Texas Memory Lets Flash Flag Fly

By John E. West

April 30, 2009

Unlike many of the companies in today’s flash-based solid state disk (SSD) market, Texas Memory Systems has been in the SSD business for more than 30 years. Back in the late 70s they were building SSDs for seismic field data collection, and in the 80s the company’s gear found its way into supercomputers from makers like Floating Point Systems and, later, Cray.

In 2000 Texas Memory Systems (TMS) began to focus on the enterprise SSD market with RAM-based devices (flash memory was still too expensive to build large drives from). These devices needed batteries to preserve data in the event of a system interruption, and some versions even included hard drives to maintain critical data. But they had application in the enterprise where they were just what the CIO ordered, for example storing and providing super fast access to metadata for large filesystems. San Diego Supercomputer Center deployed a RamSan 325 as the metadata server in its 500 TB SAM-QFS filesystem, and dramatically improved its performance — in one specific example from that case study, backup time fell from 21 hours to 34 minutes. Over the next several years this line of products was improved by the addition of native Fibre Channel and then InfiniBand interfaces, but the cost of RAM kept the capacities relatively low compared to hard drives and kept a lid on the market.

As the consumer digital media market pushed down the price of flash memory, many companies started to take advantage of this new platform to build SSDs. In 2007 TMS joined them with the RamSan 500, a 2 TB device in a system form factor with RAID protection for its flash modules, a Fibre Channel interface and a RAM cache. The RamSan solution is different from some of the other less expensive offerings in the market in that it uses single-level cell (SLC) flash memory, which is faster and more reliable than its consumer-oriented cousin, multi-level cell (MLC) flash (it’s MLC that you’ve got in your digital camera).

The RamSan 500 allowed TMS to give its customers higher bandwidth and more capacity for a lower price and opened up new markets and applications for its customers. Where its RAM-based SSD cost $300/GB and put 2 TB in 16U, the new flash-based product costs only $75/GB and put the same amount of storage in 4U. This price/performance opened new applications, and customers are today using the RamSan 500 on clusters for working storage, to host metadata for large filesystems, and to hold very large datasets in memory for data-intensive computing applications. Customers are also using RamSan devices to host video on demand application (large block random reads) and in render farms (small block random reads).

The RamSan 500 remains the core of the TMS product offering for SSD today, but the company is building out new solutions for broader application. The RamSan 20 was announced in March, and is currently shipping. The 20 is a PCI-Express-based SSD that presents with 450 GB of storage for $40/GB. It is designed for smaller clusters and can allow large memory jobs (search databases, for example) to move in core, or to be run on smaller clusters when the number of nodes in the cluster is driven by the amount of memory needed for a job. Woody Hutsell, president at Texas Memory Systems, explained that they’ve received a lot of interest from application vendors who are interested in selling both their software and the hardware to run it effectively. We’ve seen a similar model for GPU-based hardware acceleration plus application software from Acceleware. Hutsell says that this solution can make a lot of sense, especially for non-traditional HPC users where “solutions are more important than pieces.”

Although fast for reads, flash-based memory is still usually slower on writes, and this creates a problem of balance in the system. TMS addresses part of this difference in the RamSan 20 by over-provisioning capacity. Although the card presents as 450 GB of storage to the system, it actually has 640 GB of capacity. The extra chips lets the company’s support software on the PCI card accelerate writes by doing more work in parallel (TMS claims that it can support 40,000 to 50,000 IO/s on write on this device), and the extra chips also allow for better wear leveling. TMS is also concerned about eliminating the impact on host resources of the RamSan 20. The card doesn’t require system memory or CPU from its host, relying instead on its own onboard PPC processor and RAM for functions like flash table management.

Rounding out the company’s SSD offerings today is the RamSan 620, just announced this month. This is a 2U rackmount Fibre Channel device (with IB expected not later than Q3 according to Hutsell) that provides 5 TB of storage for $44/GB.

Although these devices present new opportunities for users and system architects to achieve great performance, it’s still important to remember that flash isn’t for every application today. It is still more expensive than hard disks, and there are situations where they outperform their flash-based cousins. And applications need to have a requirement for very fast random access to data sets to make flash-based SSDs the right decision. Also, error detection and correction, a technology well-understood for RAM, is even more important with flash than with RAM because flash chips are delivered with known bad blocks. As companies look at MLC and higher density SLC, they will have to deploy better ECC algorithms.

Subscribe to HPCwire's Weekly Update!

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

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire