SciNet News September 2018

September 14, 2018 in for_researchers, for_users, newsletter

SUMMARY

  • Niagara takes the number 53 spot in the June 2018 TOP500 list of supercomputers (https://www.top500.org/list/2018/06).
  • Scratch purging policy on Niagara is in effect.
  • Burst buffer available on demand.
  • Various SciNet courses and events to start next week, including a “Intro to Niagara/SciNet” session and a TechTalk on “Machine Learning Cosmic Structure Formation” on Sept 12.
  • SciNet’s Jupyterhub with access to files on Niagara is online.
  • my.SciNet website with access to your Niagara jobs records is online.
  • Courses website now accessible with your Compute Canada password.

SYSTEM NEWS

EVENTS COMING UP

Registration for SciNet courses is done by logging into https://courses.scinet.utoronto.ca with your Compute Canada username
and password.

Many of the events are at the teaching room or boardroom in the SciNet offices on the eleventh floor of the MaRS West Tower, suite 1140A (661 University Avenue, Toronto ON M5G 1M1). SciNet events are often recorded and broadcast (see the courses site for links).

  • INTRO TO SCINET AND NIAGARA
    Wednesday Sept 12, 2018, 10:00 am – 11:30 am
    SciNet Boardroom (suite 1140, 661 University Avenue, Toronto).

    This is a class of approximately 90 minutes to introduce SciNet and the new supercomputer Niagara and teach you how to use Niagara.

    Participation counts towards the SciNet HPC Certificate.

    For more information and (free) registration, go to https://courses.scinet.utoronto.ca/404

  • SCINET USER GROUP MEETING
    Wednesday Sept 12, 2018, 12:00 noon – 1:00 pm
    SciNet Boardroom (suite 1140, 661 University Avenue, Toronto).

    The SciNet Users Group (SNUG) meetings are every month on the second Wednesday (except during the summer), and involve pizza, user discussion, feedback, and a half-hour talk on topics or technologies of interest to the SciNet community.

    The TechTalk will be on

    MACHINE LEARNING COSMIC STRUCTURE FORMATION

    by George Stein (Dept. of Astronomy-UofT, CITA).

    Abstract: In modern astrophysics and cosmology, accurate simulations of the large scale structure of the universe are necessary. Usually, this is accomplished by so called N-body simulations, which calculate the full gravitational collapse of a region of the universe over its 14 billion year history. Instead of calculating this costly gravitational evolution, we trained a three-dimensional deep Convolutional Neural Network (CNN) to identify dark matter proto-haloes directly from the cosmological initial conditions, and showed that a CNN of this type can be a viable alternative in some cases. In this talk I will discuss current cosmological simulations and the invasion of machine learning techniques, with a focus on our work. For more information see https://arxiv.org/abs/1805.04537.

    For sign up and more information, see https://courses.scinet.utoronto.ca/410

  • INTRODUCTION TO COMPUTATIONAL BIOSTATISTICS WITH R (MSC1090)
    Tuesdays and Thursday, 11 am – 12 noon
    Twelve weeks starting Sept 12.

    In this course data analysis techniques utilizing the R statistical language, will be discussed and introduced, as well as, the basics of programming and scientific computing. The goal of this course is to prepare graduate students to perform scientific data analysis. Successful students will learn how to use statistical inference tools to gain insight into large and small data sets, as well as be exposed to cutting-edge techniques and best practises to store, manage and analyze (large) data. Topics include: R programming, version control, automation, modular programming and scientific visualization.

    Students willing to take the course as part of their graduate program have to enroll through Acorn/ROSI. This course is part of the IMS graduate program and to be taught at the UofT St. George campus (i.e., not in the SciNet classroom). Contact us if you wish to audit the course without credit.

    This course will be recorded, but not broadcast.

    For more information, see https://courses.scinet.utoronto.ca/399

  • INTRODUCTION TO THE LINUX SHELL
    Wednesday Sept 19, 1:00 pm – 4:00 pm
    Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    Working with many of the HPC systems in Ontario involves using the Linux/UNIX command line. This provides a very powerful interface, but it can be quite daunting for the uninitiated. In this half-day session, you can become initiated with this course. This hands on session will cover basic commands and scripting. It could be a great boon for your productivity!

    Participation counts towards the SciNet Scientific Computing Certificate.

    For sign up and more information, see https://courses.scinet.utoronto.ca/407

  • INTRODUCTION TO PROGRAMMING
    Tuesdays and Thursdays, 1:00 pm – 2:00 pm
    Four weeks starting Oct 2.
    Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    New to programming? Learn the basics of programming using python in eight one-hour sessions over the course of four weeks. Sessions will consist of a mix of lectures and hands-on exercises.

    Participation counts towards the SciNet Scientific Computing Certificate.

    For sign up and more information, see https://courses.scinet.utoronto.ca/401

  • INTRO TO SCINET AND NIAGARA
    Wednesday Oct 10, 2018, 10:00 am – 11:30 am
    SciNet Boardroom (suite 1140, 661 University Avenue, Toronto).

    This is a class of approximately 90 minutes to introduce SciNet and the new supercomputer Niagara and teach you how to use Niagara.

    Participation counts towards the SciNet HPC Certificate.

    For more information and (free) registration, go to https://courses.scinet.utoronto.ca/405

  • SCINET USER GROUP MEETING
    Wednesday Oct 10, 2018, 12:00 noon – 1:00 pm
    SciNet Boardroom (suite 1140, 661 University Avenue, Toronto).

    The SciNet Users Group (SNUG) meetings are every month on the second Wednesday (except during the summer), and involve pizza, user
    discussion, feedback, and a half-hour talk (TBA) on topics or technologies of interest to the SciNet community. We’ll likely
    discuss the upcoming Resource Allocation Competition.

    For sign up and more information, see https://courses.scinet.utoronto.ca/411

  • ADVANCED SHELL PROGRAMMING
    Wednesday Oct 17, 2018, 1:00 pm – 4:00 pm
    Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    Learn how to write bash script, use environment variables, how to control process, and much more. Requires some linux basic command line experience.

    Participation counts towards the SciNet Scientific Computing Certificate.

    For sign up and more information, see https://courses.scinet.utoronto.ca/409

  • NUMERICAL COMPUTING WITH PYTHON
    Tuesdays and Thursdays, 1:00 pm – 2:00 pm
    Four weeks, starting Nov 6, 2017 (skipping the week of Nov 12-16)
    Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    Learn about research computing even with little programming experience. Covers programming in python, best practices and
    visualization. Some experience with python is required. Four home work sets will be the basic of the evaluation.

    Participation counts towards the SciNet Scientific Computing Certificate.

    For sign up and more information, see https://courses.scinet.utoronto.ca/402

  • INTRO TO SCINET AND NIAGARA
    Wednesday Nov 14, 2018, 10:00 am – 11:30 am
    SciNet Boardroom (suite 1140, 661 University Avenue, Toronto)

    This is a class of approximately 90 minutes to introduce SciNet and the new supercomputer Niagara and teach you how to use Niagara.

    Participation counts towards the SciNet HPC Certificate.

    For more information and (free) registration, go to https://courses.scinet.utoronto.ca/406

  • INTRODUCTION TO THE LINUX SHELL
    Wednesday Nov 21, 1:00 pm – 4:00 pm
    Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    Working with many of the HPC systems in Ontario involves using the Linux/UNIX command line. This provides a very powerful interface, but it can be quite daunting for the uninitiated. In this half-day session, you can become initiated with this course. This hands on session will cover basic commands and scripting. It could be a great boon for your productivity!

    Participation counts towards the SciNet Scientific Computing Certificate.

    For sign up and more information, see https://courses.scinet.utoronto.ca/407

  • SCIENTIFIC COMPUTING FOR PHYSICISTS (PHY1610)
    Winter 2019
    Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    This course is aimed at reducing your struggle in getting started with computational projects, and make you a more efficient computational scientist. Topics include well-established best practices for developing software as it applies to scientific computations, common numerical techniques and packages, and aspects of high performance computing. While we will introduce the C++ language, in one language or another, students should already have some programming experience. Despite the title, this course is
    suitable for many physical scientists (chemists, astronomers, …).

    This course is part of the physics graduate program. Students willing to take the course as part of their graduate program have to enroll through Acorn/ROSI.

    For more information, see https://courses.scinet.utoronto.ca/398

  • QUANTITATIVE APPLICATIONS FOR DATA ANALYSIS
    Winter 2019
    University of Toronto Scarborough Campus

    In this course data analysis techniques utilizing Python and R statistical language, will be discussed and introduced, as well as the basics of programming and scientific computing. The goal of this course is to prepare graduate students to perform scientific data analysis. Successful students will learn how to use statistical inference tools to gain insight into large and small data sets, as well as be exposed to cutting-edge techniques and best practises to store, manage and analyze (large) data.

    Topics include: Python and R programming, version control, automation, modular programming and scientific visualization.

    Students willing to take the course as part of their graduate program have to enroll through Acorn/ROSI. This course is part of the EES graduate program and to be taught at the UTSc campus.

    For more information, see https://courses.scinet.utoronto.ca/403

As always, further details can be found below on the SciNet courses siteand the SciNet wiki.

SciNet News June 2018

June 4, 2018 in for_users, newsletter

SUMMARY

  • Scratch purging policy on Niagara starts July 16.
  • Niagara takes it place at the number 53 spot in the top500.
  • Preliminary fall training and education schedule

Details can be found below and are also available on the SciNet education website courses.scinet.utoronto.ca and the SciNet wiki docs.scinet.utoronto.ca.

SYSTEM NEWS

  • Scratch purging policy on Niagara starts July 16.
  • Niagara takes it place at the number 53 spot in the top500.
  • Burst Buffer, a fast storage tier between the general parallel file system and system memory, is sto;; available for groups with high I/O needs, upon request. See https://docs.scinet.utoronto.ca/index.php/Burst_Buffer

EVENTS COMING UP

Unless stated otherwise, all events listed below take place at the SciNet Teaching Room at our offices on the eleventh floor of the MaRS West Tower, suite 1140A (661 University Avenue, Toronto ON M5G 1M1).

