Thematic CERN School of Computing on Heterogeneous Architectures

9 Jun 2024, 11:00 → 15 Jun 2024, 11:30 Europe/Belgrade

Room m25/26 (ITS )

Alberto Pace (CERN), Kristina Gunne (CERN), Andrzej Nowicki (CERN)

The 14^th Thematic CERN School of Computing (tCSC 2024) will take place on June 9-15, 2024. The theme of the school is "Scientific Software for Heterogeneous Architectures" - see the academic programme for more details.

This school is organized by CERN in collaboration with the Information Technology School (ITS), situated at the Comtrade Campus in Belgrade, Serbia. CSC students will have access to modern lecture rooms, labs and a sports hall as well as restaurants and cafeterias. The campus is situated in the new Belgrade area by the Danube river, only a short busride from the old city center of Belgrade

The School is aimed at postgraduate (i.e. minimum of Bachelor degree or equivalent) students, engineers and scientists with a few years' experience in particle physics, in computing, or in related fields. We welcome applications from all countries and nationalities.

Important Dates

• Wednesday March 27 - applications deadline
• Wednesday April 10 - invitations sent to the selected participants
• Wednesday May 8  - registration fee payment deadline
• Sunday June 9 - Saturday June 15- the school

                  

Photo Gallery

School final movie

tCSC_2024_booklet.pdf

Sunday 9 June

15:30 Registration at IN Hotel

1 Welcome and introduction

2024-06-09-opening_slides.pdf

2 Self presentations

One minute-one slide self presentations

tCSC2024_self_presentations.pdf

18:00 Welcome drink at the Hotel Roof top bar

19:30 Welcome Dinner - IN Hotel

The dinner will take place at the Konoba (Tavern) Varos
Tavern "Varos"
Ban Mladenova 9
21 000 Split
Croatia
Website: http://www.konobavaros.com/hr/

Monday 10 June

07:00 Breakfast

3 Opening Session

Speaker: Alberto Pace (CERN)

4 Preparing for the HL-LHC computational challenge

In this talk we will introduce some basic concepts related to HEP data processing and analysis workflows, seeing them in action in the context of LHC experiments. We’ll also talk about the evolution of the LHC accelerator and experiments. We’ll characterise at a high level what are the consequences of those upgrades for the HEP data processing software, in particular in the context of an evolving hardware and computing infrastructure.

Speaker: Danilo Piparo (CERN)

tCSC2024_Intro.pdf

5 Announcements

11:00 Break

6 CPU Hardware Architecture and Evolution

CPU Architecture and High Performance - lecture 1

• Hardware evolution of the CPU
• Memory hierarchy, caching, NUMA
• Microarchitecture of modern CPUs

Speaker: Guilherme Amadio (CERN)

01_cpu.pdf

7 Performance Analysis on Modern CPUs

CPU Architecture and High Performance - lecture 2

• Performance analysis tools for Linux
• CPU features for performance analysis
• Top-down microarchitecture analysis

Speaker: Guilherme Amadio (CERN)

02_performance.pdf

13:30 Lunch

8 CPU Architecture and High Performance - exercise 1

Speaker: Guilherme Amadio (CERN)

exercise-1.pdf

15:30 Break

Scavenger Hunt in old Belgrade: Transport to the old town of Belgrade

17:00 Scavenger Hunt in old Belgrade

19:30 Dinner in the old town of Belgrade- Klub Knjizevnika

Francuska 7, Belgrade 11000, Serbia

Tuesday 11 June

07:00 Breakfast

9 Writing parallel software

Parallel and Optimised Scientific Software - lecture 1

• Amdahl's and Gustafson's laws
• Asynchronous execution
• Finding concurrency, task vs. data parallelism
• Using threading in C++ and Python, comparison with multi-process
• Resource protection and thread safety
• Locks, thread local storage, atomic operations

Speaker: Sebastien Ponce (CERN)

WritingParallelSoft-pres.pdf

10 Writing efficient software

Parallel and Optimised Scientific Software - lecture 2

• Why Python and C++ ?
• Recent evolutions: C++ 11/14/17
• Modern features of C++ related to performance
• Templating versus inheritance, pros and cons of virtual inheritance
• Python 3, and switching from Python 2

