RFHPC172: A Look at the New Power9, Titan V, and Snapdragon 845

In this podcast, the Radio Free HPC team looks at all new fast devices that came to market this week for HPC.

IBM Power9 Servers are here! Launched this week, the AC922 POWER9 servers will form the basis of the world’s fastest “Coral” supercomputers coming to ORNL and LLNL. Built specifically for compute-intensive AI workloads, the new POWER9 systems are capable of improving the training times of deep learning frameworks by nearly 4x allowing enterprises to build more accurate AI applications, faster.

NVIDIA Titan V brings Volta to the Desktop. Not to be outdone, NVIDIA introduced their new high end TITAN V GPU for desktop PCs. Powered by the Volta architecture, TITAN V excels at computational processing for scientific simulation. Its 21.1 billion transistors deliver 110 teraflops of raw horsepower, 9x that of its predecessor, and extreme energy efficiency.

Snapdragon 845 chip based on the ARM architecture brings AI inference to your smartphone. With A75 CPU cores clocking as high as 2.8 GHz, the new architecture also enables it to maintain the same level of performance with a 30% power reduction.

After that, we do our Catch of the Week:

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC171: Controversy in the Irish Supercomputing List

In this podcast, the Radio Free HPC team looks at a controversy stirred up by the recent Irish Supercomputing List.

The 9th Irish Supercomputer List was released this week. For the first time, Ireland has four computers ranked on the Top500 and Ireland is now ranked number one globally in terms of number of TOP500 supercomputers per capita.

Controversy: Since the publication of the List, a third party called the Irish Centre for High-End Computing (ICHEC) has expressed concerns that the press release issued by the Irish Supercomputing List is misleading. You can read their opinion here.

After that, we do our Catch of the Week:

  • Henry (with a certain measure of glee) points us to the story of the Root password security hole in Mac OS High Sierra. Why should Windows have all the fun?
  • Shahin reports that Amazon continues to add things like bare metal and other improvements to AWS. This week they introduced a Deep Learning camera for developers that comes preloaded with a Deep Lens environment for object recognition and machine learning. While this is a step in the right direction, Shahin thinks that there will be insufficient data going forward to train all the AI systems we want.

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC170: Wrapping Up the SC17 Student Cluster Competition

In this podcast, the Radio Free HPC team goes over the results of the SC17 Student Cluster Competition. This year, Nanyang Technological University from Singapore took the Top Prize.

Team Nanyang took the top spot at the SC17 Student Cluster Competition

“The Student Cluster Competition was developed in 2007 to immerse undergraduate and high school students in high performance computing. Student teams design and build small clusters, with hardware and software vendor partners, learn designated scientific applications, apply optimization techniques for their chosen architectures, and compete in a non-stop, 48-hour challenge, at the SC conference, to complete a real-world scientific workload, while impressing conference attendees and interview judges with their HPC knowledge.”

After that, we do our Catch of the Week:

In this video, Dan Olds from OrionX interviews Team Nanyang, this year’s champions of the SC17 Student Cluster Competition.

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC169: Recapping SC17 in Denver

In this podcast, the Radio Free HPC team describes the highlights of SC17 in Denver. Highlights includes the debut of ARM hardware for HPC, a Rasberry Pie Cluster, and pervasive liquid cooling.

Other highlights include:

  • Posit Research was there at the Innovation Showcase with an all-new company working to proliferate Unum based computing invented by John Gustafson from A*STAR in Singapore.
  • New high performance interconnects are proliferating with PCIe technologies, Gen-Z, etc.
  • The Student Cluster LINPACK record was smashed this year with 51.8 Teraflops. Nanyang University achieved this record number with the new NVIDIA Tesla V100 GPUs.

Shahin noted that there was a good number of exhibits centering around quantum computing.

After that, we do our catch of the week:

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC168: Running Down the Latest TOP500 List

From left, Henry Newman, Dan Olds, Shahin Khan, and Rich Brueckner are the Radio Free HPC team

In this podcast, the Radio Free HPC team reviews the latest TOP500 list in front of a live audience in Denver at SC17.

