flash memory science & technology SSD technology

Follow-Up – EETimes on SanDisk UltraDIMMs

Image representing IBM as depicted in CrunchBase
Image via CrunchBase

“The eXFlash DIMM is an option for IBM‘s System x3850 and x3950 X6 servers providing up to 12.8 TB of flash capacity. (Although just as this story was being written, IBM announced it was selling its x86 server business to Lenovo for $2.3 billion).”

Sadly it seems the party is over before it even got started in the sales and shipping of UltraDIMM equipped IBM x86 servers. If Lenovo snatches up this product line, I’m sure all the customers will still be perfectly happy but I worry about that level of innovation and product testing that led to the introduction of UltraDIMM may be slowed.

I’m not criticizing Lenovo for this, they have done a fine job taking over the laptops and desktop brand from IBM.  The motivation to keep on creating new, early samples of very risky and untried technologies seems to be more IBM’s interest in maintaining it’s technological lead in the data center. I don’t know how Lenovo figures into that equation. How much will Lenovo sell in the way of rackmount servers like the X6 line? And just recently there’s been rumblings that IBM wants to sell off it’s long history of doing semi-conductor manufacturing as well.

It’s almost too much to think R&D would be given up by IBM in semi-conductors. Outside of Bell Labs, IBM’s fundamental work in this field brought things like silicon on insulator, copper interconnects and myriad other firsts to ever smaller, finer design rules. While Intel followed it’s own process R&D agenda, IBM went its own way too always trying to find advantage it’s in inventions. Albeit that blistering pace of patent filings means they will likely never see all the benefits of that Research and Development. At best IBM can only hope to enforce it’s patents in a Nathan Myhrvold like way, filing law suits on all infringers, protecting it’s intellectual property. That’s going to be a sad day for all of us who marveled at what they demoed, prototyped and manufactured. So long IBM, hello IBM Global Services.

Enhanced by Zemanta

IBM Goes Modular And Flashy With X6 Systems – Timothy Prickett Morgan

The memory channel storage modules were developed by SanDisk in conjunction with Diablo Technologies, and are called UltraDIMM by SanDisk. The modules put flash memory created by SanDisk (which has a flash partnership with Toshiba) that has a SATA interface on a memory stick. Diablo has created a chipset that converts the SATA protocol to the DDR3 main memory protocol, and SanDisk has created a driver for a server BIOS that makes it look like a normal disk storage device to the system and its operating system. (Enterprise Tech – Timothy Prickett Morgan)
Image representing Diablo Technologies as depi...
Image by None via CrunchBase

Big news, big news coming to a server near you. A new form factor Flash Memory product has been secretly developed and is being sampled by folks out East in the High Frequency Stock Trading market (the top of the food chain in IT needs for latency speed of transactions). Timothy Prickett Morgan (formerly of The Register) has included details from IBM‘s big annoouncement of its Intel based X6 series servers. This new form factor is the result of a memory controller made by Diablo Technologies. SanDisk has engineered the actual final product that ties the memory into the Diablo designed memory controller. However this product is not available on the open market and has been going through sampling and testing with possible high end end users and customers who have need for such a large, high speed product in a DDR DRAM memory module. Sizes, and speeds are pretty large all around. The base modules come in 200GB or 400GB form factors and fit a typical DDR-3 DIMM module. IBM and SanDisk have done a number of special tweaks on the software/firmware to pull the most I/O with the lowest latency out of these modules when installed on an X6 server. The first-gen X6 will have roughly 12 DIMM slots available with some DRAM and Ultra-DIMMs populating those slots. However things get REALLY interesting when the second-gen X6 hits the market. IBM will be doubling the amount of DIMM slots to 24 and will be upping the core count available on the 4U top of the line x6 server. When that product hits the market the Ultra-DIMM will be able to populate the majority of the DIMM slots and really start to tear it up I think when it comes to I/O and analytics processing. SanDisk is the exclusive supplier, manufacturer and engineering outfit for this product for IBM with no indication yet of when/if they would ever sell it to another OEM server manufacturer.

Given the promise this technology has and that an outfit like Diablo Technologies is vaugely reminiscent of an upstart like SandForce who equally upset the Flash Memory market about 6 years ago, we’re likely to see a new trend. SATA SSDs are still slowly creeping into the consumer market, PCIe Flash memory products are being adopted by the top end consumer market (Apple’s laptops and the newest desktops). Now we’ve got yet another Flash memory product that could potentiall sweep the market the Ultra-DIMM. It will however take some time and some competing technology to help push this along (SandForce was the only game in town early on and early adopters help subsidize the late adopters with higher prices). Given how pared back and stripped down DIMM slots are generally in the consumer market it may be a while before we see any manufacturers attempt to push Ultra-DIMM as a consumer product. Same goes for the module sizes as they are shipped today. Example: the iMac 27″, Apple has gone from being easily upgraded (back in the Silver Tower, G4 CPU days) to nearly not upgradeable (MacBook Air) and the amount of space needed in their cases to allow for addition or customization through an Ultra-DIMM add-on would be severly constrained. It might be something that could be added as a premium option for the newest Mac Pro towers. And even then that’s very hopeful and wishful thinking on my part. But who knows how quickly this new form factor and memory controller design will infiltrate the computer market? It is seemingly a better moustrap in the sense of the boost one sees in performance on a more similar, more commoditized Intel infrastructure. Wait and see what happens.

Enhanced by Zemanta
data center flash memory SSD technology

All-flash IBM V7000 smashes Oracle/Sun ZFS box • The Register

Some Storage Thingy
Some Storage Thingy (Photo credit: mootown)

What would happen if we replaced those 16 disk-based V7000s with all-flash V7000s? Each of the disk-based ones delivered 32,502.7 IOPS. Let’s substitute them with 16 all-flash V7000s, like the one above, and, extrapolate linearly; we would get 1,927,877.4 SPC-1 IOPS – nearly 2 million IOPS. Come on IBM: go for it.

via All-flash IBM V7000 smashes Oracle/Sun ZFS box • The Register.

That’s right, IBM is understanding the Flash-based SSD SAN market and is making some benchmark systems to help market its disk arrays. Finally we’re seeing some best case scenarios for these high end throughput monsters. It’s entirely possible to create a 2Million IOPS storage SAN. You just have to assemble the correct components and optimize your storage controllers. What was once a theoretical maximum throughput (1M IOPs) is now achievable without anything more than a purchase order and an account representative from IBM Global Services. It’s not cheap, not by a longshot but your Big Data project or OLAP with Dashboard may just see orders of magnitude increases in speed. It’s all just a matter of money. And probably some tweaking via an IBM consultant as well (touche).

Granted that IBM doesn’t have this as a shipping product isn’t really the point. On paper what can be achieved by mixing matching enterprise storage appliances and disk arrays and software controllers is beyond what any other company is selling IS the point. There’s a goldmine to be had if anyone outside of a high frequency trading skunkworks just shares a little bit of in-house knowledge product familiarity. No doubt it’s not just the network connections that make things faster it is the IOPs that will out no matter what. Write vs. Read and latency will always trump the fastest access to an updated price in my book. But I don’t work for a high-frequency trading skunkworks either, I’m not privy to the demands made upon those engineers and consultants.  But still we are now in the best, boldest time yet of nearly too much speed on the storage front. Only thing holding us back is the network access times.