HPC Midlands - Supercomputing for Research and Industry (Hartree Centre presentation)

•Download as PPTX, PDF•

0 likes•541 views

These are the slides for my talk on HPC Midlands at the Hartree Centre's "HPC As A Service for Industry" event at STFC Daresbury Labs in January 2013. For more information on the HPC Midlands project, please see http://hpc-midlands.ac.uk

Education

Martin Hamilton, Centre Manager
hpc-midlands.ac.uk

Contents
• UK e-Infrastructure initiative
• HPC Midlands
• Industrial engagement
• Lessons learned &
opportunities

UK e-Infrastructure Programme
• OSI e-Infrastructure Working Group
• RCUK e-Infrastructure Advisory Group
• £158m BIS e-Infrastructure investment
• Tildesley Report: http://goo.gl/VSw4x
• Regional HPC consortia funded via EPSRC

What is HPC Midlands?
• HPC on demand, delivered via JANET
• Consortium of Loughborough University &
University of Leicester
• Managed service delivered by Bull
• ISV support, e.g. ANSYS and CD-adapco
• Funded by EPSRC/BIS e-Infrastructure
initiative (£1m hardware grant + ‘recurrent’)

Hera, the HPC Midlands cluster
• 3,000 cores (48 Teraflops)
• 11 chassis (18 blades each)
• 15TB RAM
• 120TB Lustre storage
• Non-blocking QDR Infiniband
• 188 compute node blades
– 2 x 2.0GHz (8 core) Sandy Bridge
– 140 with 64GB RAM (4GB/core)
– 48 with 128GB RAM (8GB/core)

HPC Midlands Applications
Large Eddy Simulation of
Premixed Combustion in Spark
Ignition Engines.

Prof. W. Malalasekera
Dr. A. Clarke
C. Ranasinghe

HPC Midlands Applications

Large Eddy Simulation of Crossflow
Vortices on an Infinite Swept Wing

V. I. Mistry
G .J. Page
J. J. McGuirk

42nd AIAA Fluid Dynamics
Conference

DOI: 10.1016/j.paerosci.2011.12.002

HPC Midlands Applications

CFD Based Study of Unconventional
Aeroengine Exhaust Systems

T. Coates
G. J. Page

30th AIAA Applied Aerodynamics
Conference

DOI: 10.2514/6.2012-2775

Industrial Engagement

Strategic partnerships
Local / Regional firms & initiatives
Science Parks and Incubators

HPC Midlands Case Study – E.ON

Key points:

• Steady state, complex geometry,
simple physics.
• Simulations with 24 to 128 cores
• 6 different meshes used from
4.5 to 80 million cells
• Speed-ups from 30 to 145 times
observed vs. in-house system

ANSYS CFX simulation of
leak in gas turbine enclosure

HPC Midlands Case Study – E.ON
Normalized Timings for HPC Runs

CPU core time per million cells
180 169.31
Grid 1 4.5M elements 24 cores
160

per 100 iterations (s)
Grid 2 8M elements 48 cores 140
120
Grid 3 15M elements 48 cores 93.18
100 84.81
Grid 4 25M elements 60 cores 80 67.62

Grid 5
60 39.18
45M elements 128 cores 36.81
40
Grid 6 80M elements 128 cores 20
0
Grid 1 … … … … …

“Straightforward to use, secure and fast” 
“Biggest advantage is for jobs that:
 Have large parameter spaces
 Are time dependent
 Have complex geometry
 Have very complex physics
 A combination of the above”

Lessons Learned
• Making the connection
• Paperwork
• Software licensing
• IPR, NDA and other TLAs
• Plumbing
• Sneakernet
• Connectivity
• Moonshot

Paperwork – Software Licensing

user@hpc-midlands.ac.uk

Plumbing - Project Moonshot
• JANET(UK) initiative
• Like eduroam, but for any protocol
• Use your existing credentials everywhere
• Securely tunnelled back to home RADIUS server
• Applicable to academia and industry?
• 18 month pilot, kickoff in April
• https://www.ja.net/products-services/janet-
futures/moonshot

The Implementing AI: High Performance Architectures webinar, hosted by KTN and eFutures, was the fourth event in the Implementing AI webinar series. The focus of the webinar was the impact of processing AI data on data centres - particularly from the technology perspective. Prof. Simon McIntosh-Smith, Professor in High Performance Computing, University of Bristol, covered Large scale HPC hardware in the age of AI.

