The HPC Cloud-Centric Guide to SC10

By Nicole Hemsoth

November 12, 2010

On Sunday we’ll be packing up and heading off to New Orleans for SC10 for a full week of live coverage, including details from select sessions and presentations, thoughts from some of the leaders in the HPC cloud space and video interviews with a number of interesting attendees from the academic and vendor communities.

Navigating the conference program requires pulling out all the stops in the “artful scheduling” department, but we’ve managed to take a look at the program and narrow in on some select sessions and other events that might be right up your alley if you’re regular reader and keep up to date with the news and ecosystem trends in the high-performance computing cloud space.

This year we were thrilled to learn that there would be a sub-set of special sections during the conference devoted to HPC in the cloud as a movement via the Birds of a Feather session series.

While there are a few of these that we want to call your attention to, there are also a few other highlights during the regular program that are worth mentioning.

This brief guide will give you a view of what to look for this year in New Orleans and might require you to make use of some virtual whiteout as you adjust your schedule accordingly.

If nothing else, the conference’s side focus on the growing role of cloud computing in HPC, in addition to other recent events, including ISC Cloud this year, shows that there is some momentum building—and that we could be witnessing a more pronounced move “cloudward” in the coming year.

For those who cannot attend SC10, we hope you’ll stay tuned to the site for an in-depth, on-spot view into what the HPC community is seeing in the clouds.

HPC Cloud Birds of Feather Sessions at SC10

Birds of a Feather (BOF) sessions have been an SC tradition and allow the conference organizers to designate a small handful of special topics to explore that are still in the HPC ecosystem, but lie just on the horizon for one reason or another.

BOFs will meet on Tuesday, Wednesday and Thursday at 12:15 p.m. and on Tuesday and Wednesday at 5:30 p.m. They will be open for all attendees, including exhibitors only badge types.

During this year’s event, the topics designated for BOFs will include HPC in Weather and Climate Modeling, Energy Efficiency in Supercomputing Center, Training and Education for High Performance Computing and, last but certainly not least, HPC in a Cloud Environment.

Low Latency, High Throughput, RDMA and the Cloud in Between

Whether in the context of traditional high-performance computing or HPC in the cloud, the issues of latency and throughput remain persistent concerns. Added to that list are matters of RDMA and offloading, which also have an impact on application performance.

Gilad Shainer from Mellanox and Jeff Layton of Dell will spend an hour on Tuesday talking about how, despite the promise of low latency Ethernet from any number of quarters, it is not the solution it needs to be to fulfill many requirements for HPC users. The presenters state that, “for enabling true low-latency Ethernet, new proposals now raise the idea of using different transport protocols on top of Ethernet, as such transports were proven to provide low latency and other important networking elements.”

While InfiniBand has already proven itself to be a viable solution for users of high-performance computing, the focus on Ethernet, especially for those who have already invested a great deal in existing networking infrastructures, is an important one.

One can expect this session on the latest progress points for RoCE will draw some spectators, as will the embedded promise of details about the InfiniBand FDR and EDR roadmap, which will be highlighted by use cases. 

The presenters also hope to hit on another topic in the wake of this more overarching discussion on low-latency Ethernet and InfiniBand by broaching, “the fourth mode of science—Science Discovery, in which HPC clouds will be required to handle the data flood generated by computation and science simulations.”

On a side note, Shainer discussed this as well as other matters in a recent interview with HPC in the Cloud that might serve as great background material for his presentation.

If this has piqued your interest, this discussion takes place on Tuesday from 12:15-1:15 p.m. in Room 397.

It sounds like a lot to pack into one short hour, but for those who are looking for a more thorough overview of RoCE in detail, you’re going to be forced to make a tough choice between Shainer and Layton’s presentation and Paul Grun’s related session entitled “RoCE: Next Generation RDMA Network” which will take place at the same bat time and channel, except in Room 280-281.

Decisions, decisions….When you get right down to it, this seems to be the toughest part of SC10. Unless you’re a journalist trying to take it all it and weave it together in no time. But enough about my struggles.
 
Building the World’s Largest HPC Cloud

William Lu from Platform Computing will be leading one of the juicer items on the HPC cloud plate during SC10 with his presentation on CERN’s cluster dedicated to crunching the massive datasets from the Large Hadron Collider (LHC) and its status as the world’s largest HPC cloud. While it is not what you’d call vendor neutral since Platform will be discussing how it built the research center’s platform, it will nonetheless be an exciting journey into the inner working of CERN’s ability to manage vastly complex virtual and physical workloads and requirements.

