EVENTS COMING UP

• Mon Aug 22: INTRODUCTION TO GPGPU WITH CUDA

  The goal of this one-day course is for incoming students, new to GPGPU but familiar with scientific programming in 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.

  For more information and to sign up, see: https://support.scinet.utoronto.ca/courses/?q=node/37

• Fri Aug 26, 12:00 noon: INTRO TO SCINET

  Learn how to use SciNet systems in approximately 90 minutes. Extremely useful for new users, but experienced users may still pick up some valuable pointers.

  More sessions may be given throughout the year depending on user interest, but only the first one (Aug 25) will feature pizza!

  Sign up on https://support.scinet.utoronto.ca/courses/?q=node/31.

• Tue Aug 30 (tentative): GPGPU WORKSHOP

  A GPGPU workshop at SciNet in which representatives from research groups will give short talks, and then we discuss.

  More info coming soon at https://support.scinet.utoronto.ca/courses/?q=node/38

• Fri Sep 9, 12:00 noon: INTRO TO SCINET

  A repeat of the Aug 26 session, but without the distraction of pizza.

  Sign up on https://support.scinet.utoronto.ca/courses/?q=node/32.

• Wed Sep 14, 12:00 noon: SNUG (SCINET USERS GROUP) MEETING
  • TechTalk by Jaime Pinto (SciNet) on “HPSS – SciNet’s storage capacity expansion”
  • User discussion
  • Pizza!

  Sign up on https://support.scinet.utoronto.ca/courses/?q=node/33

• Other SNUG meetings next term are scheduled on Oct 12, Nov 9 and Dec 14, and you can sign up on https://support.scinet.utoronto.ca/courses/?q=SNUG_Meetings.
• Fall Term: SCIENTIFIC COMPUTING COURSE

  This course will likely follow the format of a weekly 3-hour session. It will consist 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 4 lectures, which include hands-on, and can be taken separately.

  More details (including whether you can get credit for this course!) will soon be available on the courses website: https://support.scinet.utoronto.ca/courses.

SYSTEM CHANGES

• SciNet has a mailing lists for people interested in GPGPU computing. To receive information on courses, workshop and other GPGPU related events, sign up at: https://support.scinet.utoronto.ca/mailman/listinfo/scinet-gpgpu .
• To deal with the security issue in July (see below), extensive updates and tightening of security measured were performed. Users were required to change there passwords and regenerate (pass-phrase protected) ssh keys if they used these (highly recommended, see wiki). We also updated the operating system on the gpc to ensure the security hole could not be used again.
• GPC: nedit installed as a module.
• P7: any user that has access to the power-6 cluster tcs, can now give the power-7 cluster (p7) a try. This cluster currently consists of 5 IBM Power 755 servers (3 or more servers to be added later this year). Each has four 8-core 3.3GHz Power7 CPUs and 128GB RAM, and features 4-way Simultaneous MultiThreading giving 128 threads per server. Linux is the operating system. Both the GCC and IBM compilers are installed, as well as POE. LoadLeveler is used as the scheduler. Instruction on usage are on the wiki.

ADDED TO THE WIKI IN JULY

(All new wiki content below is listed and linked on the main page: http://wiki.scinet.utoronto.ca/wiki/index.php/SciNet_User_Support_Library#What.27s_New_On_The_Wiki)

• Update info on how to distinguish what cluster you are on in the .bashrc file.
• Improved help for the ish application, which can browse previously indexed directories on the hpss system.
• More material on HPSS was added.

WHAT ELSE HAPPENED AT SCINET IN JULY?

• We were hacked, as were several other hpc consortia in Canada. We needed more than 4 days of shutdown to make sure nothing vital was compromised, to reinstall and upgrade systems, and to safe-guard ourselves against any possible future attack. It seems that the attack started with a stolen user’s password. So be careful not to let your password fall in the wrong hands, and generate pass- phrase protected ssh keys for increased security!
• SciNet, in collaboration with the research groups at the Canadian Institute for Theoretical Astrophysics, the University of Toronto’s Institute for Aerospace Studies and Departments of Chemistry, Biochemistry, and Electrical and Computer Engineering, and the Medical Image Analysis Laboratory at Sunnybrooke Hospital, has successfully applied for the University of Toronto to become an NVIDIA Teaching Centre as well as an NVIDIA Research Centre. So you can expect more GPGPU training facilities from SciNet soon!