Negotiating: It's two pit bulls locked in a room and one is going to be forced to roll over. Ick. That mentality is not only outdated, but will get you nowhere. That's because, frankly, baring teeth and barking the loudest doesn't have the power it might seem to when bargaining. When deal making gets tense, no one ever wants to back down. The savviest negotiators take nothing personally; they are impervious to criticism and impossible to fluster. And because they seem unmoved by the whole situation and unimpressed with the stakes involved, they have a way of unnerving less-experienced counterparts. This can be an effective weapon when used against entrepreneurs, because entrepreneurs tend to take every aspect of their businesses very personally. Entrepreneurs often style themselves as frank, no-nonsense individuals, and they can at times have thin skin. But whenever you negotiate, remember that it pays to stay calm, to never show that an absurdly low counter-offer or an annoying stalling tactic has upset you. Use your equanimity to unnerve the person who is negotiating with you. And if he or she becomes angry or peeved, don't take the bait to strike back. Just take heart: You've grabbed the emotional advantage in the situation. Now go close that deal.

Super computer 2017

Sampath Bhargav Pinnam

HPC Midlands - E.ON Supercomputing Case Study

Martin Hamilton

How HPC and large-scale data analytics are transforming experimental science

inside-BigData.com

In this deck from DataTech19, Debbie Bard from NERSC presents: Supercomputing and the scientist: How HPC and large-scale data analytics are transforming experimental science. "Debbie Bard leads the Data Science Engagement Group NERSC. NERSC is the mission supercomputing center for the USA Department of Energy, and supports over 7000 scientists and 700 projects with supercomputing needs. A native of the UK, her career spans research in particle physics, cosmology and computing on both sides of the Atlantic. She obtained her PhD at Edinburgh University, and has worked at Imperial College London as well as the Stanford Linear Accelerator Center (SLAC) in the USA, before joining the Data Department at NERSC, where she focuses on data-intensive computing and research, including supercomputing for experimental science and machine learning at scale." Watch the video: https://wp.me/p3RLHQ-kLV Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter

40 Powers of 10 - Simulating the Universe with the DiRAC HPC Facility

inside-BigData.com

In this deck from the Swiss HPC Conference, Mark Wilkinson presents: 40 Powers of 10 - Simulating the Universe with the DiRAC HPC Facility. "DiRAC is the integrated supercomputing facility for theoretical modeling and HPC-based research in particle physics, and astrophysics, cosmology, and nuclear physics, all areas in which the UK is world-leading. DiRAC provides a variety of compute resources, matching machine architecture to the algorithm design and requirements of the research problems to be solved. As a single federated Facility, DiRAC allows more effective and efficient use of computing resources, supporting the delivery of the science programs across the STFC research communities. It provides a common training and consultation framework and, crucially, provides critical mass and a coordinating structure for both small- and large-scale cross-discipline science projects, the technical support needed to run and develop a distributed HPC service, and a pool of expertise to support knowledge transfer and industrial partnership projects. The on-going development and sharing of best-practice for the delivery of productive, national HPC services with DiRAC enables STFC researchers to produce world-leading science across the entire STFC science theory program." Watch the video: https://wp.me/p3RLHQ-k94 Learn more: https://dirac.ac.uk/ and http://hpcadvisorycouncil.com/events/2019/swiss-workshop/agenda.php Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter

Design Automation Approaches for Real-Time Edge Computing for Science Applic...

Facultad de Informática UCM

Many emerging applications require methods tailored towards high-speed data acquisition and filtering of streaming data followed by offline event reconstruction and analysis. In this case, the main objective is to relieve the immense pressure on the storage and communication resources within the experimental infrastructure. In other applications, ultra low latency real time analysis is required for autonomous experimental systems and anomaly detection in acquired scientific data in the absence of any prior data model for unknown events. At these data rates, traditional computing approaches cannot carry out even cursory analyses in a time frame necessary to guide experimentation. In this talk, Prof. Ogrenci will present some examples of AI hardware architectures. She will discuss the concept of co-design, which makes the unique needs of an application domain transparent to the hardware design process and present examples from three applications: (1) An in-pixel AI chip built using the HLS methodology; (2) A radiation hardened ASIC chip for quantum systems; (3) An FPGA-based edge computing controller for real-time control of a High Energy Physics experiment.

