Texas Memory Systems Pushes SSD Envelope

By Michael Feldman

May 18, 2011

Enterprise SSD maker Texas Memory Systems (TMS) has been kicking up some dust lately, announcing record-breaking performance results with its RamSan-630 product and launching its latest PCIe flash offering. Specifically, TMS recently put up some rather impressive numbers in two key benchmarks established by the Storage Performance Council (SPC). And then this week, the company introduced its next-generation PCIe flash memory product, the RamSan-70, which incorporates the latest NAND technology from Toshiba.

First to the SPC results. The TMS RamSan-630, a network-attached 3U enclosure that can hold up to 10 TB of flash memory, was able to deliver a record-setting 400,503 IOPS on the SPC-1 benchmark at a price performance of $1.05/IOPS. That beat the SPC competition by a wide margin, specifically, two disk-based IBM systems that approached the TMS gear in total IOPS, but were far behind in price-performance. In the first case, a 6-node IBM System Storage SAN Volume Controller v5.1 hit 380,489 total IOPS, but at $18.83/IOPS, a far cry from the RamSan-630. The less expensive IBM Power 595 with PowerVM delivered 300,993 IOPS, but still at a cost of $10.77/IOPS.

The SPC-1 benchmark is geared for random I/O, so plays into the strength of flash memory, but it’s still a rather grueling test for SSD technology. The workload uses about 60 percent writes and is sustained for eight hours. Also, the device needs to be prefilled with data at the start. That kind of environment is not very SSD-friendly. To get a sense of that, the RamSan-630 SPC-1 result is less than half the peak performance of the 1,000,000 IOPS that TMS claims for the solution.

The SPC-2 benchmark, on the other hand, is geared for sequential I/O workloads like online transaction processing (OLTP), so one wouldn’t expect great results from an SSD setup. (In fact, the RamSan-630 is the first SSD-based system submitted for this SPC benchmark). But the TMS system managed a respectable 8.3 GB/second average over the three SPC-2 workloads: large file processing, large database query and video-on-demand.

That wasn’t as fast as the top-rated IBM System Storage DS8800 system at 9.7 GB/second, but the RamSan-630’s price-performance of $49.37 undercut the $270.38 result for the IBM system rather handily. An HP and Hitachi storage system both delivered about 8.7 GB/second, but at a price-performance of around $187 they were still considerably more expensive that the RamSan box.

According to Jamon Bowen, TMS’s director of sales engineering, the SPC-2 results point to a new use case for enterprise SSD offerings. Whereas before, flash-based systems were only seen as a way to optimize storage IOPS, Bowen says we’re now beginning to see their utility for traditional bandwidth-sensitive workloads, as found in high performance computing and other applications suffering from the data deluge.

Bowen does admit that their SSD box is still going to be about three or four times as expensive as a typical JBOD array built for low cost and MB/second-level bandwidth. But according the him, when you start needing 7 or 8 GB/second for relatively modest-sized storage systems, it’s just cheaper to go the SSD route. “You can now build high bandwidth systems out of SSDs at a lower price than you can disk systems,” he says. “You don’t get the same amount of capacity, but it’s not that far behind.”

Currently there are only a few submissions from SSD vendors to the Storage Performance Council, but at least half of the new members to the group are using flash technology. “I would not be surprised if this becomes the de facto SSD proving ground,” says Bowen.

Besides the network-attached SSD arrays like the RamSan-630, TMS also offers direct-attached flash via its enterprise-class PCIe products like the 450 GB RamSan-20. Those cards are meant to be plugged into workstations and servers for applications that can make use of very fast local storage. For cluster setups, it offers a high performance I/O without the overhead imposed by a network storage solution.

In particular, for large clusters using a shared nothing architecture, the PCIe SSD offer maximum scalability, minimum I/O latency, and the best price-performance. In HPC, these apply to Hadoop-type analytics applications and direct-attached Lustre nodes. In general, any highly parallel application that can partition data along with the computation is ideal for PCIe-based SSDs says Bowen.

This week, the TMS came out with its next-generation PCIe flash card, the RamSan-70. It incorporates Toshiba’s latest 32nm SLC NAND flash chip (which has already been incorporated into the RamSan-630). Compared to the older RamSan-20, the new card delivers about 3 times the performance (330,000 sustained IOPS), twice the usable capacity (900 GB, although a 450 GB version is available too). And because of the smaller NAND geometries, the RamSan-70 takes up half the space of its predecessor and costs about 15 to 20 percent less per GB.

