insideHPC ReportIn this video from SC17, Bruce Tulloch from BitScope describes an low-cost Raspberry Pi cluster that LANL can use to simulate large-scale supercomputers.
"The BitScope Pi Cluster Modules system creates an affordable, scalable, highly parallel testbed for high-performance-computing system-software developers. The system comprises five rack-mounted BitScope Pi Cluster Modules consisting of 3,000 cores using Raspberry Pi ARM processor boards, fully integrated with network switching infrastructure."
A quest to help the systems software community work on very large supercomputers without having to actually test on them has spawned an affordable, scalable system using thousands of inexpensive Raspberry Pi nodes. It brings a powerful high-performance-computing testbed to system-software developers, researchers, and others who lack machine time on the world’s fastest supercomputers.
“It’s not like you can keep a petascale machine around for R&D work in scalable systems software,” said Gary Grider, leader of the High Performance Computing Division at Los Alamos National Laboratory, home of the Trinity supercomputer. “The Raspberry Pi modules let developers figure out how to write this software and get it to work reliably without having a dedicated testbed of the same size, which would cost a quarter billion dollars and use 25 megawatts of electricity.”
Grider conceived the new solution to provide the systems software community with an inexpensive testbed of similar scale to the largest supercomputers being deployed. Designed and built by BitScope and distributed by SICORP, the Pi Cluster Modules enable developers to build, scale and test software before launching it on Trinity, Crossroads, and other large systems. The modules have other potential applications from education to the internet of things.
The BitScope system consists of five rack-mounted Pi Cluster Modules, each with 150 four-core nodes of Raspberry Pi ARM processor boards. They are fully integrated with network switching infrastructure. With a total of 750 CPUs or 3,000 cores working together, the system gives developers exclusive time on an inexpensive but highly parallelized platform for test and validation of scalable systems software technologies.
“Having worked with Raspberry Pi for quite some time, I’ve long thought it the ideal candidate to build low-cost cloud and cluster computing solutions for research and education,” said Bruce Tulloch, CEO of BitScope. “When SICORP approached us with Gary’s plans, we jumped at the opportunity to prove the concept.”
The Raspberry Pi is an inexpensive, credit-card sized computer that can be connected to a keyboard and TV to do just about anything a typical desktop computer can do. It was developed by the Raspberry Pi Foundation in the United Kingdom to help spread computing in education and in developing countries. The price for a basic model starts at about $25 and consumes only a handful of watts of power.
Beyond HPC system-software development, additional applications for the Pi Cluster Modules include better simulation of large-scale sensor networks, with flexible I/O to connect the actual sensor devices; HPC network topology research, to improve production performance; and the internet of things.
Looking around for a solution to the challenges facing HPC systems software developers, Grider said, he “suddenly realized the Raspberry Pi was an inexpensive computer using 2 to 3 watts that you could use to build a several-thousand-node system large enough to provide a low-cost, low-power testbed to enable this R&D.” Yet he could not find a suitable densely packaged Raspberry Pi system on the market.
“It was just people building clusters with Tinker Toys and Legos,” said Grider, a widely respected innovator in the HPC world. So he turned to SICORP of Albuquerque, N.M., to collaborate on a solution. Then they jointly worked with BitScope of Australia to develop easily scaled rack-mounted units.
An Affordable Supercomputing Testbed based on Raspberry PiinsideHPC Report2017-11-19 | In this video from SC17, Bruce Tulloch from BitScope describes an low-cost Raspberry Pi cluster that LANL can use to simulate large-scale supercomputers.
"The BitScope Pi Cluster Modules system creates an affordable, scalable, highly parallel testbed for high-performance-computing system-software developers. The system comprises five rack-mounted BitScope Pi Cluster Modules consisting of 3,000 cores using Raspberry Pi ARM processor boards, fully integrated with network switching infrastructure."
A quest to help the systems software community work on very large supercomputers without having to actually test on them has spawned an affordable, scalable system using thousands of inexpensive Raspberry Pi nodes. It brings a powerful high-performance-computing testbed to system-software developers, researchers, and others who lack machine time on the world’s fastest supercomputers.