The fiftieth TOP500 list of the fastest supercomputers in the world has China overtaking the US in the total number of ranked systems by a margin of 202 to 144. It is the largest number of supercomputers China has ever claimed on the TOP500 ranking, with the US presence shrinking to its lowest level since the list’s inception 25 years ago.

Just six months ago, the US led with 169 systems, with China coming in at 160. Despite the reversal of fortunes, the 144 systems claimed by the US gives them a solid second place finish, with Japan in third place with 35, followed by Germany with 20, France with 18, and the UK with 15.

China has also overtaken the US in aggregate performance as well. The Asian superpower now claims 35.3 percent of the TOP500 flops, with the US in second place with 29.8 percent.

The top 10 systems remain largely unchanged since the June 2017 list, with a couple of notable exceptions.

Sunway TaihuLight, a system developed by China’s National Research Center of Parallel Computer Engineering & Technology (NRCPC), and installed at the National Supercomputing Center in Wuxi, maintains its number one ranking for the fourth time, with a High Performance Linpack (HPL) mark of 93.01 petaflops.

Tianhe-2 (Milky Way-2), a system developed by China’s National University of Defense Technology (NUDT) and deployed at the National Supercomputer Center in Guangzho, China, is still the number two system at 33.86 petaflops.

Piz Daint, a Cray XC50 system installed at the Swiss National Supercomputing Centre (CSCS) in Lugano, Switzerland, maintains its number three position with 19.59 petaflops, reaffirming its status as the most powerful supercomputer in Europe. Piz Daint was upgraded last year with NVIDIA Tesla P100 GPUs, which more than doubled its HPL performance of 9.77 petaflops.

The new number four system is the upgraded Gyoukou supercomputer, a ZettaScaler-2.2 system deployed at Japan’s Agency for Marine-Earth Science and Technology, which was the home of the Earth Simulator. Gyoukou was able to achieve an HPL result of 19.14 petaflops. using PEZY-SC2 accelerators, along with conventional Intel Xeon processors. The system’s 19,860,000 cores represent the highest level of concurrency ever recorded on the TOP500 rankings of supercomputers.

Titan, a five-year-old Cray XK7 system installed at the Department of Energy’s (DOE) Oak Ridge National Laboratory, and still the largest system in the US, slips down to number five. Its 17.59 petaflops are mainly the result of its NVIDIA K20x GPU accelerators.

Sequoia, an IBM BlueGene/Q system installed at DOE’s Lawrence Livermore National Laboratory, is the number six system on the list with a mark of 17.17 petaflops. It was deployed in 2011.

The new number seven system is Trinity, a Cray XC40 supercomputer operated by Los Alamos National Laboratory and Sandia National Laboratories. It was recently upgraded with Intel “Knights Landing” Xeon Phi processors, which propelled it from 8.10 petaflops six months ago to its current high-water mark of 14.14 petaflops.

Cori, a Cray XC40 supercomputer, installed at the National Energy Research Scientific Computing Center (NERSC), is now the eighth fastest supercomputer in the world. Its 1,630 Intel Xeon “Haswell” processor nodes and 9,300 Intel Xeon Phi 7250 nodes yielded an HPL result of 14.01 petaflops.

At 13.55 petaflops, Oakforest-PACS, a Fujitsu PRIMERGY CX1640 M1 installed at Joint Center for Advanced High Performance Computing in Japan, is the number nine system. It too is powered by Intel “Knights Landing” Xeon Phi processors.

Fujitsu’s K computer installed at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe, Japan, is now the number 10 system at 10.51 petaflops. Its performance is derived from its 88 thousand SPARC64 processor cores linked by Fujitsu’s Tofu interconnect. Despite its tenth-place showing on HPL, the K Computer is the top-ranked system on the High-Performance Conjugate Gradients (HPCG) benchmark.