Hpc, grid and cloud computing - the past, present, and future challengeJason Shih

HPC Infrastructure To Solve The CFD Grand Challenge

Anand Haridass

AI Hardware for Real-Time Machine Learning

Facultad de Informática UCM

"Computing systems for AI workloads have evolved towards data-center clusters of GPUs and TPUs, with architectures optimized for performing linear algebra and tunable for variable precision. As new AI paradigms emerge, more distinct divergence between hardware architectures for powering AI and other workloads are observed. GPU manufacturers are developing different architectures and chipsets for the HPC/supercomputing, cloud, edge computing, and robotics domains. FPGA vendors are also joining this ecosystem (e.g., Intel FPGAs deployed within Microsoft Azure). Moving forward, many industries and services ranging from cloud computing to consumer electronics are making hardware-accelerated AI a prominent component in their portfolio. In this talk, some examples of AI hardware architectures and available silicon technologies will be presented. The concept of co-design will be discussed. This makes the unique needs of an application domain transparent to the hardware design process. Finally, an overview of design automation tool flows will be presented to gain an understanding of how to support a high productivity framework for domain experts to design and deploy AI hardware."

Current Trends in HPCPutchong Uthayopas

What's hot

Technology Outlook - The new Era of computing

Swiss Big Data User Group

Working together with SURF Raymond Oonk Annette Langedijk SURF

CommunicatieSURF

Bartolomeo_HandoutISSRDDr. Per Christian Steimle

Available HPC resources at CSUC

CSUC - Consorci de Serveis Universitaris de Catalunya

Tesla personal super computer

Priya Manik

Terma Aerostructures

Plastindustrien

Telekom Conference_2016_Brief_DescriptionMihail Mihaylov

WSE 6A-Octo-X Terrain Mapping UAVManuel De La Cruz

Access to Open Earth Observation Data, an Overview and Outlook Raymond Sluit...

CommunicatieSURF

Param yuva iiKavish Jaiswal

Seagate-enterprise-exos-7 e2-1tb-3.5-datasheet

Mr Cuong

OCP awareness course taster session 01John Laban

3 Great Networks & Switches Whiteboard 2016

Jeff Green

Super computer 2017

Sampath Bhargav Pinnam

What's hot (14)

Technology Outlook - The new Era of computing

Working together with SURF Raymond Oonk Annette Langedijk SURF

Bartolomeo_HandoutISSRD

Available HPC resources at CSUC

Tesla personal super computer

Terma Aerostructures

Telekom Conference_2016_Brief_Description

WSE 6A-Octo-X Terrain Mapping UAV

Access to Open Earth Observation Data, an Overview and Outlook Raymond Sluit...

Param yuva ii

Seagate-enterprise-exos-7 e2-1tb-3.5-datasheet

OCP awareness course taster session 01

3 Great Networks & Switches Whiteboard 2016

Super computer 2017

Similar to HPC Midlands - Supercomputing for Research and Industry (Hartree Centre presentation)

HPC Midlands - E.ON Supercomputing Case Study

Martin Hamilton

How HPC and large-scale data analytics are transforming experimental science

inside-BigData.com

40 Powers of 10 - Simulating the Universe with the DiRAC HPC Facility

inside-BigData.com

Design Automation Approaches for Real-Time Edge Computing for Science Applic...

Facultad de Informática UCM

Hpc, grid and cloud computing - the past, present, and future challengeJason Shih

HPC Infrastructure To Solve The CFD Grand Challenge

Anand Haridass

AI Hardware for Real-Time Machine Learning

Facultad de Informática UCM

Current Trends in HPCPutchong Uthayopas

TeraGrid Communication and Computation

Tal Lavian Ph.D.

4 TeraGrid Sites Have Focal Points: SDSC – The Data Place Large-scale and high-performance data analysis/handling Every Cluster Node is Directly Attached to SAN NCSA – The Compute Place Large-scale, Large Flops computation Argonne – The Viz place Scalable Viz walls Caltech – The Applications place Data and flops for applications – Especially some of the GriPhyN Apps Specific machine configurations reflect this

Andrew J Younge - Vanguard Astra - Petascale Arm Platform for U.S. DOE/ASC Su...

Linaro

Event: Arm Architecture HPC Workshop by Linaro and HiSilicon Location: Santa Clara, CA Speaker: Andrew J Younge Talk Title: Vanguard Astra - Petascale Arm Platform for U.S. DOE/ASC Supercomputing Talk Desc: The Vanguard program looks to expand the potential technology choices for leadership-class High Performance Computing (HPC) platforms, not only for the National Nuclear Security Administration (NNSA) but for the Department of Energy (DOE) and wider HPC community. Specifically, there is a need to expand the supercomputing ecosystem by investing and developing emerging, yet-to-be-proven technologies and address both hardware and software challenges together, as well as to prove-out the viability of such novel platforms for production HPC workloads. The first deployment of the Vanguard program will be Astra, a prototype Petascale Arm supercomputer to be sited at Sandia National Laboratories during 2018. This talk will focus on the arthictecural details of Astra and the significant investments being made towards the maturing the Arm software ecosystem. Furthermore, we will share initial performance results based on our pre-general availability testbed system and outline several planned research activities for the machine. Bio: Andrew Younge is a R&D Computer Scientist at Sandia National Laboratories with the Scalable System Software group. His research interests include Cloud Computing, Virtualization, Distributed Systems, and energy efficient computing. Andrew has a Ph.D in Computer Science from Indiana University, where he was the Persistent Systems fellow and a member of the FutureGrid project, an NSF-funded experimental cyberinfrastructure test-bed. Over the years, Andrew has held visiting positions at the MITRE Corporation, the University of Southern California / Information Sciences Institute, and the University of Maryland, College Park. He received his Bachelors and Masters of Science from the Computer Science Department at Rochester Institute of Technology (RIT) in 2008 and 2010, respectively.