“It’s not like you can keep a petascale machine around for R&D work in scalable systems software,” said Gary Grider, leader of the High Performance Computing Division at Los Alamos National Laboratory, home of the Trinity supercomputer. “The Raspberry Pi modules let developers figure out how to write this software and get it to work reliably without having a dedicated testbed of the same size, which would cost a quarter billion dollars and use 25 megawatts of electricity.”
Grider conceived the new solution to provide the systems software community with an inexpensive testbed of similar scale to the largest supercomputers being deployed. Designed and built by BitScope and distributed by SICORP, the Pi Cluster Modules enable developers to build, scale and test software before launching it on Trinity, Crossroads, and other large systems. The modules have other potential applications from education to the internet of things.
The BitScope system consists of five rack-mounted Pi Cluster Modules, each with 150 four-core nodes of Raspberry Pi ARM processor boards. They are fully integrated with network switching infrastructure. With a total of 750 CPUs or 3,000 cores working together, the system gives developers exclusive time on an inexpensive but highly parallelized platform for test and validation of scalable systems software technologies.
“Having worked with Raspberry Pi for quite some time, I’ve long thought it the ideal candidate to build low-cost cloud and cluster computing solutions for research and education,” said Bruce Tulloch, CEO of BitScope. “When SICORP approached us with Gary’s plans, we jumped at the opportunity to prove the concept.”
The Raspberry Pi is an inexpensive, credit-card sized computer that can be connected to a keyboard and TV to do just about anything a typical desktop computer can do. It was developed by the Raspberry Pi Foundation in the United Kingdom to help spread computing in education and in developing countries. The price for a basic model starts at about $25 and consumes only a handful of watts of power.
Beyond HPC system-software development, additional applications for the Pi Cluster Modules include better simulation of large-scale sensor networks, with flexible I/O to connect the actual sensor devices; HPC network topology research, to improve production performance; and the internet of things.
Looking around for a solution to the challenges facing HPC systems software developers, Grider said, he “suddenly realized the Raspberry Pi was an inexpensive computer using 2 to 3 watts that you could use to build a several-thousand-node system large enough to provide a low-cost, low-power testbed to enable this R&D.” Yet he could not find a suitable densely packaged Raspberry Pi system on the market.
“It was just people building clusters with Tinker Toys and Legos,” said Grider, a widely respected innovator in the HPC world. So he turned to SICORP of Albuquerque, N.M., to collaborate on a solution. Then they jointly worked with BitScope of Australia to develop easily scaled rack-mounted units.
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletterExploring Cognitive AI with Intel Labs’ Vasudev Lal | AI Hardware & Edge AI Summit 2024insideHPC Report2024-10-16 | At the 2024 AI Hardware & Edge AI Summit in San Jose, we spoke with Vasudev Lal, Principal AI Research Scientist at Intel Labs, for an in-depth look at the latest advancements in Cognitive AI. Lal provided a tour of Intel Labs’ groundbreaking work in this rapidly evolving field, discussing how Cognitive AI is pushing the boundaries of machine learning, human-computer interaction, and intelligent systems.
In this interview, Lal shares key insights into Intel’s research, offering a glimpse into the future of AI technology and its applications across various industries.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastinsideBIGDATA Rebrands as insideAI News: Insights from Editor-in-Chief Daniel Gutierrez | July 2024insideHPC Report2024-10-14 | On July 2, 2024, insideHPC’s sister publication, insideBIGDATA, officially relaunched as insideAI News, reflecting the evolution of data science into the widespread application of artificial intelligence across industries. In this interview, Editor-in-Chief Daniel Gutierrez shares insights into the revamped platform, which offers comprehensive coverage of AI technology and implementation strategies in sectors such as healthcare, security, financial services, automation, and agriculture.
A seasoned data scientist with nearly 25 years in the field, Gutierrez brings a unique perspective to AI journalism. He discusses the key verticals insideAI News will cover, the trends driving large-scale data use, and the publication’s goal of serving as the go-to resource for enterprise executives and AI deployment specialists worldwide.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastAt ISC 2024 Quobyte Software Defined Storage Brings HPC and Hyperscaler Heritage to AI WorkloadsinsideHPC Report2024-10-14 | At ISC 2024 in Hamburg, we spoke with Bjorn Kolbeck, CEO and co-founder of Quobyte, a high-performance storage company focused on software-defined scale-out storage for AI workloads. Kolbeck shared how Quobyte brings its HPC heritage and hyperscaler technology to the enterprise, delivering efficient, software-only storage solutions for demanding HPC and AI environments.
