DANCES with Big Data: Progress Update from XSEDE Meeting

By John Russell

July 22, 2016

Moving large datasets around HPC networks such as those of XSEDE is often challenging. While Internet2, the most commonly used backbone, is fast at 100GBS, local traffic to campuses often slows to 10GBS. At this week’s XSEDE meeting the DANCES (Developing Applications with Networking Capabilities via End-to-End SDN) project leaders reported first successful testing of hardware components needed as well as results of vendor bakeoff for switches.

The NSF-funded DANCES[i] project is intended to investigate and develop the ability to add network bandwidth scheduling via software- defined networking (SDN) programmability to selected cyberinfrastructure services and applications intended to develop a software/hardware infrastructure that speeds and smoothes high-speed traffic. The recent test involved transfer of “Big Data” between two XSEDE sites, the Pittsburgh Supercomputing Center (PSC) and the National Institute for Computational Sciences (NICS) at the University of Tennessee.

Victor Hazlewood
Victor Hazlewood

“If you think about [high-performance computing] users and their end-to-end workflow, they need to do it by breaking their projects into pieces,” says Victor Hazlewood of NICS, principal author of the peer-reviewed article accompanying the presentation. “One of these pieces inevitably involves moving data into and out of where researchers do their computation.”

Today’s research wide-area networks (WAN) are more a free-for-all than their compute and storage counterparts, which are scheduled and managed, according to Hazlewood. By design, the protocol software that regulates the network data flow treats all data packets as equal. When the networks are not crowded it’s like an empty freeway. When they are crowded, though, the situation becomes much like a convoy of buses getting broken apart by other traffic. For large datasets traveling across the WAN, that can mean slow movement or even failures to move the data completely in a reasonable amount of time.

DANCES DIAGRAM

Project leaders presented a DANCES progress update and their paper[ii] at the XSEDE. In addition to reviewing the technology investigated, the paper details the evaluation of several vendors for use in the project. Corsa Technology, a Canadian technology start-up, was selected. “We were … interested in devices that supported bandwidth reservation,” said coauthor Kathy Benninger of PSC. “Part of that has to do with adoption of the newest OpenFlow versions by switch vendors.”

OpenFlow 1.3, the latest commonly deployed version of the standard network communications interface, has features that support such bandwidth management. But the problem, Benninger explains, is that “compliance” with a given version of OpenFlow doesn’t require a particular network switch to support every facet of that version’s specification. The researchers had to test a number of switch products, eventually choosing the Corsa DP6410 as most fully compatible with the DANCES requirements, successfully testing the combination in a number of data-transfer scenarios between PSC and NICS. Next, the group plans to test the full set of end-to-end system functions that will be needed to support real data transfers by XSEDE users.

Here’s an excerpt from the paper describing target device and application characteristics:

“The SDN capabilities would be provided by a network device that implements OpenFlow 1.3 including per-flow meters and per-port queuing required by DANCES. The bandwidth control and scheduling capability will be used to mitigate congestion – induced throughput problems on end site networks. The prototype environment setup has been done in two phases. The first phase was done by setting up and testing in a virtual environment and the second phase was done by creating a WAN test environment.

“The cyberinfrastructure applications chosen for investigation and potential performance enhancement by DANCES include (1) SCP and GridFTP data transfers and (2) wide area distributed file system data transfers initiated by, and integrated with, job scheduling and resource management. In addition, SLASH2, the wide area distributed file system targeted for DANCES integration, natively supports scheduled remote file replication. Coupling the SLASH2native file transfer scheduling with the network bandwidth scheduling capability of DANCES increases the efficiency of both network and file system resources.

“The project set out to deploy a prototype environment to extend the TORQUE/Moab scheduling and management currently in use at several XSEDE supercomputing sites to support networking as a schedulable resource. The wide area distributed file systems selected for bandwidth scheduling integration were the XSEDE-wide File System built on IBM’s General Parallel File System (GPFS) and SLASH2 developed at Pittsburgh Supercomputing Center. Since the award of DANCES, XSEDE has decided to no longer support the XSEDE-wide file system based on GPFS and, therefore, DANCES modified its goals to test and prototype using GridFTP, SLASH2 and now SCP as the replacement for GPFS.”