Most events will be recorded and some are broadcast, but only some of the courses can be taken remotely for SciNet certificate credits, as indicated below.

Registration for SciNet courses is done by logging into https://support.scinet.utoronto.ca/education with your SciNet account.

More SciNet courses will be announced at the end of the summer.

  • INTRO TO SCINET/NIAGARA Wednesday Sept 23, 2018, 10:00 am – 11:30 am SciNet Boardroom (suite 1140, 661 University Avenue, Toronto ON M5G 1M1).

    This is a class of approximately 60-90 minutes to introduce SciNet and the new supercomputer Niagara and teach you how to use Niagara.

    Participation counts towards the SciNet HPC Certificate.

    For more information and (free) registration, go to https://support.scinet.utoronto.ca/education/go.php/396/index.php

  • COMPUTE ONTARIO SUMMER SCHOOL EAST June 11 – 15, University of Toronto, St. George Campus New College, University of Toronto, 40 Willcocks St., Toronto, M5S 1C6

    The Compute Ontario Summer School on Scientific and High Performance Computing is an annual educational event for graduate/undergraduate students, postdocs and researchers to learn and share knowledge and experience in high performance and technical computing on modern HPC platforms.

    As in previous years, the Summer School on High Performance Computing 2014 will have three installments:

    “West” May 28 – June 1 Western University, London “Central” June 11 – June 15 University of Toronto, Toronto “East” July 30 – August 3 Queen’s University, Kingston

    Registration for the school in London and Kingston is at https://www.sharcnet.ca/summerschool/2018, while registration for the summer school in Toronto is at https://support.scinet.utoronto.ca/education/go.php/368/index.php

    The Toronto summer school hosted by SciNet will have the following three streams: Stream 1: High Performance computing; Stream 2: Data Science; Stream 3: Biomedical. Instructors are from SciNet, SHARCNET, and CAMH.

    This event will not be held at the SciNet Offices, rather it will be at New College on the Downtown St. George Campus of the University of Toronto.

    Parts of this event count towards the SciNet Certificates.

    More details and registration can be found at https://support.scinet.utoronto.ca/education/go.php/368/index.php

  • ADVANCED PARALLEL SCIENTIFIC COMPUTING Mondays and Wednesdays, 1 pm – 2pm Four weeks starting Sept 11. Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    In this course, we will be exploring advanced examples of parallel computing in scientific research. Topics include HPC tools, CUDA, molecular dynamics, Monte Carlo and random number generation, smoothed particle hydrodynamics, N-body simulations and computational fluid dynamics.

    The format of the course will be two lectures of one hour, for four consecutive weeks. The participants are expected to choose a project that involves analyzing and improving a parallel research code from one of the topic presented in the lectures, and to present their findings two weeks after the end of the course.

    The lectures of this course will be broadcast and recorded. Remote participation for credit may be possible upon request.

    Familiarity with parallel programming (MPI/OpenMP/CUDA) in a compiled language (C/C++/Fortran) is a prerequisite of this course.

    This class counts as 12 credit-hours towards the SciNet HPC Certificate. It can also be taken as a modular course by Physics grad students and as a mini course by Astrophysics students.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/317/index.php

  • INTRODUCTION TO CLINICAL BIOSTATISTICS (MSC1090) Tuesdays and Thursday, 11 am – 12 noon Twelve weeks starting Sept 12.

    In this course data analysis techniques utilizing the R statistical language, will be discussed and introduced, as well as, the basics of programming and scientific computing. The goal of this course is to prepare graduate students to perform scientific data analysis. Successful students will learn how to use statistical inference tools to gain insight into large and small data sets, as well as be exposed to cutting-edge techniques and best practises to store, manage and analyze (large) data. Topics include: R programming, version control, automation, modular programming and scientific visualization.

    Students willing to take the course as part of their graduate program have to enroll through Acorn/ROSI. This course is part of the IMS graduate program and to be taught at the UofT St. George campus (i.e., not in the SciNet classroom).

    Unfortunately, this course is full, but we are exploring the possibility of giving it again in the Winter term.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/324/index.php

  • INTRO TO SCINET Sept 13, 10:00 am – 11:30 am

    In about 90 minutes, you will learn how to use the SciNet systems. Experienced users may still pick up some valuable pointers.

    Participation counts towards the SciNet HPC Certificate.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/329/index.php

  • SCINET USER GROUP MEETING Sept 13, 12:00 noon – 1:00 pm Boardroom (MaRS West Tower, 661 University Ave, Toronto, suite 1140)

    Pizza, user discussion, and a tech talk: “ChIP-Seq analysis of the Interactive Bromodomain 1 protein (Ibd1) in Tetrahymena thermophila”, by Alejandro Saettone (Ryerson University)

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/301/index.php

  • INTRODUCTION TO NEURAL NETWORK PROGRAMMING Sept 25, 10:00 am – 3:00 pm Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    This workshop will introduce neural network programming concepts, theory, and techniques. The level of the material will be introductory, intended for those with no experience with neural networks. The programming language will be Python 2.7; experience with Python programming will be assumed. The Keras neural network framework, with a Theano back end, will be used for more-advanced programming; no experience with Keras or Theano will be expected. Students should come with the following Python packages installed on their laptops: numpy, matplotlib, scikit-learn, theano (version 0.9.0 or greater), keras.

    Participation counts towards the SciNet Data Science Certificate.

    You can also view this event’s broadcast, but this would not count towards the certificate.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/338/index.php

  • INTRODUCTION TO THE LINUX SHELL Sept 27, 10:00 am – 12:00 noon Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    Learn the basics of how to use the unix shell in two hours. Very useful for new users of SciNet that have little or no experience with unix or linux.

    Participation counts towards the SciNet Scientific Computing Certificate.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/330/index.php

  • ADVANCED NEURAL NETWORKS Mondays from 11:00 am to 12:00 noon Four weeks starting Oct 2. Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    This class will review advanced neural network programming theory and architectures. The level of the material will not be introductory, experience with neural networks will be assumed. This class is intended to continue the material covered in “Introduction to Neural Network Programming” (see above).

    Participation counts towards the SciNet Data Science Certificate.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/325/index.php

  • INTRODUCTION TO PROGRAMMING Tuesdays and Thursdays, 1:00 pm – 2:00 pm Four weeks starting Oct 10. Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    New to programming? Learn the basics of programming using python in eight one-hour sessions over the course of four weeks. Sessions will consist of a mix of lectures and hands-on exercises.

    Participation counts towards the SciNet Scientific Computing Certificate.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/328/index.php

  • ADVANCED SHELL PROGRAMMING Oct 25, 1:00 pm – 4:00 pm Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    Learn how to write bash script, use environment variables, how to control process, and much more. Requires some linux basic command line experience.

    Participation counts towards the SciNet Scientific Computing Certificate.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/331/index.php

  • INTRO TO SCIENTIFIC COMPUTING WITH PYTHON Tuesdays and Thursdays, 1:00 pm – 2:00 pm Four weeks, starting Nov 7, 2017 Teaching Room 1140A (MaRS West Tower, 661 University Ave, Toronto)

    Learn about research computing even with little programming experience. Covers programming in python, best practices and visualization. Some experience with python is required. Four home work sets will be the basic of the evaluation.

    Participation counts towards the SciNet Scientific Computing Certificate.

    The lectures of this course will be broadcast and recorded. Remote participation for credit may be possible upon request.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/327/index.php

  • SCIENTIFIC COMPUTING FOR PHYSICISTS (PHY1610) Winter 2018

    This course is aimed at reducing your struggle in getting started with computational projects, and make you a more efficient computational scientist. Topics include well-established best practices for developing software as it applies to scientific computations, common numerical techniques and packages, and aspects of high performance computing. While we will introduce the C++ language, in one language or another, students should already have some programming experience. Despite the title, this course is suitable for many physical scientists (chemists, astronomers, …).

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/326/index.php

SciNet News August 2016

August 29, 2016 in for_researchers, for_users, newsletter

The SciNet offices have moved to a new home! We are now located in the MaRS West Tower on the eleventh floor (address: 661 University Ave., Suite 1140 Toronto, ON M5G 1M1). As in our old spot, users, students, and friends are still welcome to visit our new abode, however, because we are still getting settled in, it is recommended to schedule a meeting (by emailing support AT scinet.utoronto.ca).

You will also find the announcement of SciNet course offerings in the 2016-2017 academic year. Users that take a sufficient number of courses can earn one of the SciNet certificates. In addition, many of our courses can now be taken for credit towards PhD and MSc programs by graduate students from several departments across the University of Toronto, including Institute of Medical Sciences, Ecology and Evolutionary Biology, Chemistry, Astrophysics, and Physics.

As always, details on SciNet events are available on the wiki and the SciNet education website.

SUMMARY

  • SciNet offices have moved.
  • SciNet’s Courses that can also be taken for graduate credit at the University of Toronto:
    • Advanced Parallel Scientific Computing
    • Introduction to Data Analysis with R
    • Introduction to Scientific Computing with Python
    • Scientific Computing for Physicists
  • SciNet’s Training Events: Intro to SciNet, Intro to Linux Shell, Storage and I/O in Large Scale Scientific Projects, SciNet User Group Meetings.
  • HPSS: Now also accessible via Globus.
  • GPC: New versions of ParaView and VisIt, and beagle for GPUs available as modules.
  • GPC: New web portal with records of your completed GPC jobs available for testing at https://my.scinet.utoronto.ca.
  • BGQ: Upgraded toolchain and operating system.
  • P8: We now have two Power 8 boxes with GPUs available for testing.

SCINET OFFICE MOVE

The SciNet offices have moved to a new home! We are now located in brand new offices on the eleventh floor of the new MaRS West Tower at the corner of College Street and University Avenue. Our new address is

661 University Ave
Suite 1140
Toronto, ON M5G 1M1

As was the case when we were in our old spot, users, students, and friends are still welcome to visit, however, because we are still getting settled in, we recommend that you first schedule a meeting by emailing support AT scinet.utoronto.ca.

