IBM moves flash storage contained in the box — 12.8 TB of it — to work alongside RAM and cut data movement latencies.

IBM has unveiled what it calls its next generation of cloud and massive data servers, the X6 series. The x86-based line will run Intel’s latest Ivy Bridge processor in four- and 8-way configurations and should be full of flash memory.

IBM has used flash or solid-state drives as adjuncts to spinning disks and RAID arrays outside the server. However, now it has migrated flash contained in the server case. To take action, it needed to redesign flash memory sticks to slot into the random access memory’s dual-inline memory slots on a server’s motherboard. It then designed motherboards with room for 12.8 TB of flash, in addition to 6 TB to twelve TB of RAM.

That’s a primary for IBM and a primary for the industry, IBM spokesmen said. By redesigning how solid-state memory fits into the server, IBM can add more of it than competitors and will produce a bigger combination of RAM and flash, that have similar performance characteristics, contained in the case. That, in turn, will allow for top-speed I/Os and applications designed for optimum performance when exposed to multiple TBs of flash and RAM.

The benefit of having flash alongside the RAM is that the info at the flash modules stays there when the server is shut down and is offered once it’s started up. The entire contents of a database system or NoSQL system resembling MongoDB can remain instantly available and shut to the CPUs. RAM and flash storage share an identical memory channel contained in the server case, leading to low latencies.

[Want more at the use of solid state in cloud servers? See Rackspace Adds SSDs To Cloud Servers.]

IBM calls its server flash architecture eXFlash memory channel storage.

In general, system operations accelerate when large flash storage devices are outside the server because they may quickly upload data without facing the seek time required by the top of a spinning disk drive. Putting flash storage contained in the case leads to a 10X performance improvement over flash outside the case.

Stuart McRae, high-end System x business line manager, told us IBM’s X6 server line can load data from flash storage contained in the server into RAM or CPU in 5 to ten microseconds, in comparison to a minimum of 100 microseconds for flash storage outside the case. A microsecond is one millionth of a second.

The latency between external storage and CPU is more typically measured in milliseconds, or thousandths of a second. The adaptation in thousandths or millionths of a second may not seem significant until one thinks of the savings accumulated over and over in each operation of the system. One microsecond is to a second, as an example, as one second is to 11.5 days, McRae said. The repeated performance gain of saving 5 to ten microseconds per data movement is big.

That data movement speed, along side new Intel Ivy Bridge processors, will make the X6 line an immense data analytics and NoSQL system of choice, McRae said. It contains thrice “the scalable memory” of the biggest servers offered by competitors.

These products was designed to host large numbers of virtual machines, or multiple virtualized database systems per server. That makes them a candidate for work inside cloud service suppliers, a market where IBM want a bigger share. Its x86 enterprise servers were largely absent from the general public cloud, which likes bare-bone models which are stripped of all redundant or unnecessary parts and supply easy air circulation.

In addition, IBM has designed the CPU and memory components of the X6s as “books” or modules that load into and might be pulled out of front of the server. If the landlord wants an upgrade, “you’re taking out the book and installed new memory and CPU, without touching the remainder of the system.” The X6 servers turns into portion of IBM’s x86 System x and PureSystems product lines, McRae said.

In a second announcement, IBM said it has produced the FlashSystem 840, a flash storage device to work alongside the X6 servers. It’s miles IBM’s second all-flash storage device, following the FlashSystem 820 launched in April. The 840 offers 48 TB of storage in a 2u rack device — twice up to the 820 — together with twice the information movement bandwidth.

Mike Kuhn, business line executive for FlashSystems, gave us one measure of what meaning: The 840 can perform 1.1 million I/O operations per second. That may support 40,000 bank card transactions per second.

The storage device, just like the X6 servers, is designed to support high-volume throughput in big data applications, cloud operations, or e-commerce transactions, Kuhn told us. “plenty of people are still selling flash with spinning disks. Just a handful of individuals have built a flash system from the floor up.” (IBM also produces storage systems that combine flash and disks.)

IBM acquired Texas Memory Systems in 2011 and used its engineering and intellectual property to provide the FlashSystem 820. The corporate has since expanded the Texas Memory engineering team by 400%, Kuhn said. He expects devices just like the 840 to head big data applications into real-time operations, gathering reams of knowledge on websites as thousands of users generate events, and providing feedback in time to deal with the users while they’re there.

Kuhn and McRae said final pricing had not been determined at the X6 servers or the FlashSystem 840.

Charles Babcock is an editor-at-large for InformationWeek, having joined the publication in 2003. He’s the previous editor-in-chief of Digital News, former software editor of Computerworld, and previous technology editor of Interactive Week.

Private clouds are moving rapidly from concept to production. But some fears about expertise and integration still linger. Also within the Private Clouds Step Up issue of InformationWeek: The general public cloud and the steam engine have more in common than you can think (free registration required).

More Insights