For the first time, each of the top 10 supercomputers delivered more than 10 petaflops on HPL. There are also 181 systems with performance greater than a petaflop – up from 138 on the June 2017 list. Taking a broader look, the combined performance of all 500 systems has grown to 845 petaflops, compared to 749 petaflops six months ago and 672 petaflops one year ago. Even though aggregate performance grew by nearly 100 petaflops, the relative increase is well below the list’s long-term historical trend.

A further reflection of this slowdown is the list turnover. The entry point in the latest rankings moved up to 548 teraflops, compared to 432 teraflops in June. The 548-teraflop system was in position 370 in the previous TOP500 list. The turnover is in line with what has been observed over the last four years, but is much lower than previous levels.

A total of 101 systems employ accelerator/coprocessor technology, compared to 91 six months ago. 86 of these use NVIDIA GPUs, 12 systems make use Intel Xeon Phi coprocessor technology, and 5 are using PEZY Computing accelerators. Two systems use a combination of NVIDIA GPU and Intel Xeon Phi coprocessors. An additional 17 systems now use Xeon Phi chips as the main processing unit.

Green500 Highlights

Turning to the new Green500 rankings, the top three positions are taken by newly installed systems in Japan, all of which are based on the ZettaScaler-2.2 architecture and the PEZY-SC2 accelerator. The SC2 is a second-generation 2048-core chip that provides a peak performance of 8.192 teraflops in single-precision.

The most efficient of these ZettaScaler supercomputers is the Shoubu system B installed at RIKEN’s Advanced Center for Computing and Communication. It achieved a power efficiency of 17.0 gigaflops/watt.

The number two Green500 system is the Suiren2 cluster at the High Energy Accelerator Research Organization/KEK, which managed to reach 16.8 gigaflops/watt.

The number three Green500 slot was captured by the PEZY Computing’s own Sakura system. It achieved 14.2 gigaflops/watt. All of these top three systems are positioned in the bottom half of the TOP500 rankings: Shoubu system B at position 258, Suiren2 at 306, and Sakura at 275.

The fourth greenest supercomputer is a DGX SaturnV Volta system, which is installed at NVIDIA headquarters in San Jose, California. It achieved 15.1 gigaflops/watt, and comes in at number 149 on the TOP500 list. The number five system is Gyoukou, yet another ZettaScaler-2.2 machine. It achieved an efficiency of 14.2 gigaflops/watt and it currently ranks as the fourth most powerful supercomputer in the world.

Vendor trends

A total of 471 systems, representing 94.2 percent of the total, are now using Intel processors, which is slightly up from 92.8 percent six months ago. The share of IBM Power processors is at 14 systems, down from 21 systems in June.

The number of systems using Gigabit Ethernet is unchanged at 228 systems, in large part thanks to 204 systems now using 10G Ethernet. InfiniBand technology is found in 163 systems, down from 178 systems in the previous list, and remains the second most-used system interconnect technology in the list. Intel Omni-Path technology is now in 35 systems, down from 38 six month ago.

HPE has the lead in the number of installed supercomputers at 123, which represents nearly a quarter of all TOP500 systems. This includes several systems originally installed by SGI, which is now owned by HPE. HPE accounted for 144 systems six months ago.

Lenovo follows HPE with 81 systems down from 88 systems on the June list. Inspur rose further in the ranks and has now 56 systems, up from only 20 six month ago. Cray now has 53 systems, down from 57 systems six month ago. Sugon features 51 systems in the list, up from 44 in June. IBM follows with only 19 systems remaining under their label. These are mostly BlueGene/Q supercomputers, reflecting an aging install base. The average age of IBM systems on the list is now five years.

Cray continues to be the clear performance leader, claiming 19.5 percent of the list’s aggregate performance. HPE is second with 15.2 percent of the TOP500 flops. Thanks to the number one Sunway TaihuLight system, NRCPC retains the third spot with 11.1 percent of the total performance. Lenovo is fourth with 9.1 percent of performance, followed by Inspur at 6.3 percent, IBM at 6.1 percent and Sugon at 5.2 percent. All top vendors, with the exception of Inspur and Sugon, lost performance share compared to six months ago.