Many of our courses, training sessions and user group meetings will be given in the new location, either in our new conference room or in our new classroom.

EVENTS COMING UP

Unless stated otherwise, all events take place at the SciNet Headquarters, now located on the eleventh floor of 661 University Ave., MaRS West Tower, suite 1140 (look for the bright orange reception area).

Most events will be recorded and some are broadcasted, but only some of the courses can be taken remotely for SciNet certificate credits, as indicated below. All events at SciNet are free but we ask that you enroll on the education website: https://support.scinet.utoronto.ca/education .

Courses that can be taken for credit at the UoT

  • ADVANCED PARALLEL COMPUTING
    8 Lectures, Tuesdays and Thursdays starting Sept 20, 11:00 am – 12:00 pm

    Explore advanced use case examples of parallel computing in scientific research.

    The course can be taken as a mini/modular graduate course by Physics, Astronomy, and Chemistry students.

    This course counts toward the High Performance Computing Certificate.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/263/index.php

  • INTRODUCTION TO DATA ANALYSIS WITH R
    12 Lectures, Tuesdays and Thursdays starting Oct 11, 12:00 pm – 1:00 pm

    The goal of this course is to prepare students to perform scientific data analysis. Successful students will learn how to use statistical inference tools to gain insight into large and small data sets, as well as be exposed to cutting-edge techniques and best practices to store, manage and analyze (large) data.

    For students in the Institute of Medical Science, this course can be taken as “Seminars in Translational Research” (MSC1010Y-1011Y)

    Graduate students from the Department of Ecology & Evolutionary Biology (EEB) at UofT can take this course for graduate credit as well. Interested students from the EEB department should contact Prof. Helen Rodd in advance.

    Physics, Astrophysics and Chemistry students can take this course as part of the mini/modular graduate courses.

    This course also counts toward the SciNet’s Data Science Certificate.

    Location: St.George Campus, specific room TBD.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/262/index.php

  • SCIENTIFIC COMPUTING WITH PYTHON
    8 Lectures, Tuesdays and Thursdays starting Nov 15, 11:00 am – 12:00 pm

    Learn about research computing even with little programming experience. Covers basics of programming in Python, best practices and visualization. The course will last 4 weeks with 2 lectures per week.

    The course can be taken as a mini/modular graduate course by Physics, Astronomy and Chemistry students. This course can also be taken by students for graduate credits from the Department of Ecology & Evolutionary Biology (EEB) at the UofT. Interested students from the EEB department should contact Prof. Helen Rodd in advance.

    This course counts toward the Scientific Computing Certificate and (partially) toward the High Performance Computing Certificate.

    Although in-person attendance is highly preferred, SciNet users can take this course remotely by following the lectures online and submitting the assignments.

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/264/index.php

  • SCIENTIFIC COMPUTING FOR PHYSICISTS
    Physics graduate course (PHY1610)

    Winter 2017

    This course is aimed at reducing your struggle in getting started with computational projects, and make you a more efficient computational scientist. Topics include well-established best practices for developing software as it applies to scientific computations, common numerical techniques and packages, and aspects of high performance computing. While we will introduce the C++ language, in one language or another, students should already have some programming experience. Despite the title, this course is suitable for many physical scientists (chemists, astronomers, …).

    This course can be taken for credit by Physics graduate students, and by other graduate students that are allowed to take physics courses, however, they are strongly encourage to check this with their graduate coordinators first.

    Part of this course may be given as mini-courses as well. Details on this will follow later.

    This course counts toward the Scientific Computing Certificate and the High Performance Computing Certificate.

    Although in-person attendance is highly preferred, SciNet users can take this course remotely by following the lectures online and submitting the assignments.

SciNet Training Events

  • INTRO TO SCINET
    Wed 14th Sept 10:00 am – 11:30 am

    The “Intro to SciNet” is a class of approximately 60-90 minutes where you will learn how to use the systems. Experienced users may still pick up some valuable pointers during these sessions.

    Participation counts towards the SciNet HPC Certificate.

    If you are interested to participate remotely using Google Hangout, please send an email with your gmail address to which we can send the invitation.

    Location: SciNet conference room (MaRS West Tower, 11th floor)

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/267/index.php

    Future “Intro to SciNet” dates and enrollment links:
    Oct 12, 10:00 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/269/index.php
    Nov 9, 10:00 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/268/index.php
    Dec 14, 10:00 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/270/index.php
    Jan 11, 10:00 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/296/index.php
    Mar 8, 10:00 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/297/index.php
    May 10, 10:00 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/302/index.php

  • SCINET USER GROUP MEETING
    Wed 14th Sept 12:00 pm – 1:00 pm

    The SciNet Users Group (SNUG) meetings are every month on the second Wednesday, and involve pizza, user discussion, feedback, and a half-hour talk on topics or technologies of interest to the SciNet community.

    Location: SciNet conference room (MaRS West Tower, 11th floor)

    For sign up, see https://support.scinet.utoronto.ca/education/go.php/291/index.php

    Future “SNUG” dates and enrollment links:
    Oct 12, 12:00 noon – 1:00 pm: https://support.scinet.utoronto.ca/education/go.php/292/index.php
    Nov 9, 12:00 noon – 1:00 pm: https://support.scinet.utoronto.ca/education/go.php/293/index.php
    Dec 14, 12:00 noon – 1:00 pm: https://support.scinet.utoronto.ca/education/go.php/294/index.php
    Jan 11, 12:00 noon – 1:00 pm: https://support.scinet.utoronto.ca/education/go.php/295/index.php
    Feb 8, 12:00 noon – 1:00 pm: https://support.scinet.utoronto.ca/education/go.php/298/index.php
    Mar 8, 12:00 noon – 1:00 pm: https://support.scinet.utoronto.ca/education/go.php/299/index.php
    Apr 12, 12:00 noon – 1:00 pm: https://support.scinet.utoronto.ca/education/go.php/300/index.php
    May 10, 12:00 noon – 1:00 pm: https://support.scinet.utoronto.ca/education/go.php/301/index.php

  • STORAGE AND I/O IN LARGE SCALE SCIENTIFIC PROJECTS
    Wed 21th Sept 09:30 am – 5:00 pm

    Location: SciNet classroom (MaRS West Tower, 11th floor)

    Learn how to pinpoint and alleviate bottlenecks in large data-driven research projects. Techniques such as tar, compression, ramdisk, file format options, and job scheduling techniques will be covered.

    For sign up, see https://support.scinet.utoronto.ca/education/go.php/265/index.php

  • INTRODUCTION TO THE LINUX SHELL
    Wed 19th Oct 10:00 am – 1:00 pm

    The Linux shell (command line, or however you want to phrase it) is the most basic Linux interface that there is. If you don’t know what the Linux shell is then you need to take this course!

    In this course, you will learn the basics of how to use the Unix shell in three hours. Includes a mild introduction to bash scripting as well. This course counts toward the Scientific Computing Certificate.

    Location: SciNet classroom (MaRS West Tower, 11th floor)

    For sign up and more information, see https://support.scinet.utoronto.ca/education/go.php/266/index.php

SYSTEM CHANGES

Note that most software changes for the GPC also hold for Sandy and Gravity.

  • HPSS: If you have an HPSS account, this is now also accessible from the Globus file transfer web-interface. Globus is accessible through https://globus.computecanada.ca, where you’d login with your Compute Canada account, while the name of the HPSS endpoint is “Compute Canada HPSS”, for which you separately authenticate with your SciNet account. For more information on using Globus, see https://docs.computecanada.ca/wiki/Globus .
  • GPC: New web portal with records of your completed GPC jobs available for testing at https://my.scinet.utoronto.ca.
  • GPC: Beagle GPU-version available for testing under the use.experimental module.
  • GPC: Python distribution Anaconda v4.x installed as a module
  • GPC: VisIt version 2.10.2 installed as a module
  • GPC: ParaView version 5.1.0 installed as a module
  • GPC: Versions 15.0.6 and 16.0.3 of the Intel Compilers are installed as modules.
  • GPC: Version 6.0 of Allinea Forge (containing the DDT parallel debugger, the profiler MAP and Performance Reports) installed as a module.
  • GPC: Gromacs v. 5.1.1 installed as modules (GPU and non-GPU).
  • GPC: Chemistry package Cantera installed as a module.
  • BGQ: Upgraded compute nodes and toolchain to V1R2M4 from V1R2M2. Note that this means that when recompiling applications, one has to start from scratch (i.e., no old object files).
  • BGQ: Front end nodes operating system was updated to RedHat 6.7. This should not change much for users.
  • P8: A new P8 system with two Power-8 machines, each with two NVIDIA Tesla K80 GPUs, is available for testing. For details, see wiki.scinet.utoronto.ca/wiki/index.php/P8

WHAT ELSE HAPPENED AT SCINET IN THE SUMMER OF 2016?

  • April: Participation of SciNet in Science Rendezvous 2016.
  • May: Visualization tutorial at 2016 “Chemical BioPhysics Symposium” and Best Practices at GLBIO 2016.
  • June: Participation of SciNet’s personnel at Canheit-HPCS 2016 (Edmonton, Alberta), and ISC (Frankfurt, Germany).
  • June: Participation of SciNet’s personnel at the International HPC Summer School (Ljubljana, Slovenia)
  • June: Intro to SciNet
  • June: SciNet User Group Meeting
  • June 21-26: International Summer School on HPC Challenges
  • May/August: Participation of SciNet’s personnel at the other instance of Ontario HPC Summer School 2016 (Hamilton and Ottawa)
  • July 11-15: Ontario HPC Summer School 2016 – Central
  • August 14-19: As part of the “2016 Industrial Problem Solving Workshop” to be hosted at the Fields Institute, SciNet will provide an introduction to High Performance Computing resources, as well as support for participants who decide to tackle this problems numerical utilizing supercomputer resources.

SciNet News April 2015

April 9, 2015 in for_researchers, for_users, newsletter

First of all, we want to take this opportunity to thank all of you who cleaned up their scratch directories last week when /scratch it was 99% full. You saved the file system!