Shown below is an early test DANCES WAN configuration:

DANCES-WAN-env

The authors note that Software Defined Networking has become broadly used terminology, including open and vendor proprietary protocols for network device control communication. Early in the project, OpenFlow was selected as the switch control protocol because it was the most widely deployed and supported, as well as being an open standard. DANCES team identified the version 1.3.0 specification as the minimum target for switch selection in order to provide for complex quality of service frameworks (e.g. rate limiting) that could be provided by implementation of the specification’s per-flow meters in section and per-port queues in section. The per-switch port buffer space was also an important consideration.

Product specifications from Arista, Brocade, Cisco, Dell, Noviflow, and Pica8 were assessed and determined not to “sufficiently meet all the selection criteria while remaining within the project budget.” Five OpenFlow switches were selected for field test, including an early-release Netronome Network Flow Processing Platform (NFPP-1U), a Juniper EX9204, a Juniper MX80-48T, an HP5920 (now HPE), and a Corsa DP6410, which was eventually selected for the project deployment.

Hazlewood expects that the SDN technology employed by DANCES may be adopted by major coordinated HPC networks such as XSEDE within the next five years, with more mainstream use in five to 10 years.

References

[i] Developing Applications with Networking Capabilities via End-to-End SDN (DANCES) project is a collaboration between The  University  of  Tennessee’s  National  Institute  for  Computational   Sciences (UT-NICS), Pittsburgh Supercomputing Center (PSC), Pennsylvania State University (Penn State), the National Center for Supercomputing Applications (NCSA), Texas Advanced Computing Center (TACC), Georgia Institute of Technology (Georgia Tech), the Extreme Science and Engineering Discovery Environment (XSEDE), and Internet2 to investigate and develop the ability to add network bandwidth scheduling via software- defined networking (SDN) programmability to selected cyberinfrastructure services and applications. DANCES, funded by the   National   Science   Foundation’s   Campus Cyberinfrastructure – Network Infrastructure and Engineering (CC-NIE) program award numbers 1341005, 1340953, and 1340981, has field tested five vendor network devices in order to determine which implements the DANCES requirements of the OpenFlow 1.3 standard to provide the network reservation and rate-limiting capability desired to implement the goals of DANCES.

[ii] Developing Applications with Networking Capabilities via End-to-End SDN

(DANCES), by Victor Hazlewood, Kathy Benninger, Greg Peterson, Jason Charcalla, Benny Sparks, Jesse Hanley, Andrew Adams, Bryan Learn, Robert Budden, Derek Simmel, Joseph Lappa, Jared Yanovich; http://www.psc.edu/images/news/XSEDE16-DANCES-final.pdf

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!

Nvidia Debuts Clara AI Toolkit with Pre-Trained Models for Radiology Use

March 19, 2019

AI’s push into healthcare got a boost yesterday with Nvidia’s release of the Clara Deploy AI toolkit which includes 13 pre-trained models for use in radiology. Clara, you may recall, is Nvidia’s biomedical platform Read more…

By John Russell

DARPA, NSF Seek Real-Time ML Processor

March 18, 2019

A new U.S. research initiative seeks to develop a processor capable of real-time learning while operating with the “efficiency of the human brain.” The National Science Foundation (NSF) and the Defense Advanced Re Read more…

By George Leopold

It’s Official: Aurora on Track to Be First U.S. Exascale Computer in 2021

March 18, 2019

The U.S. Department of Energy along with Intel and Cray confirmed today that an Intel/Cray supercomputer, "Aurora," capable of sustained performance of one exaflops, will be delivered by the end of 2021 to Argonne Nation Read more…