Speaker: Sebastien Ponce (CERN)

WritingEfficientSoft-pres.pdf

11 Announcements

11:00 Coffee break

12 Low-level Performance Optimization Guidelines

CPU Architecture and High Performance - lecture 3

• Main sources of performance bottlenecks
• Floating point arithmetics performance
• Advanced low-level performance tuning

Speaker: Guilherme Amadio (CERN)

tcsc24-cpu.pdf

13 CPU Architecture and High Performance - exercise 2

Speaker: Guilherme Amadio (CERN)

exercise-2.pdf

13:30 Lunch

14:30 Study or sports time

16:00 Coffee break

14 Optimizing existing large codebase

Parallel and Optimised Scientific Software - lecture 3

• Measuring performance, tools and key indicators
• Improving memory handling
• The nightmare of thread safety
• Code modernization and low level optimizations
• Data structures for efficient computation in modern C++

Speaker: Sebastien Ponce (CERN)

LargeCodeOptimization-pres.pdf

15 Data-Oriented Design

CPU Architecture and High Performance - lecture 4
- Principles of data-oriented design
- Memory access and data-type profiling
- Data structure performance optimization

Speaker: Guilherme Amadio (CERN)

18:30 Study or sports time

19:30 Dinner at ITS

16 Managing Large Databases: Challenges and Lessons

With around 100 Oracle databases and several PB of data, CERN faces numerous intriguing challenges and unique caveats in regards to managing the database infrastructure. This lecture aims to provide an overview of both past and current issues and obstacles encountered in the realm of database systems at CERN. By doing so, it seeks to offer valuable insights into potential problems to avoid and effective solutions to address encountered difficulties.

Speaker: Andrzej Nowicki (CERN)

Wednesday 12 June

07:00 Breakfast

17 Scientific computing on heterogeneous architectures

Programming for Heterogeneous Architectures - lecture 1

• Introduction to heterogeneous architectures and the performance challenge
• From general to specialized: Hardware accelerators and applications
• Type of workloads ideal for different accelerators
• Trade-offs between multi-core and many-core architectures
• Implications of heterogeneous hardware on the design and architecture of scientific software
• Embarrassingly parallel scientific applications in HPC and CERN

Speaker: Daniel Hugo Campora Perez (NVIDIA)

0_scientific_computing_on_heterogeneous_architectures.pdf

18 Optimizing existing large codebase - exercise 1

Optimisation of an existing, production grade large codebase

Speaker: Sebastien Ponce (CERN)

exercise4.pdf

19 Announcements

11:00 Coffee break

20 Optimizing existing large codebase - exercise 2

Optimisation of an existing, production grade large codebase

Speaker: Sebastien Ponce (CERN)

Student Lightning talks: The Art of Code: Pharo

Convener: Jovan Kupusović

Hinderberger_ProcessingAndCompressionFPGA.pdf

Lighting_talk_DaMolin.pdf

MC_Run_lightning_talk.pdf

tCSC_lightning_talk_AdePT 2.pdf

The art of code Pharo.pdf

Student Lightning talks: MC-Run: Scalable MC Event Generation and Analysis

Convener: Cedric Verstege (KIT - Karlsruhe Institute of Technology (DE))

Hinderberger_ProcessingAndCompressionFPGA.pdf

Lighting_talk_DaMolin.pdf

MC_Run_lightning_talk.pdf

tCSC_lightning_talk_AdePT 2.pdf

The art of code Pharo.pdf

Student Lightning talks: Data Processing and Compression on FPGAs

Convener: Peter Hinderberger (Technical University of Munich)

Hinderberger_ProcessingAndCompressionFPGA.pdf

Lighting_talk_DaMolin.pdf

MC_Run_lightning_talk.pdf

tCSC_lightning_talk_AdePT 2.pdf

The art of code Pharo.pdf

Student Lightning talks: AdePT - accelerating GEANT4 applications on GPUs

Convener: Severin Diederichs (CERN)

Hinderberger_ProcessingAndCompressionFPGA.pdf

Lighting_talk_DaMolin.pdf

MC_Run_lightning_talk.pdf

tCSC_lightning_talk_AdePT 2.pdf

The art of code Pharo.pdf

Student Lightning talks: Precision physics in current and future collider experiments