One aspect of this presentation that will be most interesting is that discussion participants who are facing challenges similar to those at CERN (those with massive datasets and intense throughput requirements) can consider the cloud. It will be interesting to see who attends and what experiences they share about their current infrastructure and budget challenges and what they’re considering in order to minimize these problems—and if cloud is really on the drawing board.

The CERN facility is one of a handful of large-scale HPC clouds that are wringing out some great news for the possibility of HPC in the cloud adoption. They have had to overcome some significant challenges and hopefully Platform’s Lu will talk in some detail about what these challenges were and how (and if) they can be mitigated enough to be acceptable solutions for other institutions.

William Lu’s presentation will take place on Wednesday, November 17th from 1:15 to 2:15 p.m. in Rooms 280-281.

For those interested, we have a recent article on CERN’s HPC cloud here for some background.
 
Enabling High Performance Cloud Computing Environments

On Thursday, November 18th from 1:15 to 2:15 Jurrie Van Den Breekel of Spirent will deliver a session on some overarching topics relevant to cloud environments for HPC users, including topics related to mult-core processors, virtualization and the current state of Ethernet—and how to build environments that maximize efficiency and performance capabilities.

The discussion will revolve around some case studies from Spirent, including their recent experiences with the European Advanced Networking Test Center (EANTC), which will reveal how actual cloud implementation on the large scale, both within the a private and bursting capacity model, works—and what challenges are present. The session also promises to provide “a close examination of how implementing a cloud approach within a datacenter affects the firewall, data center bridging, virtualization and WAN optimization.”

Again, this one sounds like it’s aiming to pack a lot into one hour but what it is providing, even if it is from a vendor’s view (why can’t there be more research team leaders presenting on their experiences with cloud on the large scale at more events?) but the pulling in implementation details to frame the discussion will make this session worthwhile.

More on the Cloud Front at SC10

Aside from the Birds of a Feather sessions, some of which touch on issues specific to HPC clouds, there are a number of presentations that are scattered around the regular conference program that will likely be of interest to some of you.

Below is a list of some items on the agenda worth paying attention to:

Disruptive Technologies: Cloud Era, Ltd.

A few weeks ago we posted a piece on an emerging company, which for now is a one-man show, that seeks to deliver the first cloud portal to the “R” language and far beyond.

While I could go on at length about this ambitious project (and what it could mean for the future of scientific and research collaboration via cloud computing) the best way to get a sense for the scope is to check out this detailed SC10 Disruptive Technologies preview article.

Karim Chine will be presenting on “Elastic-R: A Virtual Collaborative Environment for Scientific Computing and Data Analysis in the Cloud” via a tutorial/overview on Monday morning from 8:30 a.m.-12:00 p.m. in Room 389. Cloud Era, Ltd., which is the sole effort oChine, will also be featured during the Disruptive Technologies series during regular exhibition hours.

3rd Workshop on Many-Task Computing on Grids and Supercomputers (MTAGS)

Chances are, if you’re going to be at the conference on Monday and you plan on attending this workshop, you’ve already planned ahead by blocking off your entire day. From 9:00 to 5:30 in room 271, Ioan Raicu, Ian Foster and Yong Zhao will be leading a workshop to “provide the scientific community a dedicated forum for presenting new research, development and deployment efforts of large-scale, many-task computing (MTC) applications on large-scale cluster, grids, supercomputer and cloud computing infrastructure.”

This workshop has an ambitious range of topics that will explored, including what the organizers describe as “challenges that can hamper efficiency and utilization in running applications on large-scale systems…scalability, data management, I/O management, reliability at scale and application scalability” among others.

It would be interesting to take a quick census during this workshop and find out just how many of the world’s largest supercomputing center administrative folks are in attendance—and where they come from.

So, if you’ve got a national lab or university system to manage and you’re free on Monday…

Web Portals for Scientific Computing

On Monday from 1:30-5:00 p.m. in Room 287, Jano van Hemert and Jos Koetsier from the University of Edinburgh will hold court to discuss the growing trend of the use of web portals as interfaces to scientific research resources.

Far from being a general overview of the types of portals or some of their specific uses, this session will present a thorough description of actually making use of these resources. The presenters note that they will be discussing challenges to using these portals and via the tutorial they will “provide a basic hands-on tutorial on building portals and a more advanced hands-on session to deal with application and resource-specific requirements.”

Virtualization for HPC

Josh Simons, Stephan Scott, DK Panda, Bill Bryce and Scott Clark will be presenting this session in Rooms 385-386 on Tuesday, November 16 from 1:30-3:00 p.m. They plan to touch on reasons why, in the enterprise cloud adoption has remained significantly higher and how interest in virtualization for HPC has been growing.

