Cray Cluster Connect (C3) Lustre Storage for Linux

November 25, 2013

Cray’s expert storage solutions are now available for popular x86 Linux clusters.  Cray Cluster Connect (C3) offers a complete Lustre storage solution for x86 Linux. C3 brings Supercomputing-class scalability to data- and I/O-hungry Linux clusters.

Cray’s approach to I/O optimization builds on holistic system expertise developed across the entire I/O stack, from applications down to storage.  Most storage vendors have expertise in a single specialized area (typically block or file storage). Cray’s expertise spans three interdependent disciplines—compute, networking, and storage—coming together to form adaptive yet massively scalable storage solutions.

Cray optimizes I/O through:

  1. Application workload profiling:  ensuring the I/O pattern is well characterized for optimal performance
  2. End-to-end system architectures:  all aspects of the system – compute, networking, and storage – are optimized for applications
  3. I/O testing and qualification at scale:  Cray pre-qualifies and extensively tests systems against a wide variety of I/O characteristics to simulate user I/O patterns.

Cray storage solutions are well suited to data hungry supercomputers and x86 Linux clusters.  Cray works proactively with customers to understand the application and I/O requirements.  “Cray makes sure the storage solution meets the customer’s requirements, “ says Mark Swan, principal performance engineer at Cray.  “If a customer’s workloads are not taking full advantage of available bandwidth, simply adding hardware or software is a Band-Aid. However, I have seen far too many customer workloads that do not perform the kinds of optimal I/O that would take advantage of the file system’s capabilities and that has nothing to do with fast or slow zones.”

Through Cray Cluster Connect (C3), Cray delivers complete Lustre storage solutions for popular x86 Linux clusters.  Customers choose their storage platform (Cray Sonexion, DDN SFA, or NetApp E-Series) and Cray optimizes the I/O and throughput.   Unlike conventional storage solution providers, Cray is able to optimize the I/O up the entire stack, from the storage up to the client.

“Cray configures systems to be extremely flexible from a purely I/O perspective,” says senior Cray engineer, John Kaitschuck. “We can configure as much capacity and I/O bandwidth as needed by a customer for a given application and couple that with a properly configured compute capability to achieve the desired I/O-to-compute balance for a given application.”

The type of I/O and application often dictates the network and storage design. “By proper analysis of the workload profile and configuration of the required associated capacity and bandwidth, Cray is able to optimally scale I/O as a customer’s needs grow,” adds Kaitschuck.

Workloads often fight for bandwidth—and I/O contention can take many forms.  “Cray storage architectures provide multiple pathways for data so that there is no single point of failure and no single point of contention,” says Swan.  “We also devise techniques so that I/O contention in one part of the file system does not interfere with other parts of the file system.  These techniques not only isolate failures but also isolate areas of contention from affecting the entire workload.”

“Cray is continually developing utilities to analyze the I/O path,” says Steve Woods, a senior practice leader in Cray’s storage division.  “Infiniband monitoring ensures there are no errors that might affect the performance of the network and to ensure the fabric operates at its anticipated bandwidth.”

“This testing includes a wide variety of I/O characteristics,” says Woods.  “Our tests simulate user I/O patterns. This testing is constantly evolving because of changes in hardware as it relates to disk drives, arrays, servers, and even the network interconnect technologies. Not only are various combinations of hardware are tested but in addition the various levels of software are also tested to insure there are no performance regressions associated with the different layers of software.”

Optimizing the I/O system is often related to improper tuning.  Tuning can be related to clients, networks, and storage.  This may be as simple as increasing certain buffers or balancing parameters for caching or command sequence to disk. In the case of routed I/O, like LNET routing in Lustre, various parameters may need tuning to handle peak small data floods or large streaming of data.”

Cray’s solutions ensure “full performance access” to the I/O subsystem, minimizing bottlenecks in the architecture, says Woods.  “This work includes analyzing the performance of all components of the IO subsystem—from LUNs to controllers to servers to networks across routers (if required) and to the clients. Part of the analysis includes testing and turning the software stack”, he added.

Cray also offers its own scalable storage line, Cray Sonexion.  Sonexion scales I/O bandwidth in lock-step with storage capacity.   The basic unit of storage in Sonexion—including performance and capacity—is the Scalable Storage Unit (SSU).  An SSU combines high capacity, high density storage with two all-active embedded controllers serving Lustre performance.   For customers that need to add more capacity than performance, Cray offers Expansion Storage Units (ESUs).

In cases where customers need to optimize capacity or performance for a given workload, the Lustre™ File System by Cray (CLFS) can be utilized with DDN SFA or NetApp E-Series storage.  Lustre solutions by Cray are installed separately on x86 servers and provides a high degree of configuration flexibility.

Cray storage solutions help customers utilize the right storage and get results faster.  The Cray storage solution at the NCSA Blue Waters provides an example of I/O optimization at the petascale, for diverse applications. “During the Blue Waters acceptance period,” according to Swan, “Cray worked with the PSDNS turbulence application.  By analyzing its check-pointing methods, Cray optimized the I/O path to reduce check-pointing and application runtime.”

Since Cray fully tests and qualifies all solutions, users obtain an optimal experience, where I/O can be scaled as needed, and systems work as specified.   The product of the all the hard work done by Cray is that systems operate predictably.  By giving customers choice in a storage platform, the ROI matches the customers’ requirements.  In petascale scenarios, Cray reduces the storage footprint by up to 50%–and scales I/O to 1TB/s in a single file system.

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