By Tiffany Trader

HPE Extreme Performance Solutions

HPE and Intel® Omni-Path Architecture: How to Power a Cloud

Learn how HPE and Intel® Omni-Path Architecture provide critical infrastructure for leading Nordic HPC provider’s HPCFLOW cloud service.

powercloud_blog.jpgFor decades, HPE has been at the forefront of high-performance computing, and we’ve powered some of the fastest and most robust supercomputers in the world. Read more…

IBM Accelerated Insights

The Spark That Ignited A New World of Real-Time Analytics

High Performance Computing has always been about Big Data. It’s not uncommon for research datasets to contain millions of files and many terabytes, even petabytes of data, or more. Read more…

NASA’s Pleiades Simulates Launch Abort Scenarios

March 15, 2019

NASA is using flow simulations running on its Pleiades supercomputer to help design the agency’s next manned spacecraft, Orion. Crew safety is paramount, so NASA engineers are using the HPC cluster to simulate and v Read more…

By George Leopold

Nvidia Debuts Clara AI Toolkit with Pre-Trained Models for Radiology Use

March 19, 2019

AI’s push into healthcare got a boost yesterday with Nvidia’s release of the Clara Deploy AI toolkit which includes 13 pre-trained models for use in radiolo Read more…

By John Russell

It’s Official: Aurora on Track to Be First U.S. Exascale Computer in 2021

March 18, 2019

The U.S. Department of Energy along with Intel and Cray confirmed today that an Intel/Cray supercomputer, "Aurora," capable of sustained performance of one exaf Read more…

By Tiffany Trader

Why Nvidia Bought Mellanox: ‘Future Datacenters Will Be…Like High Performance Computers’

March 14, 2019

“Future datacenters of all kinds will be built like high performance computers,” said Nvidia CEO Jensen Huang during a phone briefing on Monday after Nvidia revealed scooping up the high performance networking company Mellanox for $6.9 billion. Read more…

By Tiffany Trader

Oil and Gas Supercloud Clears Out Remaining Knights Landing Inventory: All 38,000 Wafers

March 13, 2019

The McCloud HPC service being built by Australia’s DownUnder GeoSolutions (DUG) outside Houston is set to become the largest oil and gas cloud in the world th Read more…

By Tiffany Trader

Quick Take: Trump’s 2020 Budget Spares DoE-funded HPC but Slams NSF and NIH

March 12, 2019

U.S. President Donald Trump’s 2020 budget request, released yesterday, proposes deep cuts in many science programs but seems to spare HPC funding by the Depar Read more…

By John Russell

Nvidia Wins Mellanox Stakes for $6.9 Billion

March 11, 2019

The long-rumored acquisition of Mellanox came to fruition this morning with GPU chipmaker Nvidia’s announcement that it has purchased the high-performance net Read more…

By Doug Black

Optalysys Rolls Commercial Optical Processor

March 7, 2019

Optalysys, Ltd., a U.K. company seeking to advance it optical co-processor technology, moved a step closer this week with the unveiling of what it claims is th Read more…

By George Leopold

Intel Responds to White House AI Initiative

March 6, 2019

The Trump Administration’s release last month of the “American AI Initiative,” aimed at prioritizing federal R&D investments in machine intelligence, Read more…

By Doug Black

Quantum Computing Will Never Work

November 27, 2018

Amid the gush of money and enthusiastic predictions being thrown at quantum computing comes a proposed cold shower in the form of an essay by physicist Mikhail Read more…

By John Russell

The Case Against ‘The Case Against Quantum Computing’

January 9, 2019

It’s not easy to be a physicist. Richard Feynman (basically the Jimi Hendrix of physicists) once said: “The first principle is that you must not fool yourse Read more…

By Ben Criger

ClusterVision in Bankruptcy, Fate Uncertain

February 13, 2019

ClusterVision, European HPC specialists that have built and installed over 20 Top500-ranked systems in their nearly 17-year history, appear to be in the midst o Read more…

By Tiffany Trader

Intel Reportedly in $6B Bid for Mellanox

January 30, 2019

The latest rumors and reports around an acquisition of Mellanox focus on Intel, which has reportedly offered a $6 billion bid for the high performance interconn Read more…