Kolbeck also introduced Quobyte’s new distributed File Query Engine, designed to enhance the management of large-scale storage by providing high-speed querying of file system metadata for enterprises and research organizations. Additionally, he highlighted customer success stories, including Siemens Health Care, showcasing how Quobyte’s storage technology is driving innovation.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastInside Eviden’s HPC-AI Leadership: New Contracts, TOP500 Success, and Quantum Advances | ISC 2024insideHPC Report2024-10-14 | At ISC 2024, we toured the Eviden booth, a central hub of excitement as Europe’s leading HPC-AI systems vendor. Emmanuel Le Roux, Eviden Group SVP and Head of Advanced Computing & AI, kicked off the tour by sharing key updates on recent contract wins in Brazil, Denmark (Novo Nordisk), and Germany, including the extension of Eviden’s exascale contract with the Julich Supercomputing Centre.
Le Roux also highlighted Eviden’s strong performance on the TOP500 list of the world’s most powerful supercomputers and their leadership in energy efficiency, with the JEDI development system (part of the upcoming Jupiter exascale supercomputer) ranked no. 1 on the Green500 list.
The tour continued with updates from senior managers on Eviden’s diverse HPC-AI server offerings, from edge AI inference to data center servers for model training, as well as their advancements in quantum computing and the Jarvice as-a-service cloud product line.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastAdvancing HPC-AI with oneAPI: Andrew Richards on Codeplay’s Role in Intel’s Ecosystem | ISC 2024insideHPC Report2024-10-14 | At ISC 2024, we caught up with Andrew Richards, CEO of Codeplay Software, an Edinburgh-based company acquired by Intel two years ago. In this interview, Richards discusses Codeplay’s pivotal role in supporting the ecosystem for Intel’s oneAPI—a unified programming model that offers a common developer experience across various accelerator architectures.
Known for its expertise in SYCL, the open standard used in oneAPI, Codeplay plays a key part in Intel's commitment to open collaboration and standards, which are crucial in today’s heterogeneous HPC-AI environments. Learn how Codeplay and Intel are driving scientific and business advancements through their dedication to open standards and cross-industry innovation.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastCooling the Future of Exascale Systems: Andrew Whitmore on Motivair’s Thermal Solutions | ISC 2024insideHPC Report2024-10-14 | At ISC 2024, we spoke with Andrew Whitmore, Vice President of Sales at Motivair, whose cutting-edge cooling technology is powering the first three U.S. exascale-class systems: Frontier, Aurora, and El Capitan. Motivair’s comprehensive thermal management solutions span from data center cooling to direct-to-chip liquid cooling, and include their innovative Dynamic Cold Plate technology, which ensures resilient performance while minimizing fluid contamination risks and supply chain disruptions.
In this interview, Andrew shares insights into Motivair’s current offerings and their vision for the future of cooling technology, as data center sustainability becomes an increasingly complex challenge. Learn how Motivair is leading the way in thermal solutions for the most demanding HPC environments.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastHPE’s Supercomputing Vision: Insights from Chief Architect Kirk Bresniker | ISC 2024insideHPC Report2024-10-08 | At ISC 2024, we had an insightful discussion with Kirk Bresniker, Chief Architect at Hewlett Packard Labs, who shared his vision for the future of HPE’s high-performance computing (HPC) systems. Bresniker highlights HPE’s position as a global leader in supercomputing, a claim boldly displayed at their ISC booth.
In this interview, Kirk explains how HPE's integration capabilities set them apart, welcoming components from their technology partners to build cutting-edge systems. He also delves into quantum computing’s role in working alongside classical HPC, revealing how HPE is preparing for the future of supercomputing.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastAMD’s Daniele Piccarozzi on Instinct GPUs, EPYC CPUs, and ROCm Software | ISC 2024 InsightsinsideHPC Report2024-10-08 | At ISC 2024, we sat down with Daniele Piccarozzi, Senior Market Development Manager for EMEA at AMD, to discuss AMD’s advancements in the Instinct GPU accelerator and EPYC CPU product lines, and their impact on AI and the broader HPC community. Daniele also highlights AMD's ROCm open software development stack, which is designed to optimize GPU programming and accelerate AI workloads.