Like many of the latest enterprise-class SSD devices on the market, the RamSan-70 is upping its game in flash longevity and robustness. According to TMS, at 15 percent of the device’s maximum bandwidth (18 TB/day), the RamSan-70 can be expected to last 30 years. Unlike some of its competition, however, all TMS flash management is done with in-house technology. In the case of the RamSan-70, the TMS-developed flash controller logic is implemented with an on-board Virtex-6 FPGA. Bowen says the advantage of such a setup is that the FPGA can be easily reprogrammed — relative to spinning a new ASIC — when new NAND technology is adopted or if new management features are added.

TMS intends to sell the RamSan-70 through an OEM model. In truth, the sales approach is somewhat more loosely defined. According to Bowen, they’ll deal with anyone who can commit to at least 100 cards year. As mentioned before, pricing will be 15 to 20 percent less per GB than the now obsolete 450 GB RamSan-20, which sold for around $18,000. TMS has yet not announced any OEM partners.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

UCSD, AIST Forge Tighter Alliance with AI-Focused MOU

January 18, 2018

The rich history of collaboration between UC San Diego and AIST in Japan is getting richer. The organizations entered into a five-year memorandum of understanding on January 10. The MOU represents the continuation of a 1 Read more…

By Tiffany Trader

New Blueprint for Converging HPC, Big Data

January 18, 2018

After five annual workshops on Big Data and Extreme-Scale Computing (BDEC), a group of international HPC heavyweights including Jack Dongarra (University of Tennessee), Satoshi Matsuoka (Tokyo Institute of Technology), Read more…

By John Russell

Researchers Measure Impact of ‘Meltdown’ and ‘Spectre’ Patches on HPC Workloads

January 17, 2018

Computer scientists from the Center for Computational Research, State University of New York (SUNY), University at Buffalo have examined the effect of Meltdown and Spectre security updates on the performance of popular H Read more…

By Tiffany Trader

HPE Extreme Performance Solutions

HPE and NREL Take Steps to Create a Sustainable, Energy-Efficient Data Center with an H2 Fuel Cell

As enterprises attempt to manage rising volumes of data, unplanned data center outages are becoming more common and more expensive. As the cost of downtime rises, enterprises lose out on productivity and valuable competitive advantage without access to their critical data. Read more…

Fostering Lustre Advancement Through Development and Contributions

January 17, 2018

Six months after organizational changes at Intel's High Performance Data (HPDD) division, most in the Lustre community have shed any initial apprehension around the potential changes that could affect or disrupt Lustre Read more…

By Carlos Aoki Thomaz

UCSD, AIST Forge Tighter Alliance with AI-Focused MOU

January 18, 2018

The rich history of collaboration between UC San Diego and AIST in Japan is getting richer. The organizations entered into a five-year memorandum of understandi Read more…

By Tiffany Trader

New Blueprint for Converging HPC, Big Data

January 18, 2018

After five annual workshops on Big Data and Extreme-Scale Computing (BDEC), a group of international HPC heavyweights including Jack Dongarra (University of Te Read more…

By John Russell

Researchers Measure Impact of ‘Meltdown’ and ‘Spectre’ Patches on HPC Workloads

January 17, 2018

Computer scientists from the Center for Computational Research, State University of New York (SUNY), University at Buffalo have examined the effect of Meltdown Read more…

By Tiffany Trader

Fostering Lustre Advancement Through Development and Contributions

January 17, 2018

Six months after organizational changes at Intel's High Performance Data (HPDD) division, most in the Lustre community have shed any initial apprehension aroun Read more…

By Carlos Aoki Thomaz

When the Chips Are Down

January 11, 2018

In the last article, "The High Stakes Semiconductor Game that Drives HPC Diversity," I alluded to the challenges facing the semiconductor industry and how that may impact the evolution of HPC systems over the next few years. I thought I’d lift the covers a little and look at some of the commercial challenges that impact the component technology we use in HPC. Read more…

By Dairsie Latimer

How Meltdown and Spectre Patches Will Affect HPC Workloads

January 10, 2018

There have been claims that the fixes for the Meltdown and Spectre security vulnerabilities, named the KPTI (aka KAISER) patches, are going to affect applicatio Read more…

By Rosemary Francis

Momentum Builds for US Exascale

January 9, 2018

2018 looks to be a great year for the U.S. exascale program. The last several months of 2017 revealed a number of important developments that help put the U.S. Read more…

