SciNet Certificate Program

From SciNetWiki
Jump to: navigation, search

Contents

About the program

SciNet has been teaching courses on scientific technical computing and high performance computing for the Toronto-area research community for several years, and now offers recognition to attendees in the form of SciNet Certificates. There are currently two certificate offerings: A Certificate in Scientific Computing and a Certificate in High Performance Computing.

Requirements for these certificates are based on credit-hours of SciNet courses successfully completed. For a “short course” (typically a day long or shorter, with no between-course homework), a lecture hour counts as one credit hour; for a “long course” with homework due between sessions, a lecture hour counts as 1.5 credit hours. Courses offered by SciNet from January 2013 onwards will count towards these certificates. Students who have successfully taken SciNet courses before January 2013 may contact SciNet to discuss having credits applied from previously-taken courses.

Prerequisites

Some requirements such as programming skills vary per course; these are announced when the courses are announced, on SciNet's course website: https://support.scinet.utoronto.ca/education. Many courses have a hands-on component that requires participants to bring their laptop.

Who can participate in the certificate program?

SciNet courses and certificates are open to all SciNet users (for some courses, a SciNet account is not necessary, but these are exceptions). In general, any academic researcher from a Canadian research institution with significant high performance computing requirements to support his or her research may apply for an account on SciNet. For information on how to get an account, go to http://www.scinethpc.ca/2011/09/getting-a-scinet-account.

Do the certificates count as University credentials?

No, the SciNet Certificates are not University credentials, and will not appear on transcripts. However, several SciNet courses are sometimes offered in cooperation with the Astronomy and Physics departments as graduate minicourses, which can count towards course credit; interested students in other departments are encouraged to contact SciNet and their graduate coordinator.

Can I attend the courses online?

Although we are looking into this, currently, this is not possible. You can only get credit for a course by physically attending it. Recordings and slides of many of the courses are, however, posted on the SciNet wiki afterwards: Knowledge_Base:_Tutorials_and_Manuals.

Certificate in Scientific Computing

Scientific computing is now an integral part of the scientific endevour. It is an interdisciplinary field that combines computer science, software development, physical sciences and numerical mathematics.

This certificate indicates that the holder has successfully completed at least 36 credit-hours worth of coursework in general scientific computing topics such as software development, version control, testing, visualization, and data management.

Requirements

  • A total of 36 credit-hours or more from the list of SciNet courses on scientific computing given below.
  • It is also possible to include credit-hours from the list of SciNet courses on high performance computing, but these credit-hours can then no longer be counted toward a Certificate in High Performance Computing.
  • Upon completion of your certificate requirements, you must request your certificate by email to support@scinet.utoronto.ca with the list of SciNet courses on scientific computing that you have taken.

Scientific Computing Course List

Introduction to the Linux Shell (2 credit-hours)
Modern Fortran (7 credit-hours)
Scientific C++ (7 credit-hours)
Scientific Computing Course Part I – Software Development (12 credit-hours)
Scientific Computing Course Part II – Numerical Tools for Physical Scientists (12 credit-hours)
Profiling and Performance Tuning (3 credit-hours)
SciNet Developer Seminars (1 credit-hour each)

This is the latest version of this list.

Certificate in High Performance Computing

High Performance Computing, or supercomputing, is using the largest available computers to tackle big problems that would otherwise be intractable. Such computational power is needed is a wide range of fields, from bioinformatics to astronomy, and big data analytics. Since the largest available computers have a parallel architecture, using and programming high performance computing applications requires a specialized skill level.

This certificate indicates that the holder has successfully completed at least 36 credit-hours of coursework in high performance computing topics such as programming models like OpenMP, MPI, CUDA or parallel development tools like debuggers.

Requirements

  • A total of 36 credit-hours or more from the list of SciNet courses on high performance computing given below.
  • Upon completion of your certificate requirements, you must request your certificate by email to support@scinet.utoronto.ca with the list of SciNet courses on high performance computing that you have taken.

High Performance Computing Course List

Intro to Scinet (1 credit-hour)
Five-day Practical Parallel Programming (up to 32 credit-hours)
Scientific Computing Course Part III – High Performance Computing (12 credit-hours)
Parallel I/O (3 credit-hours)
Ontario HPC Summerschool (up to 24 credit-hours)
Introduction to GPGPU with CUDA (7 credit-hours)
Minicourse: Intro to GPU with CUDA (12 credit-hours)
Parallel debugging (7 credit-hours)

This is the latest version of this list.

-Last updated: May 13, 2013-

Personal tools
Namespaces
Variants
Actions
Systems
Knowledge Base
Wiki
Toolbox