In this interview, Daniele, who brings a global perspective from his experiences in Rome and the UK, provides a deep dive into AMD's approach to nurturing its HPC-AI ecosystem of customers and technology partners. Discover how AMD is shaping the future of high-performance computing and AI.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastCooling the Future: Andrew Whitmore on Motivair’s Exascale Systems & Thermal Solutions | ISC 2024insideHPC Report2024-10-08 | At ISC 2024, we caught up with Andrew Whitmore, Vice President of Sales at Motivair, whose advanced cooling technology powers the first three U.S. exascale-class systems: Frontier, Aurora, and El Capitan. Motivair’s comprehensive thermal management solutions span from data center cooling to direct-to-chip liquid cooling and the innovative Dynamic Cold Plate technology, offering resilient performance while minimizing fluid contamination risks and supply chain issues.
In this interview, Andrew shares insights into Motivair’s cutting-edge cooling solutions and the company’s vision for the future of thermal management as sustainability becomes a critical focus for data centers. Learn how Motivair is shaping the next generation of cooling technology for HPC and AI environments.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastMetas Vasu Agrawal on Liqids UltraStack GPU Dense Server SolutioninsideHPC Report2024-10-08 | Our own Doug Black spoke with Meta’s Vasu Agrawal about Liqid’s UltraStack GPU-dense server solution, with powered Dell servers equipped with up to 20 NVIDIA GPUs. Vasu explains how this state-of-the-art solution is driving AI and HPC innovation by offering unmatched GPU density and performance, allowing organizations to maximize flexibility and scalability for high-performance workloads. He specifically touches on his work with virtual reality and super photo-realism
In this video, Vasu shares how Meta is utilizing Liqid’s UltraStack with Dell-powered servers to supercharge their AI and VR projects, providing a look into the future of compute-intensive environments.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastAddressing AI Hardware-Software Challenges: Insights from Steve Conway of Intersect360 | ISC 2024insideHPC Report2024-10-08 | At ISC 2024 in Hamburg, we spoke with Steve Conway, Senior Analyst at Intersect360 Research, about the challenges organizations face when implementing the right hardware-software strategies for AI. Steve explains how these challenges have led Intersect360 to create a new AI users group, where companies can share insights, compare HPC-AI deployments, and learn from each other's experiences to avoid common pitfalls.
In this video, Steve dives into the growing need for collaboration in AI strategy and how the AI users group is helping organizations worldwide align their infrastructure with evolving AI demands.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastHPC & AI Trends in 2024: Insights from Hyperion Researchs Bob Sorensen | ISC 2024 RecapinsideHPC Report2024-10-08 | At ISC 2024 in Hamburg, we sat down with Bob Sorensen, SVP of Research at Hyperion Research, to discuss the latest trends in HPC and AI. Bob highlights key findings from Hyperion’s market studies, which show how organizations worldwide are rapidly implementing AI strategies—especially in generative AI—while grappling with a wide array of compute and infrastructure options.
In this video, Bob explores the challenges of navigating on-prem, cloud, and hybrid solutions, and how these choices are shaping the future of AI and HPC. Get exclusive insights into how businesses are tackling the complexities of AI deployment and infrastructure in 2024!
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastRevolutionizing Data Centers: DDC’s Hybrid Cooling Solutions with Chris Orlando | ISC 2024insideHPC Report2024-10-08 | At ISC 2024, we got an exclusive tour of the DDC booth with Co-founder and Chief Strategy Officer, Chris Orlando. In this insightful walkthrough, Chris explains how DDC's innovative hybrid air-liquid cooling cabinets are transforming data centers, especially for high-density HPC and AI workloads.
DDC’s cutting-edge technology ensures power efficiency, reduces build costs, and provides guaranteed airflow per kilowatt—while simplifying dense cabling deployments. Originally proven in Scale Matrix’s data centers, DDC’s platform is now available to enterprise organizations and service providers looking to upgrade to compute-intense environments. Watch to learn how DDC is paving the way for next-gen data center infrastructure!
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastAt ISC 2024: Lenovos Director of HPC and AI, Andreas Thomasch, on Reinventing HPC SustainablyinsideHPC Report2024-10-08 | At ISC 2024 in Hamburg, we caught up with Lenovo Director of HPC and AI in Europe Andreas Thomasch, who took us on a tour of the Lenovo booth while discussing the company's vision for "reinventing HPC" in ways that meet customer needs and that are sustainable.