Convener: Giacomo Da Molin (Laboratory of Instrumentation and Experimental Particle Physics (PT))

Hinderberger_ProcessingAndCompressionFPGA.pdf

Lighting_talk_DaMolin.pdf

MC_Run_lightning_talk.pdf

tCSC_lightning_talk_AdePT 2.pdf

The art of code Pharo.pdf

13:20 Lunch

21 Transport to Nikola Tesla museum - Departure at 14.15 sharp!

15:00 Visit to the Nikola Tesla Museum

https://tesla-museum.org/en/home/

22 Free time in Belgrade - visits, shopping, resting in a cafe...

The Tesla museum is located in Vracar, one of the nicest neighborhoods in Belgrade, the Krunska street where the Museum is located is one of the most famous streets in Belgrade.
Suggested activities, free to chose from
- A must see: Saint Sava Temple, the 4th biggest Eastern Orthodox church in the world, only about a 10 min walk from the Museum.
- Tasmajdan Park, with St. Mark’s Church, 10 min from museum.
- Walk to Kalemegdan park, visit to the Belgrade Fortress (2,3 km from museum)
https://www.beogradskatvrdjava.co.rs/?lang=en
- Shopping around in old Belgrade
- Go to a cafe or pub, a good cafe chain is Kafeterija

Meeting at 19.30 at Restaurant Smokvica in Molerova 33 (6 min by foot from Nikola Tesla museum)

19:30 Dinner in the old town of Belgrade - Smokvica Molerova

Restaurant Smokvica in Molerova 33

Thursday 13 June

07:00 Breakfast

23 Performant programming for GPUs

Programming for Heterogeneous Architectures - lecture 2

• From SIMD to SPMD, a programming model transition
• Thread and memory organization
• Basic building blocks of a GPU program
• Control flow, synchronization, atomics

Speaker: Daniel Campora (NVIDIA)

1_programming_for_gpus.pdf

24 Performant programming for GPUs

Programming for Heterogeneous Architectures - lecture 3

• Data locality, coalesced memory accesses, tiled data processing
• GPU streams, pipelined memory transfers
• Under the hood: branchless, warps, masked execution
• Debugging and profiling a GPU application

Speaker: Daniel Campora (NVIDIA)

2_performant_programming_for_gpus.pdf

25 Announcements

11:00 Coffee break

26 Practical vectorization

Parallel and Optimised Scientific Software - lecture 4

• Measuring vectorization level
• What to expect from vectorization
• Preparing code for vectorization
• Vectorizing techniques in C++: intrinsics, libraries, autovectorization

Speaker: Sebastien Ponce (CERN)

PracticalVectorization-pres.pdf

27 Practical vectorization - exercise

Speaker: Sebastien Ponce (CERN)

exercise5.pdf

13:30 Lunch

14:30 Study time or sports

16:30 Coffee break

28 Programming for heterogeneous architectures - exercise

Speaker: Daniel Campora (NVIDIA)

instructions_to_start_gpu_lab.pdf

29 Programming for heterogeneous architectures - exercise

Speaker: Daniel Campora (NVIDIA)

19:30 Dinner Hotel IN

Friday 14 June

07:00 Breakfast

30 Debrief of exercises

31 Design patterns and best practices

Programming for Heterogeneous Architectures - lecture 4

• Good practices: single precision, floating point rounding, avoid register spilling, prefer single source
• Other standards: SYCL, HIP, OpenCL
• Middleware libraries and cross-architecture compatibility
• Reusable parallel design patterns with real-life applications

Speaker: Daniel Campora (NVIDIA)

3_design_patterns_and_best_practices.pdf

32 Announcements

11:00 Coffee break

33 Programming for heterogeneous architectures - exercise

Speaker: Daniel Campora (NVIDIA)

34 Programming for heterogeneous architectures - exercise

Speaker: Daniel Campora (NVIDIA)

13:30 Lunch

35 Exam

15:45 Coffee break

36 Closing Session

17:20 Free time

37 CERN outreach

Speakers: Alberto Pace (CERN), Sebastien Ponce (CERN)

CERNForTCSC.pdf

19:30 Closing dinner/party at ITS IT Club

Saturday 15 June

07:30 Breakfast

38 Departures