Panasas Broadens Storage Portfolio

By John E. West

August 19, 2008

Based in Fremont, Calif., and founded in 1999, Panasas is a privately held company that builds parallel clustered storage solutions. Although the company’s legacy is in traditional high performance computing installations, Panasas now realizes 80 percent of its revenue from the commercial side of its business, where companies like Statoil, PetroChina, Ferrari, and others in finance, bioinformatics, and various data-intensive industries are seeing a need for faster processing of large quantities of data. What all of these users have in common is that the amount of data they produce and store is growing dramatically as the amount of compute capability available to solve problems grows. A quick romp through the TOP500 confirms that yes indeed, the number of processors available at both ends of the list is growing quickly, and the resulting data has to be stored somewhere.

But the phrase “stored somewhere” hides many messy details. The data has to be written fast enough to avoid slowing the pace of an ongoing simulation or delaying other work from being done on a system. Likewise, data needs to be fed into the processors fast enough to keep them busy doing work, not waiting around for data. In both cases we need to be assured that what we are reading is what was written, and likewise for the reverse.

Large-scale storage has evolved from traditional single-server NFS architectures though clustered-NFS and onto parallel clustered storage, where we are today. The single- and clustered-server NFS solutions both put a file server on the path between the data source and the sink, an architectural constraint that ultimately limits the performance these solutions. The important feature of the parallel clustered approach is that it removes the file servers from the critical data path between the sink and the source, eliminating a key performance bottleneck. Panasas competes against commercial offerings from IBM, NetApp, and others, as well as open source projects like Lustre.

The industry is highly fragmented, and the high degree of proprietary technology and lack of interoperability that reflect the state of the practice in parallel storage may soon come to an end as the work to develop the parallel NFS (pNFS) standard comes to a conclusion. NFS version 4.1 is expected to incapsulate the new pNFS standard later this year, which Panasas is keenly anticipating. The company is committed to the evolution and adoption of pNFS as a catalyst for the storage industry.

“We firmly believe that the direction for storage and the direction for our business is into that broader scale-out marketplace with parallel storage,” says Matt Reid, Panasas’s director of marketing. “And one of the key enablers of this…is pNFS. Where we’re at today is that every parallel storage offering in the marketplace is proprietary, and pNFS will remove what is somewhat of an inhibitor to broader adoption, and enable parallel storage to progress into this much larger market.”

A parallel clustered file system removes some of the architectural impediments that limit the rate at which data flows between a source and sink, but there are plenty of other issues to be dealt with. One of the most interesting of these is also related to capacity — this time not in the amount of data that needs to be managed, but in the sizes of the drives on which it is stored. The storage industry has long been dominated by RAID, a technology first introduced in a 1988 paper co-authored by Garth Gibson, currently CTO of Panasas. A RAID 5 implementation stripes data across multiple disks, improving the speed of reads and writes, and stores metadata about what is written — called parity data — to enable recovery in the event that a single disk fails.

The approach protects data from many types of errors, but can’t help at all in the event that a read error occurs when rebuilding a lost disk. This will happen to about one in every 10^14 bits, or 12.5 TB. That wasn’t such a big deal when hard drives topped out at 50 GB. As drives pass 1 TB each in shelves of six, seven, or more drives, however, unrecoverable read errors during disk rebuilds are becoming a statistical certainty, virtually guaranteeing data loss or the need to recover slightly older data from much slower tape backups. RAID 6 can fix this problem when only one disk suffers an error during rebuild, but not more, and is thus only a partial solution.

To circumvent the problem, Panasas developed a “tiered parity” solution. Tiered parity generates a parity sector from other sectors on a disk that can be used to recompute missing data in the event of a media error. This new level of protection decreases unrecoverable read errors to 1 bit in 10^18 to 10^19, an improvement of 10,000 times in the best case. Although the original announcement was made during SC07, the technology hasn’t been released in final product form until today, with its inclusion in ActiveScale 3.2. For a more detailed treatment of tiered parity technology, see our October 2007 article on the topic.

The company has also announced three new hardware products. Two of them, the ActiveStor 6000 and ActiveStor 4000, are new releases of previous products (the ActiveStor 5000 and 3000, respectively). The ActiveStor 6000 is the high-end Panasas offering, aimed at large-scale design, modeling and data visualization. This solution incorporates the most software capability in the company’s offering and features 20, 15, or 10TB shelves, a 40 GB cache/shelf, and, for the first time, integrated 10 GbE. The capacity per shelf on the 6000 is the same as the previous 5000 series, but the new product features double the cache per shelf, and with 10 GbE the company is claiming data rates of 600 MB/second per shelf, nearly twice that of the 5000 products. The ActiveStor 4000, though still pretty beefy, is a lighter version of the 6000 designed for what the company sees as a simulation and analysis use case. The 4000 is missing some of its sibling’s software capabilities, though none of its performance, and again doubles the cache size of its 3000 series predecessor.

The third new product, the ActiveStor 200, is what Reid refers to as a “product family of one,” with a 104TB unit of sale. Like the ActiveStor 6000 and 4000, the solution features the company’s ActiveScale and PanFS software stack, but offers storage for secondary applications, like backup, where good (but not great) reliability is needed at a lower price point. The lower cost is evident in the performance tradeoff for the device. The 200 features five shelves, each with a single GbE port, offering an aggregate bandwidth of 350 MB/second for the entire system.

System cost is reflected in capability. The ActiveStor 200 is priced at around $1.2/GB, while the high-end 6000 series comes in at around $5.00/GB. The company characterizes the 4000’s price point as “between the 6000 and the 200, but closer to the 6000.” The ActiveStor 200 is available now, with two large installations in the UK. Panasas says it has “orders on the books” for both the 6000 and the 4000, with delivery scheduled next month.

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!

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…

Nvidia Appoints Andy Grant as EMEA Director of Supercomputing, Higher Education, and AI

March 22, 2024

Nvidia recently appointed Andy Grant as Director, Supercomputing, Higher Education, and AI for Europe, the Middle East, and Africa (EMEA). With over 25 years of high-performance computing (HPC) experience, Grant brings a 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…

Houston We Have a Solution: Addressing the HPC and Tech Talent Gap

March 15, 2024

Generations of Houstonian teachers, counselors, and parents have either worked in the aerospace industry or know people who do - the prospect of entering the fi 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