His wide-ranging comments look at new HPC-AI servers from Lenovo, their distinctions in the marketplace, and the company's more than decade-long experience with liquid cooling, an area in which Lenovo has been a pioneer.
These HPC and AI trends will continue to be a growing topic at ISC 2024 and beyond. ___________________________________________________________________________ Stay Up-to-Date on HPC News: ➡️Facebook: facebook.com/insideHPCmediaLLC ➡️LinkedIn: linkedin.com/company/insidehpc-media-llc ➡️X: https://x.com/insideHPC
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastPreparing for ISC 2024: HPC Luminaries on Sessions, Trends, and News to Watch for in HamburginsideHPC Report2024-10-08 | The center of the HPC-AI world shifts next week to Hamburg and the ISC 2024 Conference, Sunday, May 12 – Thursday, May 16. The theme of this year’s event is “Reinventing HPC,” and indeed, High Powered Computing has been made over during the past 18 months by the overwhelming presence of generative AI, which has spurred tremendous demands on HPC-class infrastructures for model training and inference.
The hydra-headed implications of new AI capabilities – machine learning, deep learning, and LLM-driven genAI – are reverberating throughout the worlds of traditional HPC (AI for science, mod/sim) along with the broader world of HPC-AI, which is impacting every business and industrial sector throughout the public and private sectors around the world (did we leave anything out?). And then there's sustainability, not to mention social impacts.
These HPC_AI trends will be focal points at ISC 2024 and thus also of our panel of HPC experts – Earl Joseph of Hyperion Research, Lois Curfman McInnes of Argonne National Laboratory, and Addison Snell of Intersect360 Research – gathered here for a round-robin video discussion looking at the major sessions, keynotes and potential news developments in Hamburg next week. ___________________________________________________________________________ Stay Up-to-Date on HPC News: ➡️Facebook: facebook.com/insideHPCmediaLLC ➡️LinkedIn: linkedin.com/company/insidehpc-media-llc ➡️X: https://x.com/insideHPC
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastInterview with Co-Founder Sumit Puri: Liqid’s UltraStack Servers with up to 20 NVIDIA GPUsinsideHPC Report2024-10-08 | At GTC 2024 we caught up with Sumit Puri, co-founder of composable computing company Liqid, to discuss the company’s GPU-power UltraStack, which he said transforms 2U servers into high-density GPU systems powered by NVIDIA L40S GPUs.
High Power Computing and AI workloads’ insatiable demand for computing power are outstripping the capabilities of traditional servers, which are typically limited to between four and eight GPUs. The lack of GPU availability only exacerbates the challenge.
But as Puri explains in this video, Liqid UltraStack represents a shift in server design, making 2U servers into high-density GPU systems, supporting up to 20 NVIDIA L40S GPUs, which are more readily available.
He said Liqid has designed the UltraStack around Dell Technologies PowerEdge R760 & R7625 servers, in standalone and cluster-ready solutions.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastVirtual Webinar: Transforming Physics Simulations with Ansys and Microsoft AzureinsideHPC Report2024-09-20 | This video introduces Ansys Access on Microsoft Azure, a newly released option designed to streamline the transition to the cloud. This solution allows a hybrid approach, enabling you to offload heavy workloads to the cloud as and when required, providing an option that enhances your operational efficiency.At GTC 2024: Lenovo’s Blackwell-Powered ThinkSystem AI Portfolio Built in Collaboration with NVIDIAinsideHPC Report2024-09-16 | Our editor and chief sits with Pat Moley, Director of Marketing for HPC and AI at Lenovo, at the NVIDIA GTC 2024 conference in San Jose. In this interview, Pat talked about Lenovo’s new portfolio of server platforms that leverage the new NVIDIA Blackwell architecture’s generative AI Engine, alongside the faster NVIDIA NVLink interconnect and enhanced security capabilities.
ThinkSystem AI servers with NVIDIA B200 Tensor Core GPUs are set to power the era of generative AI for enterprises and the cloud. With up to 25X more real-time inferences to accelerate trillion-parameter language models, NVIDIA B200 GPUs are designed for demanding AI, data analytics, and HPC workloads.
