Decommissioned: General Purpose Cluster (GPC)

November 9, 2016 in Decommissioned Systems


The General Purpose Cluster (GPC) was our extremely large “workhorse” cluster (ranked 16th in the world at its inception in 2009, then the fastest in Canada) and was where most computations were done at SciNet until its decommission and the its successor, Niagara, went into production.  It was an IBM iDataPlex cluster based on Intel’s Nehalem architecture (one of the first in the world to make use of these chips).

The GPC consisted of 3,780 nodes (IBM iDataPlex DX360M2) with a total of 30,240 cores (Intel Xeon E5540) at 2.53GHz, with 16GB RAM per node (2GB per core) with some larger-memory nodes up to 32GB.   The nodes ran Linux.   Approximately one quarter of the cluster was connected with non-blocking DDR InfiniBand while the rest of the nodes were connected with 5:1 blocked QDR InfiniBand. The compute nodes were accessed through a queuing system that allowed jobs with a maximum wall time of 48 hours and a minimum time, in the batch queue, of 15 minutes.

A “quickstart” guide to using SciNet’s GPC could be found on our technical documentation wiki.

The GPC was decommissioned in April 2018.

HPC Postdoc, BG/Q, IBM (Toronto)

October 9, 2014 in HPC Jobs, HPC Jobs Ontario

For more information or to apply, view this webpage.

Job Responsibilities:
•        Assist research teams in making effective use of massively parallel platforms including one of the largest supercomputers in Canada and a new research platform for exascale computing
•        Develop, port, optimize and analyze scientific numerical codes using >1,024 cores on the BlueGene Q using Fortran, Python, C, C++ etc
•        Parallel-programming assistance on smaller-scale x86 and Power clusters, possibly including FPGA’s
•        Install, upgrade and lead day-to-day operations for scientific infrastructure including mathematical software packages, libraries, compilers, visualization software, schedulers etc
•        promote, advise and teach mini-courses on scientific computing and parallel programming

Required:
•        Minimum 3 years experience using and developing large-scale parallel scientific applications making use of both MPI and OpenMP for parallellization
•        Experience and understanding scientific numerical codes, compilers, code optimization, Linux kernel & OS
•        Knowledge of Fortran, C and C++ under  Linux/Unix
•        Proven communication skills, both written and verbal

High Performance Computing Specialist, SHARCNET/UOIT (Oshawa)

July 30, 2014 in HPC Jobs, HPC Jobs Ontario

University of Ontario Institute of Technology (UOIT) in partnership with the Shared Hierarchical Academic Research Computing Network (SHARCNET) is currently seeking a High-Performance Computing Technical Consultant, based out of UOIT. SHARCNET is a consortium of 18 universities and colleges operating a network of high-performance computer clusters across south western, central and northern Ontario.  SHARCNET is a partner organization of Compute Ontario and Compute Canada.  SHARCNET and these other organizations provide advanced computing resources and services for researchers.  See www.sharcnet.ca for more information.

Job Summary:

This position reports to the SHARCNET Technical Manager and is part of the technical team that provides HPC support across the consortium.  Locally, this position will report to the UOIT site lead.

For more information, or to apply, visit this link.

HPC Analyst – Research Scientist, IBM/SOSCIP (Toronto)

June 23, 2014 in HPC Jobs, HPC Jobs Ontario

Required:
• PhD in quantitative science, statistics or engineering (e.g. Mech or Electrical engineering, Computer Science, Astrophysics, Computational biology)
• Experience using and developing large-scale cloud-based data analytics tools such as R, SPSS, and Cognos
• experience and understanding scientific numerical codes, compilers, code optimization, Linux kernel & OS
• Proven knowledge of R, Fortran, C and or C++ under Linux/Unix
• Demonstrated communication skills, both written and verbal

Preferred:
• Experience in multi-disciplinary projects where team has a diverse set of domain expertise
• Experience working directly with clients and/or industry partners

Job Responsibilities:
• Assist research teams in making effective use of IBM Big Data software
• Install, upgrade and lead day-to-day operations for scientific infrastructure including mathematical software packages, libraries, visualization software

Work Locations
• Toronto, London, Oshawa, or Kingston

For more information, or to apply, see this web page.

Research Scientist, Amazon (Toronto)

April 20, 2014 in HPC Jobs, HPC Jobs Ontario

Basic Qualifications

· Master’s Degree in Computer Science or related technology field
· Proficiency in at least one modern programming language such as C, C++, C#, Java, or Perl
· 2+ years of experience in operations research or research science
· Ability to research, identify, evaluate, and implement potential changes to complex algorithms
· Familiar with fundamentals of statistics and data analysis
· Ability to find simple solutions to ambiguous large-scale problems
· Experience with production scheduling algorithms and optimization.
· Familiar with fundamentals of combinatorial optimization

To find out more about this job, or to apply, visit this web page.

Ontario Vouchers and SciNet

July 10, 2013 in for_industry

The Ontario Collaboration Vouchers program is a great new opportunity for Ontario companies to work with academic partners — like SciNet — to solve their R&D problems.