The speakers will contend that “beyond cloud computing, virtualization offers additional potential benefits for HPC, among them support for multi-tenant environments, checkpoint-resistance capabilities, dynamic provisioning, low-overhead job migration, proactive fault tolerance, multi-OS support for heterogeneous HPC facilities, and system and application resilience.”

This panel session promises to be a contentious one since this argument tends to get heated words on both sides, so expect an interesting, debate if you have a couple of hours to set aside on Tuesday.

The Past, Present and Future of HPC

Jack Dongarra (University of Tennessee), Ian Foster (Argonne) and Dorian Arnold (University of New Mexico will be presenting this epic session for early birds on Sunday, November 14th from 3:30-5:00 p.m. in Rooms 290-291. While the details about what they will cover on the future spectrum can easily be guessed, one item that one can imagine will at least be mentioned is the impact on cloud computing on current HPC paradigms.

Even if some of the material is “old hat” to some, there’s nothing like kicking off a conference series by putting the ecosystem in historical perspective and having some leading voices in the arena voice their views on what will shape the coming new era of high-performance computing. Whether its multi-core, clouds, GPGPUs, innovations in networking, progressive application developments—this is a “must attend” for anyone arriving before the real fun begins on Monday.

Scattered Clouds at SC10

Aside from the sessions themselves, there are a number of other events going on, including poster reception events that are relating some information and ideas about HPC in the cloud. The following are a few that you might want to wander into.

A research poster reception entitled, HPCaaS: High Performance Computing as a Service, with Marcel Kunze and Viktor Mauch, both from the Karlsruhe Institute of Technology,will be held Tuesday in the main lobby from 5:15 to 7:00 p.m.  The focus of this poster reception will examine issues related to the problem that there is currently “no clear concept of how to provide HPC as a Service supporting InfiniBand with a dynamic configuration in a virtualized cloud computing environment” and how there are some different strategies that have been set forth to address this.

Another such reception that is right in our range, “High Performance Computing in the Cloud: Why Not?!” will be delivered by Tobias Lindinger, Nils Gentschen Felde and Dieter Franzmuller, a team from Ludwig-Maximillans University in Munich will provide “an analysis of the suitability of clouds for high-performance computing, thereby providing a qualitative evaluation of large-scale, virtualized infrastructures in contrast to specialized clusters or supercomputers.” In addition to offering a viability overview, the team will address issues of performance degradation and virtualization hurdles and will present some recommendations based on case studies.

Other notable receptions include:

Provisioning Virtual Clusters in the Cloud Using Wrangler – a research poster reception held on Tuesday from 5:15-7:00 p.m. in the main lobby, delivered by Gideon Juve and Ewa Deelman from the Information Sciences Institute.

Cumulus: Open Source Storage Cloud for Science – another research poster reception on Tuesday from 5:15 to 7:00 p.m. in the main lobby from a team comprised of Argonne National Laboratory’s John Bresnahan and Kate Keahey as well as Tim Freeman and David LaBissoniere from the University of Chicago.

Even More Events of Note…

TeraGrid Technology Auditing Services Tools — a session with Thomas R. Furlani, Matthew D. Jones, Steven M. Gallo on Tuesday, November 16, from 12:15-1:15 p.m.

Network Automation, the Power of Choice: Making the Cloud Work for You – an exhibitor forum led by Stephen Garrison from Force10 Networks, which will take place Thursday from 11:00-11:30 a.m. in Rooms 280-281.

Cloud Networking: Evolution of Data Center Traffic Patterns – another exhibitor forum with a networking focus, this time led by Anshul Sadana from Arista Networks. Taking place Thursday from 11:30 a.m. to noon on Thursday in the same room as Force10’s presentation, 280-281.

Enough Clouds for You?

Okay, so if you weren’t convinced there was a cloud movement in HPC up until this year, it has to be a little harder to make the same argument now. There are even more sessions, particularly on the networking end, that tie in nicely with clouds for HPC as well but I’ll leave this list as it is for now. After all, if you remember way back on page one of this guide I mentioned it was going to be “brief”…Well, so much for that.

If you’re not able to attend, please make sure you stay tuned to the site for special coverage and if you haven’t already, subscribe to our newsletter to get up-to-date coverage on this and other movements in HPC and cloud. This week we’ll be upping the normal dose with two special edition newsletters (instead of one)—the first is set to hit your inbox Wednesday morning, the second early Friday morning with some final wrap-up coverage and news items the next week.