Accelerators at ORNL - Application Readiness, Early Science, and Industry Impact

inside-BigData.com

Opti struct tips_and_tricks_atcAltair

OPAL-RT Seminar on HYPERSIM

OPAL-RT TECHNOLOGIES

DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...

zionsaint

Tsinghua University: Two Exemplary Applications in China

DataStax Academy

In this talk, we will share the experiences of applying Cassandra with two real customers in China. In the first use case, we deployed Cassandra at Sany Group, a leading company of Machinery manufacturing, to manage the sensor data generated by construction machinery. By designing a specific schema and optimizing the write process, we successfully managed over 1.5 billion historical data records and achieved the online write throughput of 10k write operations per second with 5 servers. MapReduce is also used on Cassandra for valued-added services, e.g. operations management, machine failure prediction, and abnormal behavior mining. In the second use case, Cassandra is deployed in the China Meteorological Administration to manage the Meteorological data. We design a hybrid schema to support both slice query and time window based query efficiently. Also, we explored the optimized compaction and deletion strategy for meteorological data in this case.

Barcelona Supercomputing Center, Generador de Riqueza

Facultad de Informática UCM

El Barcelona Supercomputing Center (BSC) fue establecido en 2005 y alberga el MareNostrum, uno de los superordenadores más potentes de España. Somos el centro pionero de la supercomputación en España. Nuestra especialidad es la computación de altas prestaciones - también conocida como HPC o High Performance Computing- y nuestra misión es doble: ofrecer infraestructuras y servicio de supercomputación a los científicos españoles y europeos, y generar conocimiento y tecnología para transferirlos a la sociedad. Somos Centro de Excelencia Severo Ochoa, miembros de primer nivel de la infraestructura de investigación europea PRACE (Partnership for Advanced Computing in Europe), y gestionamos la Red Española de Supercomputación (RES). Como centro de investigación, contamos con más de 456 expertos de 45 países, organizados en cuatro grandes áreas de investigación: Ciencias de la computación, Ciencias de la vida, Ciencias de la tierra y aplicaciones computacionales en ciencia e ingeniería.

Valladolid final-septiembre-2010TELECOM I+D

Hardware architecture of Summit Supercomputer

VigneshwarRamaswamy

A Dataflow Processing Chip for Training Deep Neural Networks

inside-BigData.com

Project Slides for Website 2020-22.pptx

AkshitAgiwal1

Similar to HPC Midlands - Supercomputing for Research and Industry (Hartree Centre presentation) (20)

HPC Midlands - E.ON Supercomputing Case Study

How HPC and large-scale data analytics are transforming experimental science

40 Powers of 10 - Simulating the Universe with the DiRAC HPC Facility

Design Automation Approaches for Real-Time Edge Computing for Science Applic...

Hpc, grid and cloud computing - the past, present, and future challenge

HPC Infrastructure To Solve The CFD Grand Challenge

AI Hardware for Real-Time Machine Learning

Current Trends in HPC

TeraGrid Communication and Computation

Andrew J Younge - Vanguard Astra - Petascale Arm Platform for U.S. DOE/ASC Su...

Accelerators at ORNL - Application Readiness, Early Science, and Industry Impact

Opti struct tips_and_tricks_atc

OPAL-RT Seminar on HYPERSIM

DARPA ERI Summit 2018: The End of Moore’s Law & Faster General Purpose Comput...