By Alex R. Larzelere

ANL’s Rick Stevens on CANDLE, ARM, Quantum, and More

January 8, 2018

Late last year HPCwire caught up with Rick Stevens, associate laboratory director for computing, environment and life Sciences at Argonne National Laboratory, f Read more…

By John Russell

Inventor Claims to Have Solved Floating Point Error Problem

January 17, 2018

"The decades-old floating point error problem has been solved," proclaims a press release from inventor Alan Jorgensen. The computer scientist has filed for and Read more…

By Tiffany Trader

US Coalesces Plans for First Exascale Supercomputer: Aurora in 2021

September 27, 2017

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, in Arlington, Va., yesterday (Sept. 26), it was revealed that the "Aurora" supercompute Read more…

By Tiffany Trader

Japan Unveils Quantum Neural Network

November 22, 2017

The U.S. and China are leading the race toward productive quantum computing, but it's early enough that ultimate leadership is still something of an open questi Read more…

By Tiffany Trader

AMD Showcases Growing Portfolio of EPYC and Radeon-based Systems at SC17

November 13, 2017

AMD’s charge back into HPC and the datacenter is on full display at SC17. Having launched the EPYC processor line in June along with its MI25 GPU the focus he Read more…

By John Russell

Nvidia Responds to Google TPU Benchmarking

April 10, 2017

Nvidia highlights strengths of its newest GPU silicon in response to Google's report on the performance and energy advantages of its custom tensor processor. Read more…

By Tiffany Trader

IBM Begins Power9 Rollout with Backing from DOE, Google

December 6, 2017

After over a year of buildup, IBM is unveiling its first Power9 system based on the same architecture as the Department of Energy CORAL supercomputers, Summit a Read more…

By Tiffany Trader

Fast Forward: Five HPC Predictions for 2018

December 21, 2017

What’s on your list of high (and low) lights for 2017? Volta 100’s arrival on the heels of the P100? Appearance, albeit late in the year, of IBM’s Power9? Read more…

By John Russell

Chip Flaws ‘Meltdown’ and ‘Spectre’ Loom Large

January 4, 2018

The HPC and wider tech community have been abuzz this week over the discovery of critical design flaws that impact virtually all contemporary microprocessors. T Read more…

By Tiffany Trader

Leading Solution Providers

Perspective: What Really Happened at SC17?

November 22, 2017

SC is over. Now comes the myriad of follow-ups. Inboxes are filled with templated emails from vendors and other exhibitors hoping to win a place in the post-SC thinking of booth visitors. Attendees of tutorials, workshops and other technical sessions will be inundated with requests for feedback. Read more…

By Andrew Jones

Tensors Come of Age: Why the AI Revolution Will Help HPC

November 13, 2017

Thirty years ago, parallel computing was coming of age. A bitter battle began between stalwart vector computing supporters and advocates of various approaches to parallel computing. IBM skeptic Alan Karp, reacting to announcements of nCUBE’s 1024-microprocessor system and Thinking Machines’ 65,536-element array, made a public $100 wager that no one could get a parallel speedup of over 200 on real HPC workloads. Read more…

By John Gustafson & Lenore Mullin

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Flipping the Flops and Reading the Top500 Tea Leaves

November 13, 2017

The 50th edition of the Top500 list, the biannual publication of the world’s fastest supercomputers based on public Linpack benchmarking results, was released Read more…

By Tiffany Trader

How Meltdown and Spectre Patches Will Affect HPC Workloads

January 10, 2018

There have been claims that the fixes for the Meltdown and Spectre security vulnerabilities, named the KPTI (aka KAISER) patches, are going to affect applicatio Read more…

By Rosemary Francis

GlobalFoundries, Ayar Labs Team Up to Commercialize Optical I/O

December 4, 2017

GlobalFoundries (GF) and Ayar Labs, a startup focused on using light, instead of electricity, to transfer data between chips, today announced they've entered in Read more…

By Tiffany Trader

HPC Chips – A Veritable Smorgasbord?

October 10, 2017

For the first time since AMD's ill-fated launch of Bulldozer the answer to the question, 'Which CPU will be in my next HPC system?' doesn't have to be 'Whichever variety of Intel Xeon E5 they are selling when we procure'. Read more…

By Dairsie Latimer

Nvidia, Partners Announce Several V100 Servers

September 27, 2017

Here come the Volta 100-based servers. Nvidia today announced an impressive line-up of servers from major partners – Dell EMC, Hewlett Packard Enterprise, IBM Read more…

By John Russell

  • arrow
  • Click Here for More Headlines
  • arrow
Share This