HPCG Results

The TOP500 list is now incorporating the High-Performance Conjugate Gradient (HPCG) benchmark results into the list to provide a more balanced look at system performance. The benchmark incorporates calculations in sparse matrix multiplication, global collectives, and vector updates, which more closely represents the mix of computational and data access patterns used in many supercomputing codes.

As previously mentioned, the fastest system using the HPCG benchmark remains Fujitsu’s K computer, which is ranked number 10 in the overall TOP500 rankings. It achieved 602.7 teraflops on HPCG, followed closely by Tianhe-2 with a score of 580.0 teraflops. The upgraded Trinity supercomputer comes in at number three at 546.1 teraflops, followed by Piz Daint at number four with 486.4 teraflops, and Sunway TaihuLight at number five at 480.8 teraflops.

The International Space Station computer, built by HPE, is now listed in the HPCG results, making it the “highest” computer on the list.

About the TOP500 List

The first version of what became today’s TOP500 list started as an exercise for a small conference in Germany in June 1993. Out of curiosity, the authors decided to revisit the list in November 1993 to see how things had changed. About that time they realized they might be onto something and decided to continue compiling the list, which is now a much-anticipated, much-watched and much-debated twice-yearly event.

See our complete coverage of SC17

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC167: A Preview of the SC17 Student Cluster Competition

In this podcast, the Radio Free HPC team previews the SC17 Student Cluster Competition.

“In this real-time, non-stop, 48-hour challenge, teams of undergraduate and/or high school students assemble a small cluster on the exhibit floor and race to complete a real-world workload across a series of applications and impress HPC industry judges.” 

Dan’s got the inside scoop on all the new wrinkles of the competition this year, where 16 international student teams will go head-to-head. There’s an all high school squad from Indiana, returning champs from China, and an all-new cloud computing component utilizing Cycle Computing on Azure.

After that, we do our Catch of the Week:

  • Rich notes that a couple of SC17 events will be livestreamed for the folks back home:
    • SC17 Plenary on Smart Cities, Nov 13. The Smart Cities initiative looks to improve the quality of life for residents using HPC, urban informatics, and other technologies to improve the efficiency of services. The event starts at 5:30pm on Monday, Nov. 13 in the Colorado Convention Center.
    • D-Wave Quantum Computing Seminar & Livestream, Nov. 13. “We will discuss quantum computing, the D-Wave 2000Q system and software, the growing software ecosystem, an overview of some user projects, and how quantum computing can be applied to problems in optimization, machine learning, cyber security, and sampling.” Register for the in-person event at the Denver Athletic Club. You can also join them for the Livestream.
  • Shahin shares his notion that the rise of blockchain demonstrates that “distributed trust” is computationally expensive and therefore offers an emerging market for High Performance Computing. We are already seeing how computationally expensive AI processing is already boosting the HPC market.
  • Dan reminds our listeners attending SC17 to join us at the Guard and Grace Steakhouse in Denver for a live Podcast recording starting at 6:00 pm on Saturday, Nov. 11.

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC166: Speeding up Workloads with Redline Performance Solutions

Carolyn Pasti and Don Avart from RedLine Performance Solutions

In this podcast, the Radio Free HPC teams discusses performance optimization with Carolyn Pasti and Don Avart from Red Line Performance Solutions.

RedLine Performance Solutions is a world-class provider of high-performance computing solutions. Our promise: to ensure objectively engineered top quality solutions at every phase of the HPC life cycle, minimizing labor, time, and costs. Our proprietary RedLine Performance Methodology (RPM) – developed over two decades of working with HPC systems and applications and updated regularly with lessons from each new engagement – delivers unique benefits that consistently maximize customer success.

