Boosting Discovery Speed and Productivity

February 4, 2019

While price and performance have typically been used when evaluating, selecting, and operating HPC storage systems, increasingly productivity is a more important parameter.  In essence, productivity means getting the best insights and answers in the shortest possible wall clock time within an envelope of other factors such as cost, and when the result is needed.

Productivity not only takes into account processing speed, the ability to keep CPUs satiated, and fast data transfers from storage to compute resources, but also factors in reliability, availability, speed to insight, and the completion of the greatest number of compute jobs in a given time. One additional productivity criteria to consider is how long a system can remain useful without excessive downtime and maintenance. Essentially, the discovery process needs high-performance systems that can quickly run the most compute and data analysis jobs over many years without excessive downtime or maintenance.

NCSA Blue Waters: The HPC high productivity poster child

A good example that embodies what is needed in the most demanding discovery environments is the National Center for Supercomputing Applications Blue Waters system. The Blue Waters team partnered with the Cray supercomputing company for compute and data storage to help achieve the high productivity goals.

From the start, those managing the Blue Waters program focused on productivity. Its mission is to enable science and engineering that cannot be done otherwise, and to greatly improve time to insight.  Examples of things not even feasible before Blue Waters include:

  • all atom simulations of viruses using over 100 million atoms
  • simulation of the first billion years of the universe evolution and galaxy formation after the big bang, which required all the memory on the Blue Waters compute system
  • producing high resolution digital elevation models of 1/3 of the earth in less than two years.

Some teams report a factor of ten or more increases in productivity while others report an infinite increase because they can do things that prior to Blue Waters were simply not possible.

Beyond raw computer power, the system had to have a tightly integrated, high-performance storage system that could support the compute data demands. Additionally, the entire compute and storage infrastructure had to be tightly integrated, flexible, optimized, and be highly reliable to avoid diminishing the quality of services.

One great challenge with computing systems used for scientific discovery is sustaining productivity over the years. Each year, the modeling, simulations, and analysis gets more complex, more granular, and makes use of vastly more data.

According to Bill Kramer, Principle Investigator and Director of the Blue Waters Project, “Despite its scale, Blue Waters is incredibly reliable.  Over the last project year, Blue Waters had only one unscheduled system wide interruption, due to a campus wide power issue, and has a 99.7% scheduled uptime.  Mean Time Between System-wide Interrupt have been four to eight months for the past several years. Even more importantly, (knock on wood somewhere please), is the node and disk reliability.  On a daily basis, the individual node failure rate is below 1.5 nodes/day (0.006% of all the nodes) and a drive failure rate continuing below 0.43 drives/day (0.0026% of all the disks). What that means is Blue Waters is running better now than it did in the first couple of years of operations and there is no indication that will change in the next couple of years.”

Supercomputing systems must scale in compute power, as well as storage capacity and performance. In every field, new lab equipment, satellites, telescopes and sensors have produced higher-resolution and finer granular data. In the life sciences, next-generation sequencers have become faster (producing more data in a given time) and much less expensive to operate (allowing many more sequences to be done in given time). Storage systems have had to keep pace with the data volume growth. And their data analysis systems had to scale in performance.

Blue Waters has kept pace with the demands of the scientific research community. Today, Blue Waters continues to be a discovery mainstay for the open research community as a workhorse system at NCSA. Since March 2013, Blue Waters has delivered over 26 billion core-hours to scientific research. A key to this success has been the ability to scale the system’s storage bandwidth and metadata to accommodate the enormous amounts of data being analyzed.

“The storage system requirements on Blue Waters are extremely ambitious.  The 1.1 TB/s sustain I/O bandwidth for Blue Waters is still unmatched by any other production open system in the world today. Combined with 36 racks of ClusterStorTM (36+of raw PB of storage PB) and over 250 PB of near-line storage, Blue Waters is still the most data capable system in the HPC world.” explained Kramer.

The key to HPC productivity: Find the right technology partner

Building a fast and highly productive system for discovery requires the integration of many advanced technologies. For leading-edge systems, that means incorporating solutions that are new to the market. Most organizations simply do not have the across-the-board expertise in every element, nor the in-house skill set to evaluate, select, and integrate the elements and then optimize and maintain the system’s performance.

That’s where an experienced technology partner can help. Cray designed, architected, and built the Blue Waters extreme scale system and maintains and improves the system to keep it at the leading edge of productive computing.

The system’s ClusterStor storage is a good example of how Blue Waters can remain a high performer and continue to be a productive resource. The system’s productivity remains high due to the ClusterStor storage’s high-performance and high level of resiliency.  Furthermore, the Blue Waters, with the ClusterStor storage system is a leading enabler of being able to converge big data and extreme computing workload on the same system running very large modeling to extreme data intensive and machine/deep learning workload.

“The ClusterStor organization has had the privilege to partner with the Blue Waters team for the past 6 years. As a result, we have carried forward the lessons learned into the latest ClusterStor storage solutions”, according to Don Grabski, who has been with the ClusterStor product management team for 7 years.  The ClusterStor L300 and L300N line of parallel file system storage represents the latest HPC storage solution. This storage is a perfect match for the most demanding discovery workloads. ClusterStor parallel file systems balance the value equation by delivering the exact performance, speed, scalability, data protection, and availability to match an organization’s requirements and budget demands.

The systems deliver enterprise-level performance with more capacity, fewer drives, less need for IT support and more data access. Furthermore, ClusterStor technology optimizes performance productivity and system availability, accelerating time to insight.

The current ClusterStor product line includes:

  • ClusterStor L300 Storage System, which is an all-HDD Lustreâ It achieves performance requirements with the lowest number of HDDs, enclosures, and racks by maximizing the performance of each storage device.
  • ClusterStor L300N Storage System, which is a hybrid SSD/HDD solution with flash-accelerated NXD software that redirects I/O to the appropriate storage medium. It delivers cost-effective, consistent performance on mixed I/O workloads while shielding the application, file system, and users from complexity through transparent flash acceleration.
  • ClusterStor L300F Storage System, which is a scalable storage unit that provides the opportunity to add Flash storage pool creating a truly hybrid system. The L300F is designed and optimized to overcome the latency experienced by rotating media – the remaining IOPS bottleneck for Lustreâ.

The ClusterStor line includes engineered HPC Storage Solution features including:

  • Integrated software and hardware solution
  • Test validation
  • Management
  • Support automation

Taking these features together, ClusterStor storage systems enable the fast time-to-results and sustained performance needed in today’s most demanding discovery environments. The systems also scale to meet future demands by offering a way to increase capacity without impacting performance or availability seamlessly.

To learn more about increasing HPC productivity, visit:

https://www.cray.com/products/storage/clusterstor

 

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