, that connects to a server's PCIe bus. It is composed of four 24GB flash modules, presented as four separate 24GB disks, and contains a DRAM cache. It can do more than 100,000 read IOPS and 84,000 write IOPS.
The F20 also contains host bus adapter technology with a SAS disk controller for connecting up to eight internal SAS or SATA devices. Sun's ZFS can include the F20 in a hybrid storage pool. Protection is provided by a super capacitor system as with the F5100.
The F20 competes with similar PCIe-connected flash from Fusion-io and others, and acts as a read I/O accelerator for applications needing generally faster I/O from a set of disk drives. It is also conceptually similar to NetApp's Performance Acceleration Module or PAM. The F5100 is more akin to Texas Memory Systems' Flash-based RamSan products where an entire database is stored in flash.
The F5100 is priced from $45,995 but F20 pricing and availability were not revealed. ®
that's right, no need to RAID the SSD
That's right, SSDs are not "Inexpensive Disks” that should be used in a “Redundant Array” – what RAID stands for.
SSD from Sun Storage working with ZFS are for data caching, not persistent data storage.
This is different SSD from what you use in Apple laptops.
Caching tier do ECC checksum, not RAID parity checksum.
SSD eliminates the need for short stroke and wide striping of HDDs for higher performance, but does not replace HDDs for data storage.
Dunno why folks are obsessed with MTBF of SSD – nothing lasts forever, and so what if a caching device fails? You still have your data on HDDS and the broken SSDs are covered by the warranty. (Yeah, don’t buy SSD on the street.)
MTBF means Mean Time Between Failures
@Steven Jones -- Don't do it..RAID on Flash is a no-go
The update-in-place nature of RAID-5/6 striping is about the worst kind of poisonous workload for NAND-Flash
The RAID write-penalty is unacceptable on SSD because it turns every write operation into N+1 write operations, meaning (a) you'll wear out your Flash N+1 times faster, and (b) Flash write performance is awful when you turn off the DRAM write buffer, which MUST be done for any kind of parity-based RAID.
If you lose a cached write in the RAID-5 or RAID-6 scenario, you've corrupted your data -- and will probably never know it until you try to rebuild.
Furthermore, Flash writes are so slow compared to reads that the on-disk write-buffers fill up fast, and array performance goes down the tubes. FYI, this is why IBM/STEC used mirroring on thier recent SPC-1 benchmark result.
If you want to test this yourself, build a 4+1 RAID set on Intel X25-E SLC Flash and let it run for a bit against IOmeter with a 50/50 Read/Write workload. Leave the write-cache enabled.
After a couple of minutes the SSDs are performing like cheap SATA disks.
Then, write a big zip-file out to the array and pull the plug on it just after the copy completes. Pull a disk, power up the array again and test the zip file...you will find garbage for data.