RFHPC113: Alternative Processors for HPC

In this podcast, the Radio Free HPC team looks at why it’s so difficult for new processor architectures to gain traction in HPC and the datacenter.

Along the way, we try to take on the following questions:

  • ARM servers have been talked about for a few years now, but haven’t become anything near mainstream. Why is that? 
  • Can we categorize the challenges that non-x86 CPUs face?
  • Is HPC the right target market for ARM? Is ARMv8 going to move the needle?
  • What is new in ARMv8? What did Phytium announce this week at Hot Chips? What will Fujutsu use in the Post-K supercomputer?
Plus, we introduce a new regular feature for our show: The Catch of the Week.

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC112: Looking Back at IDF 2016

In this podcast, the Radio Free HPC team reviews the recent 2016 Intel Developer Forum.

“How will Intel return to growth in the face of a declining PC market? At IDF, they put the spotlight on IoT and Machine Learning. With new threats rising from the likes of AMD and Nvidia, will Chipzilla make the right moves? Tune in to find out.”

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC111: A First Look at HPE’s Pending Acquisition of SGI

In this podcast, the Radio Free HPC team looks HPE’s pending acquisition of SGI.

“At HPE, we are focused on empowering data-driven organizations,” said Antonio Neri, executive vice president and general manager, Enterprise Group, Hewlett Packard Enterprise. “SGI’s innovative technologies and services, including its best-in-class big data analytics and high performance computing solutions, complement HPE’s proven data center solutions designed to create business insight and accelerate time to value for customers.”

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC110: Machine Learning & Data Locality

In this podcast, the Radio Free HPC team looks at use cases for Machine Learning where data locality is critical for effective results.

Most of the Machine Learning hearing stories we hear involve a central data repository. Henry says he is not hearing enough about how Machine Learning is going to deal with the problem of massive data streams from things like sensors. Such data, he contends, will have to be processed at the source.

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC109: New Chip Architectures — Picking the Winners

In this podcast, the Radio Free HPC team welcomes Shahin Khan from OrionX to a discussion on chip architectures for HPC.

More and more new alternative architectures were in evidence at ISC in Germany this year, but what does it take for a chip architecture to be a winner? Looking back, chips like DEC Alpha had many advantages over the competition, but it did not survive.

Henry contends that a viable chip architecture needs a memory controller that has the muster to keep up with new memory hierarchies. Rich thinks it isn’t about features so much and that survival is all about the chip’s ecosystem.

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC108: First Look at the Sunway TaihuLight 93 Petaflop Supercomputer

In this podcast, Shahin Khan from OrionX joins the Radio Free HPC team for a look at the new TOP500 list of the world’s fastest supercomputers.

As announced this morning at ISC 2016, the fastest supercomputer on the TOP500 is new Sunway TaihuLight System in Wuxi, China. Developed by the National Research Center of Parallel Computer Engineering & Technology (NRCPC), TaihuLight scored a whopping 93 Petaflops on the LINPACK benchmark. To put that in perspective, that is nearly three times faster than the previous #1 system, the Tianhe-2 supercomputer, which has moved to #2 after ruling the roost for some three years or so TaihuLight is also five times faster than Titan, the 17 Petaflop machine at ORNL, which is still the fastest machine in the USA.

Sunway in Wuxi, China

The Sunway TaihuLight Suercomputer in Wuxi, China is the world’s fastest supercomputer.

Here is the rundown:

  • Linpack: 93 Petaflops (Rmax)
  • Peak performance: 125.4 Petflops (Rpeak)
  • Processor: Sunway SW26010 1.4 GHz processor
  • Cores per socket: 260
  • Instruction Set: RISC instruction set developed by Sunway
  • Interconnect: their TOP500 submission says “Sunway design” but Mellanox supplied the Host Channel Adapter (HCA) and switch chips. Sunway may not call it InfiniBand, but that is exactly what it is. China has political reasons for characterizing the overall system domestic technology.
  • Cabinets: 40 Water-cooled cabinets, each with 3 Petaflops of peak performance
  • Power consumption: 15.27 Megawatts
  • Mflops/watt: 6051

Download the MP3 * Subscribe on iTunes * RSS Feed

Sign up for our insideHPC Newsletter

RFHPC107: HPC News Roundup for Friday the 13th

In this podcast, the Radio Free HPC team looks at the news highlights for the week leading up to Friday the 13th of May, 2016:

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC105: The GTC’16 Run Down

In this podcast, the Radio Free HPC team recaps the GPU Technology Conference.

Since Rich is traveling around in some desert somewhere, Dan and Henry go it alone and discuss what happened at last week’s GTC’16 conference. They talk about the new Pascal (P1000) GPU, NVIDIA’s new server, and what happened at the concurrent OpenPOWER conference. A good time is had by all….”

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC104: A Preview of GTC 2016

In this podcast, the Radio Free HPC team congratulates Dan Olds on becoming a Partner at OrionX. Then we go on to preview the GPU Technology Conference coming up April 4-7 in Silicon Valley.

GTC is the largest and most important event of the year for GPU developers. Join us this year as we showcase the most vital work in the computing industry today – Artificial Intelligence and Deep Learning, Virtual Reality and Self Driving Cars. GTC attracts developers, researchers, and technologists from some of the top companies, universities, research firms and government agencies from around the world.

Download the MP3 * Subscribe on iTunes * RSS Feed