Matching funds up to $20,000 – and possibly more with eligibility for other programs – are available to help you work on short (6-12 month), focused projects with academic partners like SciNet. This opportunity – part of the Ontario Government’s new Ontario Collaboration Vouchers for Innovation and Productivity (VIP) program – may be the best opportunity yet to take advantage of SciNet’s R&D computing expertise for speeding up and scaling up your research computing.
A small ($5,000 or less) cash investment will leverage four times as much investment when paired with in-kind matching such as the time you spend working with us. That means you can get access to our large computing facilities (use up to a quarter of a million processor- hours) ; use our large storage (up to 25TB of storage) ; take advantage of our training options (up to 250 student-hours); and work with our consultants one-on-one to tackle your biggest R&D computing challenges (up to 250 consultant-hours). And we’ll start a pilot project with you immediately while the six-to-eight week application process is underway.

Want to know more? Read our brochure about the vouchers program, and contact us to start solving your R&D computing problems.

HPC Systems Administrator (Toronto)

June 30, 2013 in HPC Jobs, HPC Jobs Ontario

Currently seeking a Senior LAMP Administrator for our valued client located downtown Toronto. As the Senior LAMP Administrator, you will be responsible for configuring, monitoring and maintaining two High Performance Cluster (HPC) environments. Additionally, as the Senior LAMP Administrator, you will be responsible for ensuring that all servers and related storage systems operate at optimum levels. Due to the close contact you will have with internal groups consisting of Scientists and Researchers, the prime candidate must possess excellent verbal and written communication skills. The Senior LAMP Administrator should be comfortable working independently and collaboratively within a team environment, should be vocal in bringing solutions to table discussions, as should be willing to work flexible hours.

For more information and to apply, see this website.

SciNet Applications Analyst

April 28, 2013 in Uncategorized

SciNet

Come work for us at SciNet, helping solve the biggest computing problems in Canadian Research!

Applications are due May 6th.

Under general direction, the SciNet Applications Analyst provides senior IT services in parallel programming for the SciNet High Performance Computing (HPC) consortium which serves researchers at the University of Toronto including faculty, postdoctoral fellows and graduate students in all disciplines and fields (e.g. science and engineering, finance, languages, etc.). The incumbent will be involved in “massively-parallel computing” with the task of running single jobs that simultaneously make use of up to 30,000 cores. S/he will also work with researchers and research teams to plan, develop, install and optimize the initial SciNet systems for various research programs and will provide technical consultation to researchers on their system needs for research operations. SciNet operates the two largest High Performance Computing (HPC) systems in Canada and provides HPC resources and support to researchers at the University of Toronto, the affiliated research hospitals and other Canadian universities.

To find out more and apply, go to https://utoronto.taleo.net/careersection/10000/jobdetail.ftl?lang=en&job=1300283

Supercomputing to Save the Planet

April 9, 2013 in in_the_news

Dworkin and his team of six graduate students use SciNet to run complex simulations that measure combustion chemistry, particle diffusion, fuel temperature, pressure, and a host of other variables.
With rapid industrialization in developing nations like China and India, Dworkin’s research could have a profound influence on climate change and pollution-borne illness.

Read more at The Eye-Opener

SciNet’s Top 5 of 2012

December 17, 2012 in blog, blog-general

scinet-new-year

 

It’s been a great, busy, year here at SciNet in 2012;  here’s our take on a SciNet Year in Review as a countdown to what’s already shaping up to be an exciting 2013…

5!  SciNet Training

Training and education have always been important to SciNet; it’s one thing to provide computer resources, but we enable research at scale by teaching researchers how to make use of the computers for their work.   In 2012:

  • We held 100 hours of courses, seminars, and Tech Talks,
  • Representing 1,138 student hours of teaching; and
  • Started a new SciNet Certificate program so that attendees could get recognition for all the effort they were putting in.

..4! SciNet Websites

We launched our new and clearer SciNetHPC.ca website this year, which will be the one-stop-shop for news, events and features about the science being enabled here at SciNet; but don’t worry, our technical wiki, wiki.SciNetHPC.ca, will still be there for all your documentation and training material needs.  Some website highlights:

  • Our new Careers page keeps track of Ontario jobs for researchers with HPC experience;
  • Our technical wiki served it’s 750,000th page view this year, and that doesn’t even count the downloads of PDFs of training materials or views of video-recorded educational sessions.

…3! Big Storage for SciNet Users

This was the year we made available our large HPSS system for long-term storage available to our users.  This sophisticated large-scale storage system allows our users reliable near-line storage for very-large data sets.  On tape we already have

  • Over 63 million files, and
  • Over 1.4 petabytes of usage

….2! SOSCIP and the Blue Gene/Q

As part of the SOSCIP project for accelerating research and innovation across Southern Ontario, SciNet took delivery of and is running and supporting the SOSCIP Blue Gene/Q systems.   These new systems are:

…..1! Future SciNet-ers; Outreach to High Schools

And the top highlight of the year has to go to working with the great high school students at SATEC in Toronto, who built a supercomputer of their own, learned to program it with MPI and OpenMP, and demo’ed it to their local MPP.    These students will be the supercomputing experts and data scientists of the future, and it was a pleasure to work with them.

So thanks for all your emails, tweets, and support through the year, have a wonderful holiday, and…

Happy New Year!