RedLine is partnering with Radio Free HPC on Project Cyclops, an effort to build the world’s fastest single node on the HPCG benchmark. Listen in as Don and Carolyn share their methodology for workload performance optimization and what it takes to make clusters really perform up their potential in the real world.

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC165: Countdown to SC17 in Denver

In this podcast, the Radio Free HPC team previews the ancillary events leading up to SC17, which takes place Nov. 12-17 in Denver.

Here is the countdown to SC17:

  • HP-CAST, Nov. 10-11. The High Performance Consortium for Advanced Scientific and Technical (HP-CAST) computing users group works to increase the capabilities of Hewlett Packard Enterprise solutions for large-scale, scientific and technical computing. HP-CAST provides guidance to Hewlett Packard Enterprise on the essential development and support issues for such systems. HP-CAST meetings typically include corporate briefings and presentations by HPE executives and technical staff (under NDA), and discussions of customer issues related to high-performance technical computing.

 

  • Intel HPC Developer Conference, Nov 11-12.  This conference brings together top industry leaders, luminaries, and tech leaders to discuss Parallel Programming, High Productivity Languages, Artificial Intelligence, Systems, Enterprise, Visualization Development, and more. This is a premier training event for learning industry best practices, tools, and techniques, to maximize software efficiency on Intel architecture. Learn more.

 

  • Radio Free HPC LIVE, Nov 11. The Radio Free HPC podcast is a joint production by industry pundits focused on High Performance Computing. While there are other HPC podcasts out there, we wanted to create something that was both informative and fun to listen to. Join us for our annual live podcast recording this year at SC17. The event starts at 6:00 pm on Nov. 12 at the Guard and Grace Steak House in Denver.

  • SC17, Nov 12-17. The international conference on high performance computing kicks off Sunday with Tutorial Sessions.

 

  • StartupHPC Meetup at SC17, Nov 12. StartupHPC is a grass roots community for STEM- and HPC-inspired entrepreneurs, corporations, venture capital firms, academia, government agencies, and support organizations. The StartupHPC Meetup takes place at the Rock Bottom Brewery at 5-9 pm on Sunday, Nov.12. Please register so we know you are coming but feel free to just show up. This is a free no-host gathering and the idea is to be relaxed and network.

 

  • Dell HPC Community Meeting, Nov. 13. The Dell EMC HPC Community is a worldwide technical forum that fosters the exchange of ideas among researchers, computer scientists, technologists, and engineers and promotes the advancement of innovative, powerful HPC solutions.

 

  • DDN User Group at SC17, Nov 13. The DDN User Group brings the HPC community together with DDN executive and technical leadership to share how leading global organizations are leveraging innovative new approaches, best practices, and technologies to raise the bar on HPC discoveries. This year, we are hosting the user group at the Embassy Suites, which is a three minute-walk across the street from the Colorado Convention Center.

 

  • SC17 Plenary on Smart Cities, Nov 13. The Smart Cities initiative looks to improve the quality of life for residents using HPC, urban informatics, and other technologies to improve the efficiency of services. The event starts at 5:30pm on Monday, Nov. 13 in the Colorado Convention Center.

 

  • Beowulf Bash, Nov 13. The annual event takes place at the Lucky Strike bowling alley on Monday, Nov. 13 from 9pm to 12 midnight. “Got an urge to try something new? Lucky Strike Denver has you covered. With a tantalizing menu of chef driven cuisine, hand-crafted cocktails, and live local DJs, Lucky Strike Denver breathes new life into the typical bowling experience with chic modern decor and a one-of-a-kind atmosphere.”

 

  • Hyperion Research SC17 Briefing, 7:30am – 9:00am, Nov. 15. When does moving your organization’s HPC workloads and data to the cloud make sense? Hyperion Research will present new, real-world findings on the use of on premise data centers versus public, private and hybrid clouds for HPC workloads–including drivers, barriers and near-term plans. Spectra Logic will discuss current trends affecting data storage, especially in cloud environments.

