Praktikum: Evaluierung moderner HPC-Architekturen und -Beschleuniger (IN0012, IN2106, IN4294)
Dieses Praktikum wird nur in englischer Sprache angeboten.
Experimental Evaluation of modern Computing Systems and Accelerators
This lab course will be held in English.
Bachelor/Master practical course (IN0012, IN4294 IN0014, IN2107, IN4602).
Due to popular demand we have decided to introduce a preliminary meeting. We will be covering the course organization and some of the topics of next semester.
This lab is organized in collaboration with LRZ and LMU.
It is provided seperately for bachelor and master students, but carried out as one event.
What you will learn from this course ?
The LRZ recently started building a testbed environment with novel architectures and accelerators called BEAST (Bavarian Energy, Architecture and Software Testbed). This will help in shaping future large LRZ systems (such as successors of SuperMUC-NG). In this lab you will gain hands on experience on the BEAST testbed.
- have the chance to work on newest HPC technology such as CPUs or GPUs from Intel, AMD, NVidia, Marvell, Fujitsu (in particular A64FX - powers the fastest supercomputer in the world (Fugaku) by TOP500 rankings as of June 2020),
- explore various features of novel technology on the node level,
- learn about node level performance optimizations,
- attend to vendor talks on their newest hardware,
- learn various programming models on the node level for multicore CPUs and accelerators, such as OpenMP (for GPU), SIMD, synchronization,
- - for later assignments - make use of OpenCL, SYCL, or CUDA.
The lab is split in two types of assignments:
- In the first assignments, students are expected to work on all architectures.
- For the second part, students work on specific architectures exclusively (assigned to student groups according to preferences).
Students will work in groups, doing reports for each task as well as short presentations of their evaluation results (2 groups will be selected every week to present their reports). Grading is done based on delivered reports and short presentations. This includes source code used for the evaluations, which has to be added to the reports.
Registration for the lab will happen via the matching system.
- Good knowledge of C or C++ on Linux
- Understanding of terms in computer architecture: pipelining, SIMD, multi-core, SMT, TLB, processor caches, replacement policies (before registering in the matching system, ask yourself if you can explain these terms!)
- Interest in computer architecture, benchmarking, and low-level code optimization
Schedule / Dates
Preliminary Meeting: TBD
Weekly meetings during the semester: TBD (probably every Thursday 3pm - 5pm).
First date: TBD