Tsinghua University: Two Exemplary Applications in China

Barcelona Supercomputing Center, Generador de Riqueza

Valladolid final-septiembre-2010

Hardware architecture of Summit Supercomputer

A Dataflow Processing Chip for Training Deep Neural Networks

Project Slides for Website 2020-22.pptx

More from Martin Hamilton

Keep taking the tablets? The graduation of the iPad generation

Martin Hamilton

Over the last ten years, smartphones and tablets and pervasive Internet connectivity have taken over our lives, and more importantly the lives of our children. In this talk for Loughborough University's 2018 Learning and Teaching conference I reflect on how higher education could respond to this trend. I invite the audience to consider what children who have always had access to the sum total of human knowledge will will expect and need from universities in the future, and how we might best support learners from disadvantaged backgrounds to thrive

The Intelligent Campus - Where the Internet of Things meets AI - HESCA June 2018

Martin Hamilton

The Digital Book Thief has a Napster Moment - Edinburgh Near Future Library S...

Martin Hamilton

We’ve grown accustomed to being able to call up any information, anywhere, any time - but what happens when our digital landlord forecloses? When that service we entrusted our data to goes to the Startup Graveyard? In this talk for the University of Edinburgh's Near Future Library Symposium I highlight some of the risks to our culture and the scientific record, and what the near future library can do to help

Martin Hamilton - The wind from nowhere - Horizon scanning in an uncertain ag...

Martin Hamilton

Does it feel like the world has recently become much less certain and predictable? We tend to expect that each day will be like the last, more or less, but incremental change is increasingly feeling like a thing of the past. In this talk for the 2018 Scientia EMEA User Conference I look at how we can read the signals around us to better predict the future, and consider the impact of technologies and trends like blockchain, artificial intelligence and Brexit on further and higher education

From Blockchain to Brexit - edtech trends for 2018 - BETT 2018

Martin Hamilton

In this talk for BETT 2018 I take a look at a few of the socio-technical trends that are set to have a big impact on universities and colleges in 2018 from blockchain to Brexit, and data vandalism to UK spaceports. I look at some approaches that institutions can take to help plan for an uncertain future, and consider how the community can mobilise to protect the progressive values that now often seem to be under threat.

Martin Hamilton - Digital skills: You won't believe what happened next!

Martin Hamilton

Martin Hamilton - Librarians in Outer Space - CILIP invited talk

Martin Hamilton

We are becoming used to living in an interconnected world, with vast amounts of data at our fingertips, but what happens when our preconceptions are challenged? What happens when the things that we take for granted simply don’t work any more? How can librarians rise to the challenge? In this invited talk for CILIP, I reflect on the impact for libraries and librarians of some of the defining narratives of the late Anthropocene era: from climate change and failed states to cheap space travel and artificial intelligence

Martin Hamilton - The impact of technology on the higher education sector - L...

Martin Hamilton

Martin Hamilton - Robots and AI, the calm before the Singularity? - BCS invit...

Martin Hamilton

In this invited talk for the BCS I look at the state of the art in robots and artificial intelligence, and what this tells us about the near-term future that our children and grandchildren will live in. Will the imminent arrival of AI powered self-driving vehicles mean redundancies for truck and taxi drivers? What will these people do next? I also show how robots and AI are already becoming commonplace, working in places like shops, restaurants and distribution centres

Martin Hamilton - What did your AI make today? - BCS invited talk

Martin Hamilton

Artificial intelligence and machine learning are used extensively by Internet services to help us find and classify information. This is how products like Google Photos can "magically" find all your photos of cats or trees - or cats up trees! But there's a new wave of AI called generative adversarial neural networks that's all about using AI to make things. In this invited talk for the British Computer Society (BCS) I reflect on what we could use this technology for and the implications for people and society.

Blockchain in research and education - UKSG Webinar - September 2017

Martin Hamilton

HPC in the cloud comes of age - Red Oak HPC Seminar

Martin Hamilton

What does HPC in the cloud look like in 2017, and how did we get there? In this talk for Red Oak's HPC Seminar, I look at the origins of cloud HPC, and how it has become mainstream through technologies like Amazon Web Services and OpenStack. I also offer a sneak preview of the 2017 UK national e-Infrastructure survey results, and some thoughts about what's next in cloud HPC, from hyperscale providers to the momentum behind container technologies from Docker and the Open Container Initiative.

Imagining Mars University - Universities UK 2017 conference

Martin Hamilton

Back to the future - Future Proof IT 2017

Martin Hamilton

Tech in exams - SQA Assessment Expert Group - June 2017

Martin Hamilton

Through the Overton Window - Health Education England horizon scanning worksh...

Martin Hamilton

A new life awaits you in the off world colonies - UCISA Spotlight on Digital ...

Martin Hamilton

Help! My robot is a teacher! - Future Edtech 2017

Martin Hamilton

Towards a UK Edtech Strategy - Edtech Vision 2020

Martin Hamilton

Bridging the digital divide - Digital Skills Summit 2017

Martin Hamilton

More from Martin Hamilton (20)