Below you will find the salient points of what else happened or changed recently and what will happen in the near future. Among these is the announcement of the Ontario High Performance Computing Summerschool, which has three installments, one of which is to be held in Toronto in July. As always, details on SciNet events are available on the wiki and the SciNet education website.

 

UPCOMING EVENTS

Unless stated otherwise, all events take place at the SciNet Conference Room on the second floor of 256 McCaul Street, Toronto. All events at SciNet are free for users but we ask that you enroll on the education website: https://support.scinet.utoronto.ca/education.

  • INTRO TO SCINET
    Wednesday April 8, 10:30 am – 11:30 am

    The “Intro to SciNet” is a class of approximately 90 minutes where you will learn how to use the systems. Experienced users may still pick up some valuable pointers during these sessions.

    Participation counts towards the SciNet HPC Certificate.

    For more information and enrollment, go to https://support.scinet.utoronto.ca/education/go.php/172/index.php

    More intro sessions to be held in the near future: May 13, 10:30 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/173/index.php

    Jun 10, 10:30 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/174/index.php

  • SCINET USER GROUP MEETING
    Wednesday April 8, 12:00 noon – 1:00 pm

    The SciNet Users Group (SNUG) meetings are every month on the second Wednesday, and involve pizza, user discussion, feedback, and a half-hour talk on topics or technologies of interest to the SciNet community.

    TechTalk: “Smoothed Particle Hydrodynamics” by Marcelo Ponce (SciNet)

    In this talk I’m planning to review the main concepts used in “Smoothed Particle Hydrodynamics”, a technique widely used for simulating fluid dynamics. In particular I will focus in the basic concepts and their applications to astrophysical simulations, such as, the effects on accretion disks surrounding a gravitational recoiled black hole, and “mergerburst” events where the interaction between stars may trigger luminosity outburst (suhc as the outburst of V838 Monocerotis observed some years ago).

    For more information and enrollment, go to https://support.scinet.utoronto.ca/education/go.php/184/index.php

    Future SNUGs: May 13, 2015, 12:00 noon – 1:00 pm, TechTalk TBD
    Jun 10, 2015, 12:00 noon – 1:00 pm, TechTalk TBD

  • INTRODUCTION TO CUDA
    Tuesday April 14, 9:00 am – 5:00 pm

    The goal of this course is for incoming students, new to GPGPU but familiar with scientific programming in C or C++, to leave being able to start writing simple kernels for their own problems, and understand the tools, techniques and libraries that will be needed to improve and optimize the results.

    Participation counts towards the SciNet HPC Certificate.

    For more information (soon) and enrollment, go to https://support.scinet.utoronto.ca/education/go.php/179/index.php

  • RELATIONAL DATABASE BASICS
    Wednesday May 6, 9:30 – 5:00 pm

    As a form to manage data, relational databases are widely used in the back-end of many websites as well as for data mining computations. In this course, we will look at the basics of relational databases: What are they, and what are they useful for? We will also consider why they are not all that much used in the context of scientific computing, using SQLite and Python.

    Participation counts towards the SciNet Scientific Computing Certificate.

    For more information (soon) and enrollment, go to https://support.scinet.utoronto.ca/education/go.php/186/index.php

  • XEON PHI DEVELOPER TRAINING
    Tuesday May 19 and Wednesday May 20, 9:30 am – 4:30 pm

    On May 19 and 20, 2015, Intel is giving a two-day workshop in Toronto for software developers on the foundation needed for modernizing their code to take advantage of parallel architectures found in both the Intel Xeon processor and the Intel Xeon Phi co-processor.

    Note that a substantial difference with the one-day workshop on the same topic given in Toronto on October 27, 2014, is the second day, which consists of a hands-on lab in which you get to use the material and techniques presented in the first day.

    This is an external event organized by Intel. For more information and registration please go to

    Day 1: http://events.constantcontact.com/register/event?llr=kpiwi7pab&oeidk=a07eapw56wscf327149

    Day 2: http://events.constantcontact.com/register/event?llr=kpiwi7pab&oeidk=a07eapwcqo3f9b22629

  • COMPUTE ONTARIO RESEARCH DAY
    Thursday May 21 Conestoga College Institute of Technology and Advanced Learning Cambridge, Ontario

    A day filled with high performance computing related research done in Ontario. Have an interesting research story for which you used high performance computing (such as the facilities at SharcNet, SciNet, and HPCVL)? Want to share you experience with other Ontario HPC users? Consider giving a talk at the meeting.

    More details and registration at https://www.sharcnet.ca/events/CORD2015

  • ONTARIO HPC SUMMER SCHOOL 2015

    The Compute Ontario Summer School on High Performance Computing provides attendees with opportunities to learn and share knowledge and experience in high performance and technical computing.

    As in previous years, the Summer School on High Performance Computing 2015 will have three installments:

    “West” May 25 – 29 Western U. (London) “Central” July 13 – 17 U. Toronto (Toronto) “East” July 27 – 31 Queen’s U. (Kingston)

    The format will be a five day workshop with mixed lectures and hands-on sessions on a number of selected subjects such as MPI, CUDA, OpenMP, Python, and Visualization (the subjects will vary slightly depending on the location). These sessions will be organized in two parallel streams.

    Parts of this event may count towards the SciNet HPC Certificate.

    More details regarding the summer school program, location, and registration (for either of the three installments), will become available at https://www.sharcnet.ca/events/ss2015

  • HPCS 2015: HIGH PERFORMANCE COMPUTING SYMPOSIUM IN MONTREAL
    Montreal, June 15-19

    Registration is now open for the 29th annual HPCS conference (Canada’s foremost supercomputing conference). This conference will take place in Montreal, June 15-19, the first two days of which are tutorials on advanced research computing and high performance computing. This year’s theme is `Advanced Computing and Big Data – Driving Competitiveness and Discovery’.

    More information and registration can be found at http://www.hpcs.ca.

  • INTERNATIONAL SUMMER SCHOOL ON HPC CHALLENGES IN COMPUTATIONAL SCIENCES
    Sunday June 21 – Friday June 26 Toronto, Canada

    The application deadline for this event has passed. For more information see https://ihpcss2015.computecanada.ca.

 

SYSTEM NEWS

Note that most software changes for the GPC also hold for ARC, Sandy and Gravity.

  • BGQ: software stack upgraded to version V1R2M2.
  • BGQ: module hdf5/1812-v18-mpich2-gcc is deprecated; please use hdf5/1814-v18-mpich2-gcc instead.
  • BGQ: module fftw/3.3.3-gcc4.8.1 is deprecated; please use fftw/3.3.4-gcc4.8.1 instead.
  • GPC: Allinea Forge (DDT & MAP) available as module ddt/5.0.This includes Allinea Performance Reports.
  • GPC: OpenBLAS 0.2.13 installed as modules.
  • GPC and BGQ: Namd 2.10 installed as modules.
  • GPC: Gromacs 4.6.7 installed as a module.
  • GPC: Gromacs 5.0.4 installed as an experimental module (“module load use.experimental gromacs/5.0.4”).
  • GPC: Stacks/1.29 installed as a module.

WHAT’S NEW ONLINE?

The SciNet Wiki contains technical information on how to use the systems and can be found at https://support.scinet.utoronto.ca/wiki. The SciNet Education and Training site can be found at https://support.scinet.utoronto.ca/education and contains the material of SciNet’s training and education activities.

  • Wiki: Documentation of SciNet command line utilities to help with storage management, jobs and queues, and modules.
  • Wiki: PDF of the January SNUG Techtalk about “Data Management and Transfer Tools at SciNet”
  • Education: pdfs and recordings of the lectures of the “Scientific Computing” mini-courses.
  • Education: pdf of the TechTalk on “Job and Queue Management at SciNet”.

SciNet News February 2015

February 26, 2015 in for_researchers, for_users, newsletter

ANNOUNCEMENTS

Dear SciNet users,

To keep users posted on what’s going on at SciNet, we send monthly emails with the new features, planned courses, events and accomplishments at SciNet.

Last month, the SciNet team was augmented with another analyst, Mike Nolta. Welcome, Mike!

Below you will find the salient points of what else happened or changed recently and what will happen in the near future. Among these is the announcement and call for participant applications for the International High Performance Computing Summer School, to be held in Toronto. As always, details on SciNet events are available on the wiki and the SciNet education website.

– The SciNet team

 

UPCOMING EVENTS

Unless stated otherwise, all events take place at the SciNet Conference Room on the second floor of 256 McCaul Street, Toronto. All events at SciNet are free for users but we ask that you enroll on the education website: https://support.scinet.utoronto.ca/education .

 

  • INTRO TO SCINET
    Wednesday March 11, 10:30 am – 11:30 am

The “Intro to SciNet” is a class of approximately 90 minutes where you will learn how to use the systems. Experienced users may still pick up some valuable pointers during these sessions.

Participation counts towards the SciNet HPC Certificate.

For more information and enrollment, go to https://support.scinet.utoronto.ca/education/go.php/171/index.php

Future “Intro to SciNet” dates and enrollment links:

Apr 8, 10:30 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/172/index.php

May 13, 10:30 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/173/index.php

Jun 10, 10:30 am – 11:30 am: https://support.scinet.utoronto.ca/education/go.php/174/index.php

 

  • SCINET USER GROUP MEETING
    Wednesday March 11, 12:00 noon – 1:00 pm

The SciNet Users Group (SNUG) meetings are every month on the second Wednesday, and involve pizza, user discussion, feedback, and a half-hour talk on topics or technologies of interest to the SciNet community.

TechTalk: “Job and Queue Management Tools” by Mike Nolta (SciNet)

For more information and enrollment, go to https://support.scinet.utoronto.ca/education/go.php/183/index.php

Future SNUGs:
Apr 8, 2015, 12:00 noon – 1:00 pm, TechTalk TBD

May 13, 2015, 12:00 noon – 1:00 pm, TechTalk TBD

Jun 10, 2015, 12:00 noon – 1:00 pm, TechTalk TBD

 

  • SCIENTIFIC COMPUTING COURSE: HIGH PERFORMANCE COMPUTING
    March/April 2015

Part 3 of the Scientific Computing Course (part 1 was given in January and part 2 is in progress). Each part consists of eight lectures of one hour.