If you are at the show, please do come say hello! Don’t know about any of you, but man–I’m excited for this one!
 

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

HPC Financial Update (Feb. 2017)

February 22, 2017

In this recurring feature, we’ll provide you with financial highlights from companies in the HPC industry. Check back in regularly for an updated list with the most pertinent fiscal information. Read more…

By Thomas Ayres

Rethinking HPC Platforms for ‘Second Gen’ Applications

February 22, 2017

Just what constitutes HPC and how best to support it is a keen topic currently. Read more…

By John Russell

HPC Technique Propels Deep Learning at Scale

February 21, 2017

Researchers from Baidu’s Silicon Valley AI Lab (SVAIL) have adapted a well-known HPC communication technique to boost the speed and scale of their neural network training and now they are sharing their implementation with the larger deep learning community. Read more…

By Tiffany Trader

IDC: Will the Real Exascale Race Please Stand Up?

February 21, 2017

So the exascale race is on. And lots of organizations are in the pack. Government announcements from the US, China, India, Japan, and the EU indicate that they are working hard to make it happen – some sooner, some later. Read more…

By Bob Sorensen, IDC

HPE Extreme Performance Solutions

O&G Companies Create Value with High Performance Remote Visualization

Today’s oil and gas (O&G) companies are striving to process datasets that have become not only tremendously large, but extremely complex. And the larger that data becomes, the harder it is to move and analyze it – particularly with a workforce that could be distributed between drilling sites, offshore rigs, and remote offices. Read more…

ExxonMobil, NCSA, Cray Scale Reservoir Simulation to 700,000+ Processors

February 17, 2017

In a scaling breakthrough for oil and gas discovery, ExxonMobil geoscientists report they have harnessed the power of 717,000 processors – the equivalent of 22,000 32-processor computers – to run complex oil and gas reservoir simulation models. Read more…

By Doug Black

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Drug Developers Use Google Cloud HPC in the Fight Against ALS

February 16, 2017

Within the haystack of a lethal disease such as ALS (amyotrophic lateral sclerosis / Lou Gehrig’s Disease) there exists, somewhere, the needle that will pierce this therapy-resistant affliction. Read more…

By Doug Black

HPC Technique Propels Deep Learning at Scale

February 21, 2017

Researchers from Baidu’s Silicon Valley AI Lab (SVAIL) have adapted a well-known HPC communication technique to boost the speed and scale of their neural network training and now they are sharing their implementation with the larger deep learning community. Read more…

By Tiffany Trader

IDC: Will the Real Exascale Race Please Stand Up?

February 21, 2017

So the exascale race is on. And lots of organizations are in the pack. Government announcements from the US, China, India, Japan, and the EU indicate that they are working hard to make it happen – some sooner, some later. Read more…

By Bob Sorensen, IDC

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Drug Developers Use Google Cloud HPC in the Fight Against ALS

February 16, 2017

Within the haystack of a lethal disease such as ALS (amyotrophic lateral sclerosis / Lou Gehrig’s Disease) there exists, somewhere, the needle that will pierce this therapy-resistant affliction. Read more…

By Doug Black

Azure Edges AWS in Linpack Benchmark Study

February 15, 2017

The “when will clouds be ready for HPC” question has ebbed and flowed for years. Read more…

By John Russell

Is Liquid Cooling Ready to Go Mainstream?

February 13, 2017

Lost in the frenzy of SC16 was a substantial rise in the number of vendors showing server oriented liquid cooling technologies. Three decades ago liquid cooling was pretty much the exclusive realm of the Cray-2 and IBM mainframe class products. That’s changing. We are now seeing an emergence of x86 class server products with exotic plumbing technology ranging from Direct-to-Chip to servers and storage completely immersed in a dielectric fluid. Read more…

By Steve Campbell

Cray Posts Best-Ever Quarter, Visibility Still Limited

February 10, 2017

On its Wednesday earnings call, Cray announced the largest revenue quarter in the company’s history and the second-highest revenue year. Read more…

By Tiffany Trader

For IBM/OpenPOWER: Success in 2017 = (Volume) Sales

January 11, 2017

To a large degree IBM and the OpenPOWER Foundation have done what they said they would – assembling a substantial and growing ecosystem and bringing Power-based products to market, all in about three years. Read more…

By John Russell

US, China Vie for Supercomputing Supremacy

November 14, 2016

The 48th edition of the TOP500 list is fresh off the presses and while there is no new number one system, as previously teased by China, there are a number of notable entrants from the US and around the world and significant trends to report on. Read more…