After that, we do our Catch of the Week:

  • Henry notes that someone has already picked up his idea about using Blockchain to fight Fake News. In fact, they are doing an Initial Coin Offering (ICO) for their PUBLIQ cryptocurrency to fund the project.
  • Rich notes that Intel is now shipping their Optane SSDs, which the company claims outperform NAND devices by 4x. Henry says the things are pretty expensive though.
  • Dan gave us an update Project Cyclops, which is quickly evolving into a liquid-cooled powerhouse that you can check out for yourself at the Supermicro booth #1611 at SC17.

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC164: Is the Cloud Killing Tape?

In this podcast, the Radio Free HPC team asks the question: Is the Cloud Killing the Tape Market? The issue came up this week that Spectra Logic is laying off 9 percent of its workforce.

“There may be a silver lining for Spectra Logic, with a legacy of tape storage. The technology has had its ups and downs over the years, and things look positive for it once again as cloud providers have adopted it as a cost-effective way to handle the growing amount of data being generated, TechTarget, a marketing firm, reported.”

After that, we do our Catch of the Week:

  • Dan announced that the Project Cyclops hit a major milestone this week with a Hello World after he successfully installed CentOS.
  • Shahin announced that the StartupHPC Meetup will take place at the Rock Bottom Brewery at 5-9 pm on Sunday, Nov.10.  Please register so we know you are coming but feel free to just show up.  This is a free no-host gathering and the idea is to be relaxed and network.
    • For this year, we’d like to do two things:
      • Gather informally and have time to network and deepen our friendships.
      • Offer private meetings so startups and those who are thinking about entrepreneurship can have the time and privacy to discuss what they need in order to get from point A to point B.
  • Dan reminds our listeners attending SC17 to join us at the Guard and Grace Steakhouse in Denver for a live Podcast recording starting at 6:00 pm on Saturday, Nov. 9.
  • Rich pointed us to news about how supercomputer modeling aided the LIGO discovery of gravity waves from a neutron star collision.

Download the MP3 * Subscribe on iTunes * RSS Feed

RFHPC163: Announcing Project Cyclops – One Node to Rule them All

In this podcast, Dan Olds from OrionX describes Project Cyclops – a benchmarking quest to build the world’s fastest single node.

Codenamed “Cyclops”, the single-node supercomputer demonstrates the computational power that individual scientists, engineers, artificial intelligence practitioners, and data scientists can deploy in their offices. Cyclops has achieved results of (score) for the HPCC benchmark, a mark of (score) for the HPCG benchmark, and a score of (score) for the Graph500 benchmark. These results set new single node high water marks for the (list benchmark we achieved records on).

We’ve done more than 150 podcasts talking about topics in high performance computing,” said Dan Olds, partner at OrionX.net Research. “It was time we actually put something together and take a run at some of these benchmarks.” Rich Brueckner commented, “We couldn’t have done this without our sponsors, who include Intel, Kingston, Seagate, and Ebullient Cooling Solutions. They’ve been incredibly helpful.”

Project Cyclops is building a system based on a chassis provided by Supermicro and is fueled by a pair of Intel Xeon E5 v4 processors, 128 GB of Kingston DDR4 memory, and M2 SSD drives from Seagate. An innovative liquid cooling system from CoolIT Systems will chill the processors and GPU accelerators. RedLine Performance Solutions will provide system testing and will help optimize the system for benchmarking and workload runs.

After that, we do our Catch of the Week:

  • Dan does his new Weekly Rant feature, asking why people don’t put their contact info in their email signature.
  • Shahin notes that the new iPhone X reportedly has face recognition capabilities, though their demo didn’t work at the launch event. Henry says this is just another step and not an end of the quest for better security.
  • Henry says the recent Equifax debacle marks the sixth time that his info has been hacked.
  • Rich points us to this story on HPC and Kubernetes.

Download the MP3 * Subscribe on iTunes * RSS Feed