Note that these parts can be taken as “mini-courses” by astrophysics graduate students and as “modular courses” by physics and chemistry graduate students.

Participation in part 3 counts towards the SciNet HPC Certificate.

For more information (soon) and enrollment, go to https://support.scinet.utoronto.ca/education/go.php/178/index.php

Enrollment for part 1 and 2 is closed.

 

  • DEADLINE APPLICATION INTERNATIONAL SUMMER SCHOOL ON HPC IN TORONTO
    Wednesday March 11, 2015

See below for information regarding this summer school.

  • INTRODUCTION TO CUDA
    Tuesday April 14, 9:00 am – 5:00 pm

The goal of this course is for incoming students, new to GPGPU but familiar with scientific programming in C or C++, to leave being able to start writing simple kernels for their own problems, and understand the tools, techniques and libraries that will be needed to improve and optimize the results.

Participation counts towards the SciNet HPC Certificate.

For more information (soon) and enrollment, go to https://support.scinet.utoronto.ca/education/go.php/179/index.php

 

  • RELATIONAL DATABASE BASICS
    Wednesday May 6, 9:30 – 5:00 pm

As a form to manage data, relational databases are widely used in the back-end of many websites as well as for data mining computations. In this course, we will look at the basics of relational databases: What are they, and what are they useful for? We will also consider why they are not all that much used in the context of scientific computing, using SQLite and Python.

Participation counts towards the SciNet Scientific Computing Certificate.

For more information (soon) and enrollment, go to https://support.scinet.utoronto.ca/education/go.php/186/index.php

 

  • ONTARIO HPC SUMMER SCHOOL CENTRAL 2015
    Dates: TDB

Details to be posted soon.

 

  • COMPUTE ONTARIO RESEARCH DAY
    Thursday May 21, Conestoga College Institute of Technology and Advanced Learning, Cambridge, Ontario

A day filled with high performance computing related research done in Ontario. Have an interesting research story for which you used high performance computing (such as the facilities at SharcNet, SciNet, and HPCVL)? Want to share you experience with other Ontario HPC users? Consider giving a talk at the meeting.

More details to follow.

 

  • HPCS 2015: HIGH PERFORMANCE COMPUTING SYMPOSIUM IN MONTREAL
    Montreal, June 15-19

Registration is now open for the 29th annual HPCS conference (Canada’s foremost supercomputing conference). This conference will take place in Montreal, June 15-19, the first two days of which are tutorials on advanced research computing and high performance computing. This year’s theme is `Advanced Computing and Big Data – Driving Competitiveness and Discovery’.

More information and registration can be found at http://www.hpcs.ca .

 

  • INTERNATIONAL SUMMER SCHOOL ON HPC CHALLENGES IN COMPUTATIONAL SCIENCES
    Sunday June 21 – Friday June 26, Toronto, Canada

In this summer school, leading American, Canadian, European and Japanese computational scientists and HPC technologists will offer instruction on a variety of topics, including HPC challenges by discipline (e.g., bioinformatics, computer science, chemistry, and physics), HPC programming proficiencies, performance analysis & profiling, algorithmic approaches & numerical libraries, data-intensive computing and scientific visualization.

Compute Canada/Calcul Canada invites students enrolled in science and engineering fields, and researchers in those fields, at Canadian post-secondary institutions to apply for one of 10 spots allocated to Canada.

The application deadline is March 11, 2015.

For more information see https://ihpcss2015.computecanada.ca

 

SYSTEM NEWS

  • BGQ now a single 4-rack system. bgqdev-fen1 is the single login/devel/submission node.
  • GPC: Cmake 3.1.0 available as a module.
  • GPC: kernel upgraded to 2.6.32-504.3.3.el6.x86_64. Its base OS remains unchanged.
  • GPC: emacs 24.4 available as a module.
  • GPC: intel/15.0 module is deprecated — use 15.0.1 instead.
  • GPC, BGQ: git/1.9.5 is available as a module.
  • GPC: Nwchem 6.5 installed as a moduled.

 

WHAT’S NEW ONLINE?

The SciNet Wiki can be found at http://wiki.scinethpc.ca and contains technical information on how to use the systems. The SciNet Education and Training site can be found at https://support.scinet.utoronto.ca/education and contains the material of SciNet’s training and education material.

  • Wiki: PDF of the January SNUG Techtalk about “Data Management and Transfer Tools at SciNet” Wiki: How to run parallel R.
  • Wiki: BGQ page updated to reflect its new 4-rack configuration.
  • Education: recordings of the first lectures on in the “Scientific
  • Computing” and “Numerical Tools” mini-course.

 

WHAT HAPPENED AT SCINET IN THE LAST MONTH?

  • Jan 14 and Feb 11: Intro to SciNet
  • Jan 21: Introduction to the Linux Shell
  • Jan: Eight lectures of the mini-course “Scientific Computing”
  • Feb: Four lectures of the mini-course “Numerical Tools for Physical Sciences”
  • Feb 11: SciNet User Group Meeting with TechTalk on “Data Management and Transfer Tools”
  • Feb 13: SciNet analysts participated as judges in the final of the Big Data Challenge for High School Students.

 

 

 

SciNet News April 2014

April 21, 2014 in for_researchers, for_users, newsletter

ANNOUNCEMENTS

We are pleased to announce that users can now keep track of their progress towards a SciNet certificate through their account on the Courses website.

UPCOMING EVENTS

Unless stated otherwise, all events take place in the SciNet Boardroom, 2nd floor of 256 McCaul Street, Toronto. Events are free for users but registration is required: https://support.scinet.utoronto.ca/education.

In a Nutshell

SCINET USER GROUP (SNUG) MEETING- Wednesday April 9, 12:00 noon – 1:00 pm

INTRO TO SCINET- Wednesday May 14, 10:30 am – 11:30 am

SCINET USER GROUP (SNUG) MEETING- Wednesday May 14, 12:00 noon – 1:00 pm

5TH INTERNATIONAL SUMMER SCHOOL ON HPC CHALLENGES IN COMPUTATIONAL
SCIENCES- Sunday June 1 – Friday June 6 (Budapest, Hungary)

COMPUTE ONTARIO SUMMER SCHOOL ON HIGH PERFORMANCE COMPUTING/CENTRAL- Monday June 9 – Friday June 13 (University of Toronto, St. George Campus)

SCINET USER GROUP (SNUG) MEETING – Wednesday Jun 18, 12:00 noon – 1:00 pm

HPCS 2014: HIGH PERFORMANCE COMPUTING SYMPOSIUM IN HALIFAX – Monday June 23 – Friday June 27

Details

  • SCINET USER GROUP (SNUG) MEETING
    Wednesday April 9, 12:00 noon – 1:00 pmTechTalk:
    Keep inventory of your data on different file systems with ISH
    by Ramses van Zon

    Want to be able to browse remote data on SciNet and HPPS as if you’re in a linux shell? With ‘ish’ , an inventory shell, you can use ls, cd, find, etc., to access and browse metadata of remote files and tar-balls. It allows you keep and peruse an inventory of your files stored in different locations.

    For more information and enrollment, go to the course website.

  • INTRO TO SCINET
    Wednesday May 14, 10:30 am – 11:30 amA class of approximately 60 minutes where you will learn how to use the SciNet systems. Experienced users may still pick up some valuable pointers during these sessions.

    For more information and enrollment, go to the course website.

  • SCINET USER GROUP (SNUG) MEETING
    Wednesday May 14, 12:00 noon – 1:00 pmTechTalk:
    Getting Started with Cilk Plus
    by Chris Szalwinski

    Intel’s Cilk Plus augments the C and C++ languages with three keywords for parallel programming on multi-core systems. These keywords inform the compiler that pertinent code may be parallelized. This talk will include code snippets showing how to implement map-reduce and fork-join patterns and will compare run times with those of serial code.

    For more information and enrollment, go to the course website.

  • 5TH INTERNATIONAL SUMMER SCHOOL ON HPC CHALLENGES IN COMPUTATIONAL SCIENCES
    Sunday June 1 – Friday June 6, Budapest, HungaryIn this summer school, leading American, Canadian, European and Japanese computational scientists and HPC technologists will offer instruction on a variety of topics, including HPC challenges by discipline (e.g., bioinformatics, computer science, chemistry, and physics), HPC programming proficiencies, performance analysis and profiling, algorithmic approaches & numerical libraries, data-intensive computing and scientific visualization.

    Compute Canada/Calcul Canada invites students enrolled in science and engineering fields at Canadian post-secondary institutions to apply for one of 10 (expenses-paid) spots allocated to Canada.

    The deadline for application has passed.

    For more information see http://www.prace-ri.eu/International-Summer-School-2014.

  • COMPUTE ONTARIO SUMMER SCHOOL ON HIGH PERFORMANCE COMPUTING/CENTRAL
    Monday June 9 – Friday June 13, University of Toronto, St. George CampusThe Compute Ontario Summer School on High Performance Computing provides attendees with opportunities to learn and share knowledge and experience in high performance and technical computing.

    As in previous years, the Summer School on High Performance Computing 2014 will have three installments:

    “West” May 26 – 30 Waterloo

    “Central” June 9 – 13 Toronto

    “East” July (TBD) Ottawa

    The format will be a four day workshop with mixed lectures and hands-on sessions on a number of selected subjects, such as MPI, CUDA, OpenMP, and Visualization (these will vary slightly depending on the location).

    This event will *not* be held at the SciNet Headquarters.

    Parts of this event may count towards the SciNet HPC Certificate.

    More details regarding the summer school will be announced here

  • SCINET USER GROUP (SNUG) MEETING
    Wednesday Jun 18, 12:00 noon – 1:00 pmTechTalk: TBA

    For more information and enrollment, go to
    the course website.

