Node-Level Performance Engineering

This course covers performance engineering approaches on the compute node level. Even application developers who are fluent in OpenMP and MPI often lack a good grasp of how much performance could at best be achieved by their code. This is because parallelism takes us only half the way to good performance. Even worse, slow serial code tends to scale very well, hiding the fact that resources are wasted. This course conveys the required knowledge to develop a thorough understanding of the interactions between software and hardware. This process must start at the core, socket, and node level, where the code gets executed that does the actual computational work. We introduce the basic architectural features and bottlenecks of modern processors and compute nodes. Pipelining, SIMD, superscalarity, caches, memory interfaces, ccNUMA, etc., are covered. A cornerstone of node-level performance analysis is the Roofline model, which is introduced in due detail and applied to various examples from computational science. We also show how simple software tools can be used to acquire knowledge about the system, run code in a reproducible way, and validate hypotheses about resource consumption. Finally, once the architectural requirements of a code are understood and correlated with performance measurements, the potential benefit of code changes can often be predicted, replacing hope-for-the-best optimizations by a scientific process.

This course provides - via lectures, demos, and hands-on labs - scientific training in Computational Science, and in addition, the scientific exchange of the participants among themselves.

Veranstaltungsort

Online course
Organizer: HLRS, University of Stuttgart, Germany

Anmeldung

Veranstaltungsbeginn

18. Jun 2024
08:45

Verstaltungsende

21. Jun 2024
13:00

Sprache

Englisch

Einstiegslevel

Fortgeschritten

Themenbereiche

Performance-Optimierung & Debugging

Zurück zur Liste

Prerequisites and content levels

Prerequisites:

Familiarity with Linux and Linux editors is recommended.
Basics/principles of programming in C or Fortran and basic OpenMP.

Content levels:

Intermediate: 8 hours 15 minutes
Advanced: 8 hours 15 minutes
Community: 2 hours 45 minutes

Learn more about course curricula and content levels.

Instructors

Dr. habil. Georg Hager and Dr.-Ing. Jan Eitzinger (NHR@FAU, Friedrich-Alexander-Universität Erlangen-Nürnberg)
Bert Wesarg (ZIH Uni. Dresden) for the tools-day

Agenda

- preliminary -

Day 1 (Tue, June 18)

08:45 Enter the Zoom meeting
09:00 Welcome – Intro
09:30 Computer architecture for software developers part 1
10:15 Coffee break
10:30 Computer architecture for software developers part 2
10:45 Hands-on: Divide benchmark
11:30 Tools: Topology and affinity, frequency
12:00 Lunch
13:00 Hands-On: topology, affinity
13:45 Introduction to the Roofline Model
15:00 Coffee break
15:15 Quiz/Q&A/open end

Day 2 (Wed, June 19)

08:45 Enter the Zoom meeting
09:00 Tools: Performance counters
09:45 Hands-on: performance counters and memory bandwidth
10:30 Coffee break
10:45 Roofline case study: stencils
11:30 Performance Engineering: Basic skills
12:15 Lunch
13:15 Hands-on: Dense matrix-vector multiplication (I)
13:45 Optimal use of parallel resources: ccNUMA
14:15 Hands-on: Dense matrix-vector multiplication (II)
14:45 Roofline case study: Tall & skinny matrix-matrix multiplication
15:15 Coffee break
15:30 Quiz/Q&A/open end

Day 3 (Thu, June 20)

08:45 Enter the Zoom meeting
10:00 Hands-on: SIMD in MiniMD
11:00 Coffee break
11:15 Roofline case study: Sparse matrix-vector multiplication
12:15 Lunch
13:15 Hands-on: Matrix-free CG solver
14:45 Coffee break
15:00 The ECM performance model
16:00 Quiz/Q&A/open end

Day 4 (Fri, June 21: Tools Day)

08:45 Enter the Zoom meeting
09:00 Performance engineering from the application point of view
09:30 Monitoring with Score-P
09:45 Hands-on: MiniMD with Score-P
10:15 Coffee break
10:30 Profile exploration and measurement scoring
10:45 Trace exploration with Vampir
11:15 Coffee break
11:30 Hands-on: Load imbalance: SpMV
12:15 Wrap-up

Handouts

Before the course, the course material and an updated agenda will be available here.

An older version of this course with most of the material (including the audio information) can also be viewed in the online self-study materials.

Registration-information

Register via the button at the top of this page.
We encourage you to register to the waiting list if the course is full. Places might become available.

Fees

Students without master’s degree or equivalent: 32.50 Euro
PhD students or employees at a German university or public research institute: 67.50 Euro
PhD students or employees at a university or public research institute in an EU, EU-associated or PRACE country other than Germany: 135 Euro
PhD students or employees at a university or public research institute outside of EU, EU-associated or PRACE countries: 270 Euro
Other participants, e.g., from industry, other public service providers, or government: 690 Euro

Link to the EU and EU-associated (Horizon Europe), and PRACE countries.

Our course fees include coffee breaks (in classroom courses only).

Contact

Lucienne Dettki, phone 0711 685 63894, lucienne.dettki(at)hlrs.de

HLRS Training Collaborations in HPC

HLRS is part of the Gauss Centre for Supercomputing (GCS), together with JSC in Jülich and LRZ in Garching near Munich. EuroCC@GCS is the German National Competence Centre (NCC) for High-Performance Computing. HLRS is also a member of the Baden-Württemberg initiative bwHPC.

This course is provided within the framework of the bwHPC training program.

Further courses

See the training overview and the Supercomputing Academy pages.