By Tiffany Trader

Lighting up Aurora: Behind the Scenes at the Creation of the DOE’s Upcoming 200 Petaflops Supercomputer

December 1, 2016

In April 2015, U.S. Department of Energy Undersecretary Franklin Orr announced that Intel would be the prime contractor for Aurora: Read more…

By Jan Rowell

D-Wave SC16 Update: What’s Bo Ewald Saying These Days

November 18, 2016

Tucked in a back section of the SC16 exhibit hall, quantum computing pioneer D-Wave has been talking up its new 2000-qubit processor announced in September. Forget for a moment the criticism sometimes aimed at D-Wave. This small Canadian company has sold several machines including, for example, ones to Lockheed and NASA, and has worked with Google on mapping machine learning problems to quantum computing. In July Los Alamos National Laboratory took possession of a 1000-quibit D-Wave 2X system that LANL ordered a year ago around the time of SC15. Read more…

By John Russell

Enlisting Deep Learning in the War on Cancer

December 7, 2016

Sometime in Q2 2017 the first ‘results’ of the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C) will become publicly available according to Rick Stevens. He leads one of three JDACS4C pilot projects pressing deep learning (DL) into service in the War on Cancer. Read more…

By John Russell

IBM Wants to be “Red Hat” of Deep Learning

January 26, 2017

IBM today announced the addition of TensorFlow and Chainer deep learning frameworks to its PowerAI suite of deep learning tools, which already includes popular offerings such as Caffe, Theano, and Torch. Read more…

By John Russell

HPC Startup Advances Auto-Parallelization’s Promise

January 23, 2017

The shift from single core to multicore hardware has made finding parallelism in codes more important than ever, but that hasn’t made the task of parallel programming any easier. Read more…

By Tiffany Trader

CPU Benchmarking: Haswell Versus POWER8

June 2, 2015

With OpenPOWER activity ramping up and IBM’s prominent role in the upcoming DOE machines Summit and Sierra, it’s a good time to look at how the IBM POWER CPU stacks up against the x86 Xeon Haswell CPU from Intel. Read more…

By Tiffany Trader

Leading Solution Providers

Nvidia Sees Bright Future for AI Supercomputing

November 23, 2016

Graphics chipmaker Nvidia made a strong showing at SC16 in Salt Lake City last week. Read more…

By Tiffany Trader

BioTeam’s Berman Charts 2017 HPC Trends in Life Sciences

January 4, 2017

Twenty years ago high performance computing was nearly absent from life sciences. Today it’s used throughout life sciences and biomedical research. Genomics and the data deluge from modern lab instruments are the main drivers, but so is the longer-term desire to perform predictive simulation in support of Precision Medicine (PM). There’s even a specialized life sciences supercomputer, ‘Anton’ from D.E. Shaw Research, and the Pittsburgh Supercomputing Center is standing up its second Anton 2 and actively soliciting project proposals. There’s a lot going on. Read more…

By John Russell

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

IDG to Be Bought by Chinese Investors; IDC to Spin Out HPC Group

January 19, 2017

US-based publishing and investment firm International Data Group, Inc. (IDG) will be acquired by a pair of Chinese investors, China Oceanwide Holdings Group Co., Ltd. Read more…

By Tiffany Trader

Dell Knights Landing Machine Sets New STAC Records

November 2, 2016

The Securities Technology Analysis Center, commonly known as STAC, has released a new report characterizing the performance of the Knight Landing-based Dell PowerEdge C6320p server on the STAC-A2 benchmarking suite, widely used by the financial services industry to test and evaluate computing platforms. The Dell machine has set new records for both the baseline Greeks benchmark and the large Greeks benchmark. Read more…

By Tiffany Trader

What Knights Landing Is Not

June 18, 2016

As we get ready to launch the newest member of the Intel Xeon Phi family, code named Knights Landing, it is natural that there be some questions and potentially some confusion. Read more…

By James Reinders, Intel

KNUPATH Hermosa-based Commercial Boards Expected in Q1 2017

December 15, 2016

Last June tech start-up KnuEdge emerged from stealth mode to begin spreading the word about its new processor and fabric technology that’s been roughly a decade in the making. Read more…

By John Russell

Intel and Trump Announce $7B for Fab 42 Targeting 7nm

February 8, 2017

In what may be an attempt by President Trump to reset his turbulent relationship with the high tech industry, he and Intel CEO Brian Krzanich today announced plans to invest more than $7 billion to complete Fab 42. Read more…

By John Russell

  • arrow
  • Click Here for More Headlines
  • arrow
Share This