  • HPCS 2014: HIGH PERFORMANCE COMPUTING SYMPOSIUM IN HALIFAX
    Monday June 23 – Friday June 27The 28th annual HPCS conference, will take place in Halifax, June 3-27 (of which the first two days are tutorials). This year’s theme is `HPC Serving Society’. Sessions will focus on Bioinformatics, Big Data & Analytics, Computational Materials Science, Ocean & Atmosphere Modelling, Computational Chemistry, Data Cave Visualization, and more.

    We are seeking submissions for contributed works and posters relating to the conference theme. Submissions for contributed works will be due on April 30.

    More information is available at the conference website, http://2014.hpcs.ca.

System News

  • GPC: Ffmpeg v2.1.3, an audio and video software solution, installed as a module.
  • GPC: Python 3.3.4 installed as a module.
  • P7: Vapor 2.3.0, The Visualization and Analysis Platform for Ocean, Atmosphere, and Solar Researchers, installed as a module.
  • BGQ: Hpn-ssh, high-performance enabled ssh, installed as a module.
  • GPC: Petsc 3.4.4,the Portable, Extensible Toolkit for Scientific computation, installed as a module.
  • BGQ: OpenFOAM 2.3.0 installed as a module.
  • GPC: Quake 0.3.5, a program to detect and correct errors in DNA sequencing reads, installed as a module.
  • TCS: Cdo 1.6.1, a collection of command line Operators to manipulate and analyse climate and NWP model data, installed as a module.

What’s New On The SciNet Websites?

The SciNet Wiki can be found at http://wiki.scinethpc.ca, and the SciNet Education and Training site can be found at https://support.scinet.utoronto.ca/education.

  • Education site: Users can now keep track of their progress towards the SciNet certificates as they are take the courses. See the ‘Certificates’ tab on the education web site. For clarity, courses that count toward the certificates now have a course number.
  • Wiki and Education: Slides of the Seminar on Intro to GPU Computing Using CUDA.
  • Wiki and Education: Slides and Recordings of the lectures of the “High Performance Scientific Computing” mini-course.
  • Wiki and Education: Slides of the TechTalk “NetCDF4 binary files with Python, C++ and R”

What Else Happened at SciNet in the Last Month?

  • Mar 6: SciNet developer seminar on “OpenCL Middleware” by AJ Guillon
  • Mar 12: SciNet User Group meeting with TechTalk on “NetCDF4 binary files with Python, C++ and R” by Bertrand Brelier.
  • Mar 12: Introduction to GPU computing using CUDA
  • Mar 11-Apr 3: Lectures on “High Performance Scientific Computing” (Part 3 of the SciNet Scientific Computing Course)

SciNet News March 2014

March 12, 2014 in for_researchers, for_users, newsletter

To keep users posted on what’s going on at SciNet, we send monthly emails with the new features, planned courses and events and accomplishments at SciNet.

Below you will find the salient points of what has happened or changed during the last month and what will happen in the near future. Among these is the announcement of the dates of the “Compute Ontario Summer School on HPC”. There will an installment in Waterloo in May, one in Toronto in June and one in Ottawa in July.

As always, details are available on the wiki or the SciNet education website.

– The SciNet team

Events Coming Up

Unless stated otherwise, all events take place at the SciNet Headquarters, Rm 235 of 256 McCaul Street, Toronto. All events below are free for users but we ask that you sign up (“enroll”) on the education website.

  • Thursday March 6, 2:00 pm – 3:00 pm

    SCINET DEVELOPER SEMINAR

    OPENCL MIDDLEWARE

    By AJ Guillon

    AJ will demonstrate new C++11 OpenCL middleware that enables professional software development for heterogeneous computing. This framework significantly reduces programming complexity and extends OpenCL with new features. AJ will also provide an overview of the internal design of the middleware and supporting software stack.

    For more information and enrollment, go to the course website.

  • Sunday March 9, 2014

    APPLICATION DEADLINE INTERNATIONAL SUMMER SCHOOL ON HPC

    See below for more information about this expenses-paid summer school on challenges in HPC, held in June in Budapest, Hungary.

  • Wednesday March 12, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    The SciNet Users Group (SNUG) meetings are every month on the second Wednesday, and involve (free) pizza, user discussion, feedback, and a one or two short talks on topics or technologies of interest to the SciNet community.

    This time, we will have

    • TechTalk:

      Introduction to NetCDF4 binary files with Python, C++ and R

      Bertrand Brelier (IBM@SciNet)

      During the presentation, Bertrand will review some features of NetCDF4, which is a portable binary file format and corresponding library. He will then show an example how to read and write a NetCDF4 compound variable in Python, C++ and R.

    • User discussion
    • Pizza!

    For more information and enrollment, go to the course website.

  • Wednesday March 12, 14:30 pm – 15:30 pm

    INTRODUCTION TO GPU COMPUTING USING CUDA

    Scott Northrup (SciNet)

    The CUDA (Compute Unified Device Architecture) parallel programming model harnesses the power of Nvidia graphics processing units (GPUs) to enable dramatic increases in computing performance for graphics, 3D content, video and scientific processing-intensive applications. This talk will provide an introduction to CUDA and examples of applications within research. Participants should be familiar with C/C++ development.

    For more information and enrollment, go to the course website.

    This talk is also offered remotely, for details see: https://www.westgrid.ca/events/introduction-to-gpu-computing-using-cuda

  • March 11, 13, 18, 20, 25, 27, and April 1 and 3: 11am-12noon

    HIGH PERFORMANCE SCIENTIFIC COMPUTING

    Lectures of Part 3 of the Scientific Computing Course

    Topics: parallel programming concepts, shared memory parallel programming with openmp, distributed memory parallel programming with mpi, hybrid programming.

    Participation counts towards the SciNet HPC Certificate.

    For more information and enrollment, go to the course website.

  • Tuesdays, almost bi-weekly, 2:30 pm – 3:30 pm

    COAST-TO-COAST SEMINAR SERIES

    The Coast to Coast Seminar is an hour-long presentation given on a scientific topic and is made accessible to audiences at a number of remote sites across Canada through collaboration technology.

    SciNet is a local seminar location for this series.

    The topic of the Spring 2014 Coast to Coast seminar series is “Technology for Aging Well” and it is built around a pan-Canadian project titled AGE-WELL (Aging Gracefully across Environments using Technology to Support Wellness, Engagement and Long Life.)

    The focus of the series will be a discussion how to use technology in helping support the Canadian aging population to ensure that all Canadians can grow older with dignity and grace.

    Mar 11: Geoff Fernie (U.Toronto)

    Mar 25: Alex Mihailidis (U.Toronto) and Rafik A. Goubran (Carleton U)

    For more information see http://www.irmacs.sfu.ca/events/coast-coast-seminars or https://www.westgrid.ca/events/coast-to-coast-seminar-series

  • Wednesday April 9, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • Wednesday May 14, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • Sunday June 1 – Friday June 6, Budapest, Hungary

    5TH INTERNATIONAL SUMMER SCHOOL ON HPC CHALLENGES IN COMPUTATIONAL SCIENCES

    In this summer school, leading American, Canadian, European and Japanese computational scientists and HPC technologists will offer instruction on a variety of topics, including HPC challenges by discipline (e.g., bioinformatics, computer science, chemistry, and physics), HPC programming proficiencies, performance analysis and profiling, algorithmic approaches & numerical libraries, data-intensive computing and scientific visualization.

    Compute Canada/Calcul Canada invites students enrolled in science and engineering fields at Canadian post-secondary institutions to apply for one of 10 (expenses-paid) spots allocated to Canada.

    Interested students should apply by March 9, 2014.

    For more information see http://www.prace-ri.eu/International-Summer-School-2014.

  • Monday June 9 – Friday June 13, University of Toronto, St. George Campus

    COMPUTE ONTARIO SUMMER SCHOOL ON HIGH PERFORMANCE COMPUTING/CENTRAL

    The Compute Ontario Summer School on High Performance Computing provides attendees with opportunities to learn and share knowledge and experience in high performance and technical computing.

    As in previous years, the Summer School on High Performance Computing 2014 will have three installments:

    “West” May 26 – 30 Waterloo

    “Central” June 9 – 13 Toronto

    “East” July (TBD) Ottawa

    The format will be a four day workshop with mixed lectures and hands-on sessions on a number of selected subjects, such as MPI, CUDA, OpenMP, and Visualization (these will vary slightly depending on the location).

    This event will *not* be held at the SciNet Headquarters.

    More details regarding the summer school and how to register, will be announced in the future.

  • Monday June 23 – Friday June 27

    HPCS 2014: HIGH PERFORMANCE COMPUTING SYMPOSIUM IN HALIFAX

    The 28th annual HPCS conference, will take place in Halifax, June 3-27 (of which the first two days are tutorials). This year’s theme is `HPC Serving Society’. Sessions will focus on Bioinformatics, Big Data & Analytics, Computational Materials Science, Ocean & Atmosphere Modelling, Computational Chemistry, Data Cave Visualization, and more.

    We are seeking submissions for contributed works and posters relating to the conference theme. Submissions for contributed works will be due on April 30.

    More information is available at the conference website, http://2014.hpcs.ca.

System News

  • GPC: Ray 2.3.1 installed as a module.
  • GPC: Ghostscript added to the Xlibraries module.
  • GPC: CP2K (trunk of 25 Feb, 2014) installed as a module.
  • GPC: Valgrind 3.9.0 with openmpi 1.6 and intelmpi support installed as a modules.
  • ARC/Gravity: Cuda 5.5 installed as a module (cuda 4.1 still the default).

What’s New On The Wiki?

  • Slides and Recordings of 6 lectures of the “Numerical Tools for Physical Scientists” mini-course.
  • Slides of the TechTalk “Introduction to the Intel Xeon Phi”

What Else Happened at SciNet in the Last Month?

  • Feb 12: Intro to SciNet
  • Feb 12: SciNet User Group meeting with TechTalk on the Intel Xeon Phi.
  • Feb 4-27: Numerical Tools for Physical Scientists mini-course, 6 lectures.