By Doug Black

Looking for Light Reading? NSF-backed ‘Comic Books’ Tackle Quantum Computing

January 28, 2019

Still baffled by quantum computing? How about turning to comic books (graphic novels for the well-read among you) for some clarity and a little humor on QC. The Read more…

By John Russell

Why Nvidia Bought Mellanox: ‘Future Datacenters Will Be…Like High Performance Computers’

March 14, 2019

“Future datacenters of all kinds will be built like high performance computers,” said Nvidia CEO Jensen Huang during a phone briefing on Monday after Nvidia revealed scooping up the high performance networking company Mellanox for $6.9 billion. Read more…

By Tiffany Trader

Contract Signed for New Finnish Supercomputer

December 13, 2018

After the official contract signing yesterday, configuration details were made public for the new BullSequana system that the Finnish IT Center for Science (CSC Read more…

By Tiffany Trader

Deep500: ETH Researchers Introduce New Deep Learning Benchmark for HPC

February 5, 2019

ETH researchers have developed a new deep learning benchmarking environment – Deep500 – they say is “the first distributed and reproducible benchmarking s Read more…

By John Russell

Leading Solution Providers

SC 18 Virtual Booth Video Tour

Advania @ SC18 AMD @ SC18
ASRock Rack @ SC18
DDN Storage @ SC18
HPE @ SC18
IBM @ SC18
Lenovo @ SC18 Mellanox Technologies @ SC18
NVIDIA @ SC18
One Stop Systems @ SC18
Oracle @ SC18 Panasas @ SC18
Supermicro @ SC18 SUSE @ SC18 TYAN @ SC18
Verne Global @ SC18

IBM Quantum Update: Q System One Launch, New Collaborators, and QC Center Plans

January 10, 2019

IBM made three significant quantum computing announcements at CES this week. One was introduction of IBM Q System One; it’s really the integration of IBM’s Read more…

By John Russell

IBM Bets $2B Seeking 1000X AI Hardware Performance Boost

February 7, 2019

For now, AI systems are mostly machine learning-based and “narrow” – powerful as they are by today's standards, they're limited to performing a few, narro Read more…

By Doug Black

The Deep500 – Researchers Tackle an HPC Benchmark for Deep Learning

January 7, 2019

How do you know if an HPC system, particularly a larger-scale system, is well-suited for deep learning workloads? Today, that’s not an easy question to answer Read more…

By John Russell

HPC Reflections and (Mostly Hopeful) Predictions

December 19, 2018

So much ‘spaghetti’ gets tossed on walls by the technology community (vendors and researchers) to see what sticks that it is often difficult to peer through Read more…

By John Russell

Arm Unveils Neoverse N1 Platform with up to 128-Cores

February 20, 2019

Following on its Neoverse roadmap announcement last October, Arm today revealed its next-gen Neoverse microarchitecture with compute and throughput-optimized si Read more…

By Tiffany Trader

Move Over Lustre & Spectrum Scale – Here Comes BeeGFS?

November 26, 2018

Is BeeGFS – the parallel file system with European roots – on a path to compete with Lustre and Spectrum Scale worldwide in HPC environments? Frank Herold Read more…

By John Russell

France to Deploy AI-Focused Supercomputer: Jean Zay

January 22, 2019

HPE announced today that it won the contract to build a supercomputer that will drive France’s AI and HPC efforts. The computer will be part of GENCI, the Fre Read more…

By Tiffany Trader

Microsoft to Buy Mellanox?

December 20, 2018

Networking equipment powerhouse Mellanox could be an acquisition target by Microsoft, according to a published report in an Israeli financial publication. Microsoft has reportedly gone so far as to engage Goldman Sachs to handle negotiations with Mellanox. Read more…

By Doug Black

  • arrow
  • Click Here for More Headlines
  • arrow
Do NOT follow this link or you will be banned from the site!
Share This