Resource-Aware Computing - Summer 15

From Sccswiki
Revision as of 15:41, 22 May 2015 by Poeppl (talk | contribs)
Jump to navigation Jump to search
Term
Summer 15
Lecturer
Univ.-Prof. Dr. Michael Bader, Dr. Josef Weidendorfer, Emily Mo-Hellenbrand, M.Sc., Alexander Pöppl, M.Sc., Isaías Alberto Comprés Ureña
Time and Place
- Initial Meeting: January 21st, 16:00, room 02.09.023
- Kick-off Meeting: April 16th, 16:00, room 01.06.011
- Flash Talks: April 28th, 15:00 to 16:00, room 02.07.023
- Presentations: see Schedule section below.
Audience
Students from Master Informatics (IN2107), Computational Science and Engineering (IN2183), and Bachelor Informatics (IN0014)
Tutorials
-
Exam
-
Semesterwochenstunden / ECTS Credits
2 SWS (2S) / 4 Credits
TUMonline
Master Seminar Resource-Aware Computing (IN2183,IN2107,IN0014)



Capacity

  • Max. number of participants: 12

Description

In the last ten years the period of vast increases in processing power mostly achieved by increasing the clock frequency of a processor has come to an end. Instead, computer architectures are getting more complex in order to accommodate the growing demand for processing power. Modern CPUs typically have a wide range of SIMD instructions for fine-grained data parallelism, and are capable of executing several threads on each of their several cores. Memory accesses are passed through multiple cache levels to hide memory access latencies. In addition to that, hardware specialized in performing massively parallel computations is getting more and more popular. Examples are GPUs and accelerators such as the Xeon Phi. In the HPC context, several nodes, each with its own CPU(s) and GPU(s) may be joined into a cluster.

Regular programming techniques and paradigms are no longer sufficient to fully utilize this hardware. Frameworks such as OpenCL take the structure and heterogeneity of the underlying hardware into account and provide the programming environment to expose all available resources, such as GPUs and accelerators. Novel approaches (such as invasive computing) expose the allocation of resources to the users, allowing them to request required resources and, by offering the reallocation of resources at runtime, enabling them to adapt to changing computing demands.

The behavior of the hardware at runtime also needs to be considered. Modern Cluster architectures are not necessarily capable to run at peak utilization 100% of the time. To avoid the overheating of the hardware and the resulting degradation of the silicon, the clock frequency of the CPU may be drastically reduced, or single nodes may even be shut down completely for a time. Taking these problems into account is an additional challenge developers face today.

Background: The seminar will (partly) discuss research executed in the collaorative research unit Invasive Computing funded by the German Research Foundation (DFG).

Topic Assignments

Supervisor Topic Assigned To Reviewer 1 Reviewer 2
Alex Invasive algorithms Michael Riesch Evgeny SungJae
Isaías Resource management for heterogeneous architectures Nathaniel Knapp Tamas Matthias
Emily Faults and Fault-Tolerance Christoph Hartlmüller Martin Michal
Isaías Invasive Network-on-a-Chip architectures (iNoCs) Evgeny Agamirzov Felix Oleksandr
Isaías Resource aware runtime systems Tamas Borbáth SungJae Michael
Emily The Invasic Framework Martin Schwörer Matthias Nathaniel
Emily Resource-aware computing concepts Felix Scheffler Michal Christoph
Josef Dark silicon (hardware, current Trends, etc) SungJae Jung Oleksandr Evgeny
Josef Power Saving by Dynamic Voltage and Frequency Scaling (DVFS) and Power Capping (PC) Matthias Kruk Michael Tamas
Alex Characterization of problems and extraction of patterns Michal Szymczak Nathaniel Martin
Alex Heterogeneous computing frameworks: OpenCL Oleksandr Shchur Christoph Felix

Note: An addition of topics is possible for foreign students that arrive once the semester has started.

Literature

Consider the Invasive Computing homepage. We encourage independent research and review of the available literature.

Schedule

Week Time Place Description Material
- 21.01.2015, 16:00 PM MI 02.09.023 Initial Meeting Slides
1st 16.04.2015, 16:00 PM MI 01.06.011 Kick-Off Meeting Slides
3rd 27.04.2015, 23:59 PM - Submission of paper outline -
3rd 28.04.2015, 15:00 PM MI 02.07.023 Flash-Talk Meeting (2-min talk, no slides) -
7th - - Submission of paper draft to supervisor (voluntary) -
8th 05.06.2015, 23:59 PM - Submission of paper (release candidate) for review -
9th 12.06.2015, 23:59 PM - Submission of 2 reviews -
10th 16.06.2015, 14:00 PM MI 02.07.023 Presentation session 1 -
10th 18.06.2015, 14:00 PM t.b.a. Presentation session 2 -
11th 23.06.2015, 14:00 PM MI 02.07.023 Presentation session 3 -
11th 25.06.2015, 14:00 PM t.b.a. Presentation session 4 -
12th 30.06.2015, 14:00 PM MI 02.07.023 Presentation session 5 -
12th 02.07.2015, 14:00 PM t.b.a. Presentation session 6 -
12th 03.07.2015, 23:59 PM - Submission of final paper -

Important Deadlines

  • April 27th, 23:59 PM: submission of paper topic and outline.
  • June 5th, 23:59 PM: submission of paper (release candidate) for review.
  • June 12th, 23:59 PM: submission of 2 reviews.
  • July 3rd, 23:59 PM: submission of final paper.

Note: Failure to meet these deadlines may result in grade deduction or failure of the course.

Requirements

  • Course language: English
  • Independent literature research
  • Paper: 5-10 pages (max 10 pages) in total. IEEE format (see link below) required.
  • Peer review process: 2 reviews from each participant
  • Presentation: 30 minutes talk + 15 minutes discussion
  • Mandatory attendance: Participants must attend all presentations. Absence can be excused for "good" reasons with a prior notification to the supervisors. For absence due to sickness, a doctor's attest must be provided.

Grading

  • Major components: final paper + presentation
  • Other factors: attendance, meeting the deadlines, release candidate paper, plagiarism
  • Bonus factors: commitment, activeness (interaction with supervisor, participation in discussions, etc.), self-implemented code/examples, etc.

Links

IEEE format

Template is updated. 2-columns. The main file to compile is "dm-template.tex".