SciNet News February 2014

February 12, 2014 in for_researchers, for_users, newsletter

To keep users posted on what’s going on at SciNet, we send monthly emails with the new features, planned courses and events and accomplishments at SciNet (Want to share a testimonial? Email it to support@scinet.utoronto.ca).

Below you will find the salient points of what has happened or changed during the last month and what will happen in the near future. Among these is the participation of Compute Canada in the International High Performance Computing Summer School in Budapest.

As always, details are available on the wiki or the SciNet education website.

– The SciNet team

Events Coming Up

Unless stated otherwise, all events take place at the SciNet Headquarters, Rm 235 of 256 McCaul Street, Toronto. All events below are free for users but we ask that you sign up (“enroll”) on the education website.

  • February 4, 6, 11, 13, 25, 27, and March 4 and 6: 11am-12noon

    NUMERICAL TOOLS FOR PHYSICAL SCIENTISTS

    Lectures of Part 2 of the Scientific Computing Course

    Topics: Modeling, floating point computations, random numbers and Monte Carlo, ODEs and Molecular Dynamics, linear algebra, fast Fourier transforms

    Participation counts towards the SciNet Scientific Computing Certificate.

    For more information and enrollment, go to the course website.

  • Wednesday February 12, 10:30 am – 11:30 am

    INTRO TO SCINET

    A class of approximately 90 minutes where you will learn how to use the systems. Experienced users may still pick up some valuable pointers during these sessions.

    Participation counts towards the SciNet HPC Certificate.

    For more information and enrollment, go to the course website.

  • Wednesday February 12, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    The SciNet Users Group (SNUG) meetings are every month on the second Wednesday, and involve (free) pizza, user discussion, feedback, and a one or two short talks on topics or technologies of interest to the SciNet community.

    This time, we will have

    • TechTalk:

      Intel Xeon Phi: An X86 Coprocessor with 50+ cores

      Scott Northrup (SciNet)

    • User discussion
    • Pizza!

    For more information and enrollment, go to the course website.

  • Tuesdays, almost bi-weekly, 2:30 pm – 3:30 pm

    COAST-TO-COAST SEMINAR SERIES

    The Coast to Coast Seminar is an hour-long presentation given on a scientific topic and is made accessible to audiences at a number of remote sites across Canada through collaboration technology.

    SciNet is a local seminar location for this series.

    The topic of the Spring 2014 Coast to Coast seminar series is “Technology for Aging Well” and it is built around a pan-Canadian project titled AGE-WELL (Aging Gracefully across Environments using Technology to Support Wellness, Engagement and Long Life.)

    The focus of the series will be a discussion how to use technology in helping support the Canadian aging population to ensure that all Canadians can grow older with dignity and grace.

    Feb 25: Ron Baecker (U.Toronto) and David Kaufman (Simon Fraser U)

    Mar 11: Geoff Fernie (U.Toronto)

    Mar 25: Alex Mihailidis (U.Toronto) and Rafik A. Goubran (Carleton U)

    For more information see
    http://www.irmacs.sfu.ca/events/coast-coast-seminars
    or
    https://www.westgrid.ca/events/coast-to-coast-seminar-series

  • Sunday March 9, 2014

    DEADLINE APPLICATION INTERNATIONAL SUMMER SCHOOL ON HPC

    See below.

  • Wednesday March 12, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • Wednesday March 12, 14:30 pm – 15:30 pm

    INTRODUCTION TO GPU COMPUTING USING CUDA

    Scott Northrup (SciNet)

    The CUDA (Compute Unified Device Architecture) parallel programming model harnesses the power of Nvidia graphics processing units (GPUs) to enable dramatic increases in computing performance for graphics, 3D content, video and scientific processing-intensive applications. This talk will provide an introduction to CUDA and examples of applications within research. Participants should be familiar with C/C++ development.

    For more information and enrollment, go to the course website.

    This talk is also offered remotely, for details see:
    https://www.westgrid.ca/events/introduction-to-gpu-computing-using-cuda

  • March 11, 13, 18, 20, 25, 27, and April 1 and 3: 11am-12noon

    HIGH PERFORMANCE SCIENTIFIC COMPUTING

    Lectures of Part 3 of the Scientific Computing Course

    Topics: parallel programming concepts, shared memory parallel programming with openmp, distributed memory parallel programming with mpi, hybrid programming.

    Participation counts towards the SciNet HPC Certificate.

    For more information and enrollment, go to the course website.

  • Wednesday April 9, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • Wednesday May 14, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • Sunday June 1 – Saturday June 6, Budapest, Hungary

    5TH INTERNATIONAL SUMMER SCHOOL ON HPC CHALLENGES IN COMPUTATIONAL
    SCIENCES

    In this summer school, leading American, Canadian, European and Japanese computational scientists and HPC technologists will offer instruction on a variety of topics, including HPC challenges by discipline (e.g., bioinformatics, computer science, chemistry, and physics), HPC programming proficiencies, performance analysis & profiling, algorithmic approaches & numerical libraries, data-intensive computing and scientific visualization.

    Compute Canada/Calcul Canada invites students enrolled in science and engineering fields at Canadian post-secondary institutions to apply for one of 10 spots allocated to Canada.

    Interested students should apply by March 9, 2014.

    For more information see http://www.prace-ri.eu/International-Summer-School-2014.

  • Summer 2014

    ONTARIO SUMMER SCHOOL ON HIGH PERFORMANCE COMPUTING 2014

    Preparations are underway for the 2014 edition of the Ontario Summer School on High Performance Computing. This summer school provides attendees with opportunities to learn and share knowledge and experience in high performance and technical computing.

    As in previous years, the Ontario Summer School on High Performance Computing 2014 will likely have three installments, one in Western Ontario, one in Central Ontario and one in Eastern Ontario.

    Dates and locations are still to be determined, and will be announced in a future SciNet News email.

System News

  • GPC: As a precaution, emails by the Moab/Torque scheduler have been disabled because of a potential security vulnerability.
  • GPC: Version 4.1.2 of the intelmpi implementation installed as a module.
  • GPC: Intel compiler 14.0.1 available as a module.
  • GPC: Python 2.7.5 module now supports IPython Notebook.
  • GPC: CP2K 2.4.0 installed as a module.
  • GPC: Intel tools 2013 installed as module inteltools.

What’s New On The Wiki?

  • Slides and Recordings of all 8 lectures of the “Scientific Software Development” mini-course
  • Improved description of the Application Process for SciNet accounts
  • Slides of the TechTalk “IPython Notebook on GPC”

What Else Happened at SciNet in the Last Month?

  • Jan 15 & 16: Scheduled maintenance downtime
  • Jan 14 & 28: Coast-to-coast seminars
  • Jan 15: SciNet User Group meeting with TechTalk on the IPython Notebook
  • Jan 7-30: Scientific software development mini course

SciNet News January 2014

January 15, 2014 in for_researchers, for_users, newsletter

To keep users posted on what’s going on at SciNet, we send monthly emails with the new features, planned courses and events and accomplishments at SciNet.

Taking of accomplishments, we are compiling user success stories and testimonials (http://www.scinethpc.ca/testimonials). If you want to share your SciNet-powered success stories, please let us know by emailing to support@scinet.utoronto.ca.

Below you will find the salient points of what has happened or changed during the last month and what will happen in the near future. Among these is the start of the term-long graduate course in Scientific Computing, which can be taken for credit by physics, astrophysics and chemistry graduate students.

As always, details are available on the wiki or the SciNet education website.

Happy new year!

– The SciNet team

Events Coming Up

Unless stated otherwise, all events take place at the SciNet Headquarters, Rm 235 of 256 McCaul Street, Toronto. All events below are free for users but we ask that you sign up (“enroll”) on the education website.

  • January 15, 7:00 am – January 16, 6:00 pmSCHEDULED MAINTENANCE DOWNTIME

    Systems will be taken down around 7 am on Wednesday Jan 15, andare expected to be back up by 6 pm on Thursday Jan 16. Check the wiki for updates.

  • January 18SCRATCH PURGING

    Because of the maintenance downtime we will postpone the scratch
    purging from the 15th of January to the 18th of January.

  • Jan-Mar, Tuesdays, almost bi-weekly, 2:30 pm – 3:30 pmCOAST-TO-COAST SEMINAR SERIES

    The Coast to Coast Seminar is an hour-long presentation given on a scientific topic and is made accessible to audiences at a number of remote sites across Canada through collaboration technology.

    SciNet is a local seminar location for this series.

    The topic of the Spring 2014 Coast to Coast seminar series is “Technology for Aging Well” and it is built around a pan-Canadian project titled AGE-WELL (Aging Gracefully across Environments using Technology to Support Wellness, Engagement and Long Life.) The focus of the series will be a discussion how to use technology in helping support the Canadian aging population to ensure that all Canadians can grow older with dignity and grace.

    Jan 14: Arlene Astell (U.Toronto), Uwe Glaesser (Simon Fraser U) Jan 28: Andrew Sixmith (Simon Fraser U) Feb 25: Ron Baecker (U.Toronto) and David Kaufman (Simon Fraser U) Mar 11: Geoff Fernie (U.Toronto) Mar 25: Alex Mihailidis (U.Toronto) and Rafik A. Goubran (Carleton U)

    For more information see
    the irmacs site or the westgrid seminar site

  • Wednesday January 15, 12:00 noon – 1:00 pmSCINET USER GROUP (SNUG) MEETING

    The SciNet Users Group (SNUG) meetings are every month on the second Wednesday, and involve (free) pizza, user discussion, feedback, and a one or two short talks on topics or technologies of interest to the SciNet community.

    This time, we will have

    • TechTalk:IPython Notebook: Accessing IPython running on the GPC from your browser
    • User discussion
    • Pizza!

    For more information and enrollment, go to the course website.

  • Winter 2014SCIENTIFIC COMPUTING COURSE

    Many computational projects start off with knowledge of the science you want to do, and with a bit of programming experience. It can be an arduous journey to get to a (maintainable) piece of code which you trust to compute the right thing. This course is aimed at reducing your struggle, and make you a more efficient computational scientist. Topics include well-established best practices for developing software as it applies to scientific computations, common numerical techniques and packages (so you don’t reinvent the wheel), and aspects of high performance computing.