Lenovo’s liquid- and air-cooled platforms support a range of different workload needs, supported by the company’s knowledge and experience in helping customers get the right configuration for their generative AI strategies.
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastAdvancing AI with Microsoft Azure at NVIDIA GTC with John LeeinsideHPC Report2024-09-16 | At GTC 2024, we spoke with John Lee, Microsoft Azure’s Principal Lead, AI Platforms & Infrastructure, about the incorporation of NVIDIA’s new Blackwell GPU architecture within Azure’s AI capabilities portfolio. John is not only a long-time HPC veteran, he’s also an AI and cloud computing enthusiast.
Here he talks about Microsoft as one of the first organizations to bring the power of NVIDIA Grace Blackwell GB200 and advanced NVIDIA Quantum-X800 InfiniBand networking to the cloud, delivering trillion-parameter foundation models for natural language processing, computer vision, speech recognition, and more.
Citing his two decades of experience in HPC (prior to Microsoft, he was at Intel and Cray), he said, “What makes HPC so different from rest of the industry is scalability, how big you run a job. And in AI, that directly translates because when you're trying to model these incredibly large and complex language models, scalability matters. So if you take a look at Azure, we've been building up this AI supercomputing infrastructure capability for many, many years.”
Expand Your HPC Knowledge: 🌐Visit the insideHPC website! insidehpc.com 👉Advertise With Us! insidehpc.com/advertise 🎧Listen to Our Podcasts: insidehpc.com/category/resources/podcastInterview: Cray CEO Peter Ungaro on The Exascale ReportinsideHPC Report2021-07-23 | In this video, Mike Bernhardt from The Exascale Report interviews Cray CEO Peter Ungaro.Xyratex at HPC WorkshopinsideHPC Report2021-07-23 | ...Effective MapReduce on LustreinsideHPC Report2021-07-23 | Unlisted videoComputational Engineering - Maintaining Technological Superiority for the Nations DefenseinsideHPC Report2021-07-23 | In this video, Dr. Reed Mosher presents Computational Engineering - Maintaining Technological Superiority for the Nation's Defense. Recorded at the National HPCC Conference 2011 in Newport.Slidecast: Opscode Hosted Chef Goes GA for the CloudinsideHPC Report2021-07-23 | In this slidecast, Opscode co-founder & CEO Jesse Robbins discusses the company's latest offering called Hosted Chef.QLogic Tri-Labs Linux Capacity Cluster AwardinsideHPC Report2021-07-23 | In this podcast, QLogic's Joe Yaworski discusses the company's recent Tri-Labs Linux Capacity Cluster Award. The company will provide InfiniBand for three major supercomputer centers, providing over 3 Petaflops of HPC computing capacity.Demo: Mathematica for OpenCL and CudainsideHPC Report2021-07-23 | Recorded at SIGGRAPH 2011.Intels Raj Hazra Talks ExascaleinsideHPC Report2021-07-23 | In this video, Intel's Raj Hazra talks about Exascale research at the company.sharanHI.movinsideHPC Report2021-07-23 | ...New Sustainability Metrics for Datacenters from the Green GridinsideHPC Report2021-07-23 | In this video, Nicolas Dube from the Green Grid describes new sustainability metrics for datacenters. Just as PUE has become a standard for measuring power efficiency, the new metrics take into account the environmental impact of a datacenter in terms of Energy Reuse Effectiveness (ERE), Water Usage Efficiency (WUE), and Carbon Usage Effectiveness (CUE).ERDCs Advanced Materials InitiativeinsideHPC Report2021-07-23 | In this video, Charles Robert Welch from the US Army Corps of Engineers, Engineers Research & Development Center (ERDC) presents: ERDC's Advanced Materials Initiative - Using HPC Simulations to Develop Super Fibers and Super Ceramics. Recorded at the HPC User Forum in San Diego on Sept. 8, 2011.HP EcoPod - Worlds Most Efficient DatacenterinsideHPC Report2021-07-23 | In this video, HP's Ed Turkel presents HP Pod 240a: World's Most Efficient Datacenter. Recorded at the HPC User Forum on Sept. 9, 2011 in San Diego.AMD HPC Industry Roundtable - Q&A SessioninsideHPC Report2021-07-23 | What: HPC Industry Roundtable (continental breakfast to be served) When: Wednesday, November 2, 2011 Where: Four Seasons Hotel, located at 757 Market Street, San Francisco, CA,
Moderator: Gartner: Carl Claunch, VP and Distinguished Analyst, Gartner
Participants: AMD: Chuck Moore, Corporate Fellow and Technology Group CTO Appro: Anthony Kenisky, VP of North American Sales Cray: Margaret "Peg" Williams, Sr. Vice President of HPC Systems Penguin Computing: Charles Wuischpard, President & CEO SGI: Bill Mannel, Vice President of Product Marketing Supermicro: Don Clegg, Vice President of Marketing and Business Development
Charlie: Trademarked HPC as a service.
