Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


Call for Participation: MSST Mass Storage Conference 2018

The 34th International Conference on Massive Storage Systems and Technologies (MSST 2018) has issued its Call for Participation. The event takes place May 14-16 in Santa Clara, California. “The conference invites you to share your research, ideas and solutions, as we continue to face challenges in the rapidly expanding need for massive, distributed storage solutions. Join us and learn about disruptive storage technologies and the challenges facing data centers, as the demand for massive amounts of data continues to increase. Join the discussion on webscale IT, and the demand on storage systems from IoT, healthcare, scientific research, and the continuing stream of smart applications (apps) for mobile devices.”

Agenda Posted: OpenPOWER 2018 Summit in Las Vegas

The OpenPOWER Summit has posted its speaker agenda. Held in conjunction with IBM Think 2018, the event takes place March 19 in Las Vegas. “The OpenPOWER Foundation is an open technical community based on the POWER architecture, enabling collaborative development and opportunity for member differentiation and industry growth. The goal of the OpenPOWER Foundation is to create an open ecosystem, using the POWER Architecture to share expertise, investment, and server-class intellectual property to serve the evolving needs of customers and industry.”

Video: Computing Challenges at the Large Hadron Collider

CERN’s Maria Girona gave this talk at the HiPEAC 2018 conference in Manchester. “The Large Hadron Collider (LHC) is one of the largest and most complicated scientific apparata ever constructed. “In this keynote, I will discuss the challenges of capturing, storing and processing the large volumes of data generated at CERN. I will also discuss how these challenges will evolve towards the High-Luminosity Large Hadron Collider (HL-LHC), the upgrade programme scheduled to begin taking data in 2026 and to run into the 2030s, generating some 30 times more data than the LHC has currently produced.”

Video: Deep Reinforcement Learning and Systems Infrastructure at DeepMind

In this video from HiPEAC 2018 in Manchester, Dan Belov from DeepMind describe the company’s machine learning technology and some of the challenges ahead. “DeepMind Inc. is well known for state of the art Deep Reinforcement Learning (DRL) algorithms such as DQN on Atari, A3C on DMLab and AlphaGo Zero. I would like to take you on a tour of challenges we encounter when training DRL agents on large workloads with hundreds of terabytes of data. I’ll talk about why DRL poses unique challenges when designing distributed systems and hardware as opposed to simple supervised learning. Finally I’d like to discuss opportunities for DRL to help systems design and operation.”

Adaptive Computing rolls out Moab HPC Suite 9.1.2

Today Adaptive Computing announced the release of Moab 9.1.2, an update which has undergone thousands of quality tests and includes scores of customer-requested enhancements. “Moab is a world leader in dynamically optimizing large-scale computing environments. It intelligently places and schedules workloads and adapts resources to optimize application performance, increase system utilization, and achieve organizational objectives. Moab’s unique intelligent and predictive capabilities evaluate the impact of future orchestration decisions across diverse workload domains (HPC, HTC, Big Data, Grid Computing, SOA, Data Centers, Cloud Brokerage, Workload Management, Enterprise Automation, Workflow Management, Server Consolidation, and Cloud Bursting); thereby optimizing cost reduction and speeding product delivery.”

Intel Rolls out new 3D NAND SSDs

Today, Intel announced the Intel SSD DC P4510 Series for data center applications. As a high performance storage device, the P4510 Series uses 64-layer TLC Intel 3D NAND to enable end users to do more per server, support broader workloads, and deliver space-efficient capacity. “The P4510 Series enables up to four times more terabytes per server and delivers up to 10 times better random read latency at 99.99 percent quality of service than previous generations. The drive can also deliver up to double the input-output operations per second (IOPS) per terabyte.”

Interview: European cHiPSet Event focuses on High-Performance Modeling and Simulation for Big Data Applications

The cHIPSet Annual Plenary Meeting takes place in France next month. To learn more, we caught up with the Vice-Chair for the project, Dr. Horacio González-Vélez, Associate Professor and Head of the Cloud Competency Centre at the National College of Ireland. “The plenary meeting will feature a workshop entitled “Accelerating Modeling and Simulation in the Data Deluge Era”. We are expecting keynote presentations and panel discussions on how the forthcoming exascale systems will influence the analysis and interpretation of data, including the simulation of models, to match observation to theory.”

TACC Podcast Looks at AI and Water Management

In this TACC podcast, Suzanne Pierce from the Texas Advanced Computing Center describes her upcoming panel discussion on AI and water management and the work TACC is doing to support efforts to bridge advanced computing with Earth science. “It’s about letting the AI help us be better decision makers. And it helps us move towards answering, discussing, and exploring the questions that are most important and most critical for our quality of life and our communities so that we can develop a future together that’s brighter.”

Video: Intel and NVIDIA at Congressional Hearing on Artificial Intelligence

In this video, Information Technology Subcommittee Chairman Will Hurd begins a three-part hearing on Artificial Intelligence. “Over the next three months, the IT Subcommittee will hear from industry professionals such as Intel and NVIDIA as well as government stakeholders with the goal of working together to keep the United States the world leader in artificial intelligence technology.”

MIT helps move Neural Nets back to Analog

MIT researchers have developed a special-purpose chip that increases the speed of neural-network computations by three to seven times over its predecessors, while reducing power consumption 94 to 95 percent. “The computation these algorithms do can be simplified to one specific operation, called the dot product. Our approach was, can we implement this dot-product functionality inside the memory so that you don’t need to transfer this data back and forth?”