    The course consists of three parts:

    • Part 1: Scientific Software Development & Design
    • Part 2: Numerical Tools for Physical Scientists
    • Part 3: High Performance Scientific Computing

    Each part consists of eight lectures of one hour.

    Note that these parts can be taken as “mini-courses” by astrophysics graduate students and as “modular courses” by physics and chemistry graduate students at the University of Toronto. (Feel left out? Talk to your graduate office!)

    Participation in parts 1 and 2 counts towards the SciNet Scientific Computing Certificate.

    Participation in part 3 counts towards the SciNet HPC Certificate.

    For more information and enrollment, go to the course websites for
    part 1, part 2, and part 3.

  • January 7, 9, 14, 16, 21, 23, 28, 30: 11:00 am – 12:00 noonSCIENTIFIC SOFTWARE DEVELOPMENT

    Lectures of Part 1 of the Scientific Computing Course

    Topics: C++, version control, make, modular programming, testing, debugging, profiling

    For more information and enrollment, go to the
    course website.

  • Wednesday February 12, 10:30 am – 11:30 amINTRO TO SCINET

    A class of approximately 90 minutes where you will learn how to use the systems. Experienced users may still pick up some valuable pointers during these sessions.

    Participation counts towards the SciNet HPC Certificate.

    For more information and enrollment, go to the course website.

  • Wednesday February 12, 2014, 12:00 noon – 1:00 pmSCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • February 4, 6, 11, 13, 25, 27, and March 4 and 6: 11am-12noonNUMERICAL TOOLS FOR PHYSICAL SCIENTISTS

    Lectures of Part 2 of the Scientific Computing Course

    Topics: Modelling, floating point computations, random numbers and Monte Carlo, ODEs and Molecular Dynamics, linear algebra, fast Fourier transforms

    For more information, go to the course website.

  • Wednesday March 12, 2014, 12:00 noon – 1:00 pmSCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • Wednesday March 12, 14:30 pm – 15:30 pmINTRODUCTION TO GPU COMPUTING USING CUDA

    Scott Northrup (SciNet)

    The CUDA (Compute Unified Device Architecture) parallel programming model harnesses the power of Nvidia graphics processing units (GPUs) to enable dramatic increases in computing performance for graphics, 3D content, video and scientific processing-intensive applications. This talk will provide an introduction to CUDA and examples of applications within research. Participants should be familiar with C/C++ development.

    For more information, go to the course web site.

    This talk is also offered remotely, see here for details.

  • March 11, 13, 18, 20, 25, 27, and April 1 and 3: 11am-12noonHIGH PERFORMANCE SCIENTIFIC COMPUTING

    Lectures of Part 3 of the Scientific Computing Course

    Topics: parallel programming concepts, shared memory parallel programming with openmp, distributed memory parallel programming with mpi, hybrid programming.

    For more information and enrollment, go to the course website.

  • Wednesday April 9, 2014, 12:00 noon – 1:00 pmSCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • Wednesday May 14, 2014, 12:00 noon – 1:00 pmSCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

System News

  • The new resource allocations will take effect on January 10, 2014, for groups who were awarded an allocation.
  • Scheduled maintenance downtime on January 15-16 (see above).
  • GPC: discovar, a genome assember, installed as a module.
  • GPC: allpaths-lg, a short read genome assembler, installed as a module.
  • GPC: Of the ab initio quantum chemistry package gamess, the version of May 1, 2013 has been installed as a module.
  • TCS: HDF4 file format library version 4.2.6 installed.
  • TCS: Newest IBM compilers (xlf 14.1 and xlc 12.1) are now the default.
  • TCS: zlib and slib compression libraries installed as module ‘compression’
  • BGQ: cmake version 2.8.12.1 installed as a module.
  • BGQ: serial and parallel HDF5 v1.8.12 libraries installed as modules.

What Else Happened at SciNet in the Last Month?

  • Dec 11: Intro to SciNet session
  • Dec 11: SNUG w/TechTalk about Resources at SciNet

SciNet News December 2013

December 9, 2013 in for_researchers, for_users, newsletter

To keep users posted on what’s going on at SciNet, every month, we compile the new features, planned courses and events and accomplishments at SciNet.

Speaking of accomplishments, we are compiling user success stories and testimonials. If you want to share your SciNet-powered success stories, please let us know by emailing to support AT scinet DOT utoronto DOT ca.

Below you will find the salient points of what has happened or changed during the last month and what will happen in the near future. Among these is the start of the term-long graduate course in Scientific Computing, which can be taken for credit by physics, astrophysics and chemistry graduate students at the University of Toronto.

As always, details are available on the wiki or the SciNet education website.

Happy new year!

– The SciNet team

Events Coming Up

Unless stated otherwise, all events take place at the SciNet Headquarters, Rm 235 of 256 McCaul Street, Toronto. All events below are free for users but we ask that you sign up (“enroll”) on the education website.

  • Wednesday December 11, 2013, 10:30 am – 11:30 am

    INTRO TO SCINET

    A class of approximately 90 minutes where you will learn how to use the systems. Experienced users may still pick up some valuable pointers during these sessions.

    Participation counts towards the SciNet HPC Certificate.

    For more information and enrollment, go to the course website.

  • Wednesday December 11, 2013, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    The SciNet Users Group (SNUG) meetings are every month on the second Wednesday, and involve (free) pizza, user discussion, feedback, and a one or two short talks on topics or technologies of interest to the SciNet community.

    This time, we will have

    • TechTalk:

      New and Existing Resources Available at SciNet

    • User discussion
    • Pizza!

    For more information and enrollment, go to the course website.

  • Winter 2014

    SCIENTIFIC COMPUTING COURSE

    Many computational projects start off with knowledge of the science you want to do, and with a bit of programming experience. It can be an arduous journey to get to a (maintainable) piece of code which you trust to compute the right thing. This course is aimed at reducing your struggle, and make you a more efficient computational scientist. Topics include well-established best practices for developing software as it applies to scientific computations, common numerical techniques and packages (so you don’t reinvent the wheel), and aspects of high performance computing.

    The course consists of three parts:

    • Part 1: Scientific Software Development & Design
    • Part 2: Numerical Tools for Physical Scientists
    • Part 3: High Performance Scientific Computing

    Each part consists of eight lectures of one hour.

    Note that these parts can be taken as “mini-courses” by astrophysics graduate students and as “modular courses” by physics and chemistry graduate students at the University of Toronto. (Feel left out? Talk to your graduate office!)

    Participation in parts 1 and 2 counts towards the SciNet Scientific Computing Certificate.

    Participation in part 3 counts towards the SciNet HPC Certificate.

    For more information and enrollment, go to the course websites for
    part 1, part 2, and part 3.

  • January 7, 9, 14, 16, 21, 23, 28, 30: 11:00 am – 12:00 noon

    SCIENTIFIC SOFTWARE DEVELOPMENT

    Lectures of Part 1 of the Scientific Computing Course

    Topics: C++, version control, make, modular programming, testing, debugging, profiling

    For more information and enrollment, go to the
    course website.

  • Wednesday January 8, 2014, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • Wednesday February 12, 2014, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • February 4, 6, 11, 13, 25, 27, and March 4 and 6: 11am-12noon

    NUMERICAL TOOLS FOR PHYSICAL SCIENTISTS

    Lectures of Part 2 of the Scientific Computing Course

    Topics: Modelling, floating point computations, random numbers and Monte Carlo, ODEs and Molecular Dynamics, linear algebra, fast Fourier transforms

    For more information, go to the course website.

  • Wednesday March 12, 2014, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • March 11, 13, 18, 20, 25, 27, and April 1 and 3: 11am-12noon

    HIGH PERFORMANCE SCIENTIFIC COMPUTING

    Lectures of Part 3 of the Scientific Computing Course

    Topics: parallel programming concepts, shared memory parallel programming with openmp, distributed memory parallel programming with mpi, hybrid programming.

    For more information and enrollment, go to the course website.

  • Wednesday April 9, 2014, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

  • Wednesday May 14, 2014, 12:00 noon – 1:00 pm

    SCINET USER GROUP (SNUG) MEETING

    TechTalk: TBA

    For more information and enrollment, go to the course website.

System News

  • SciNet network connection will be down on Monday 9 Dec from 7:00 am to 11:00 am. Jobs will continue to run but users will not be able to connect to any SciNet systems.
  • The user-contributed x86 system “Sandy” is now available for job submission by other SciNet users as well. Jobs would be scheduled with default priority, depending on node availability. This Sandybridge x86 cluster has 76 nodes with 16 cores and 64GB of RAM per node. For more information on the system and how to use it, see the wiki.
  • Access to the user-contributed GPU system “Gravity” can now be requested by users (similar to ARC). Gravity is a cluster of 49 nodes, each node with 12 cores, 32 GB, and two NVIDIA Tesla M2090 GPUs. The GPUs have CUDA capability 2.0, 512 CUDA cores and 6 GB of RAM. For more information on the system and how to use it, see the wiki.
  • BGQ: Python 2.7.3 installed, with numpy support.
  • GPC: ffmpeg, a movie encoder, installed as a module.
  • P7: cmake 2.8.8, as build system, installed.
  • GPC, P7: Visit 2.6.3, a visualization program, installed as a module.

Added to the Wiki

All new wiki content below is listed and linked on the wiki main page.

  • Slides and Recordings of (almost) all lectures of “Research Computing with Python”
  • Links to recent TechTalks
  • BGQ Software list updated

What Else Happened at SciNet in the Last Month?

  • November 5 – 28, 2013: Final four lectures of “Research Computing with Python”
  • November 13, 2013: SNUG Meeting with TechTalk on Molecular Motors by Peter Colberg
  • November 18 and 19, 2013: Remotely hosted HPCVL’s “Introduction to HPC”.
  • November 22 and 23, 2013: Two unrelated events, a power glitch and a fabric manager dying caused running jobs to get lost. Both occurred at night, but systems were back up around 10am in the morning.