Questions:
Will there be memory locality improvements for GPU architectures? Are grass roots efforts necessary for HPC? Is ARM a valid architecture for HPC? Bigger issues: With the Chinese rise in supercomputing, is the US prepared to win and should the government step up efforts?Interview: Todd Gresham from Xyratex on the Companys Big Play in HPCinsideHPC Report2021-07-23 | In this video, Todd Gresham from Xyratex discusses the company's push into HPC.Interview: Ken Claffey from Xyratex on how ClustStor Changes the Storage Game in HPCinsideHPC Report2021-07-23 | In this video, Ken Claffey from Xyratex discusses how ClustStor changes the storage playing field in HPC.Interview: Ken Claffey from Xyratex on how ClustStor Changes the Storage Game in HPCinsideHPC Report2021-07-23 | In this video, Ken Claffey from Xyratex discusses how ClustStor changes the storage playing field in HPC.Interview: Todd Gresham from Xyratex on the Companys Big Play in HPCinsideHPC Report2021-07-23 | In this video, Todd Gresham from Xyratex discusses the company's push into HPC.Whamclouds Brent Gorda on Lustre, Moving Forward in 2012insideHPC Report2021-07-23 | In this video, Brent Gorda, CEO of Whamcloud describes how the Lustre file system has started to move into the enterprise.Ira Weiny Vid 032612.MOVinsideHPC Report2021-07-23 | ...Evolving File Systems - APIs for Exascale and Big DatainsideHPC Report2021-07-23 | In this video, Galen Shipman leads a panel discussion on Evolving File Systems - APIs for Exascale and Big Data.
Recorded at the Open Fabrics Workshop on March 26, 2012 in Monterey, CA. Learn more at: openfabrics.org/press-room/upcoming-events/109.htmlintelligentlightHI.movinsideHPC Report2021-07-23 | ...Bright Cluster Manager and PBS Pro: Extending Clusters into the CloudinsideHPC Report2021-07-23 | In this video, Robert Stober from Bright Computing presents: Bright Cluster Manager and PBS Professional: Extending Clusters into the Cloud with Data Aware Scheduling.
"Mr. Stober will show how to easily extend on-premise clusters into the cloud, and manage these resources as part of the local cluster. The combined products deliver a seamless, intuitive solution for provisioning, scheduling, monitoring and managing the extended cluster and data within one intuitive GUI, and/or with a cluster management shell."
Recorded at the PBS Works User Group meeting in San Jose on Oct. 2, 2012. Learn more at: http://www.pbsworks.com/pbsug/agenda.htmlPeter Braam on Exascale I/O at EIOW at SC12insideHPC Report2021-07-23 | In this video, Peter Braam from Xyratex presents at the EIOW exascale meeting at SC12 in Salt Lake City. This segment shows the last 19 minutes of the talk, and to our knowledge, there is no footage or recording of the beginning.Gundam Style at Adaptive ComputinginsideHPC Report2021-07-23 | Gundam Style at Adaptive Computing booth at SC12.Steve Barber 2013 Interview (original Full Version)insideHPC Report2021-07-23 | ...Todd Gresham InterviewinsideHPC Report2021-07-23 | ...Ken Claffey InterviewinsideHPC Report2021-07-23 | I created this video with the YouTube Video Editor (http://www.youtube.com/editor)Rick Rasmussen InterviewinsideHPC Report2021-07-23 | ...Simon Johnson InterviewinsideHPC Report2021-07-23 | ...Steve Barber on 2012 at Xyratex Plus Strategy for 2013insideHPC Report2021-07-23 | ...Steve Barber on the Evolution of XyratexinsideHPC Report2021-07-23 | ...