Opening Ceremony

• Sebastian Lopienski (CERN)
• Frederic Hemmer (CERN)
• Ivica Puljak (University of Split, FESB)

Introduction to efficient computing

• Andrzej Nowak

Technologies and Platforms - lecture 1 - The evolution of computing hardware and what it means in practice - The seven dimensions of performance - Controlling and benchmarking your computer and software - Software that scales with the hardware - Advanced performance tuning in hardware

Writing parallel software

• Sebastien Ponce

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

Announcements

Hardware evolution and heterogeneity

• Andrzej Nowak

Technologies and Platforms - lecture 2 - Accelerators, co-processors, heterogeneity - Memory architectures, hardware caching and NUMA - Compute devices: CPU, GPU, FPGA, ASIC etc. - The role of compilers

Modern programming languages for HEP

• Sebastien Ponce

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

Data-oriented design

• Andrzej Nowak

Technologies and Platforms - lecture 3 - Hardware vectorization in detail – theory vs. practice - Software design for vectorization and smooth data flow - How can compilers and other tools help?

Technologies and Platforms - exercise 1

• Andrzej Nowak

Technologies and Platforms - exercise 2

• Andrzej Nowak

Special evening talk: Future of the Universe and of Humanity

• Ivica Puljak (University of Split, FESB)

Technologies and Platforms - exercise 3

• Andrzej Nowak

Technologies and Platforms - exercise 4

• Andrzej Nowak

Announcements

Scientific computing on heterogeneous architectures

• Daniel Campora (Nikhef National institute for subatomic physics (NL))

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

Student presentations

Parallel and Optimised Scientific Software - exercise 3

• Sebastien Ponce

Parallel and Optimised Scientific Software - exercise 4

• Sebastien Ponce

Programming for GPUs

• Daniel Campora (Nikhef National institute for subatomic physics (NL))

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 - Debugging and profiling a GPU application

Optimizing existing large codebase

• Sebastien Ponce

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++

Announcements

Parallel cross-architecture programming

• Daniel Campora (Nikhef National institute for subatomic physics (NL))

Programming for Heterogeneous Architectures - lecture 3 - Data locality, coalesced memory accesses, tiled data processing - Control flow, synchronization, atomics - Other standards: SYCL, HIP, OpenCL - Middleware libraries and cross-architecture compatibility

Programming for Heterogeneous Architectures - exercise 1

• Daniel Campora (Nikhef National institute for subatomic physics (NL))

Programming for Heterogeneous Architectures - exercise 2

• Daniel Campora (Nikhef National institute for subatomic physics (NL))

Announcements

Design patterns and best practices

• Daniel Campora (Nikhef National institute for subatomic physics (NL))

Programming for Heterogeneous Architectures - lecture 4 - GPU streams, pipelined memory transfers - Good practices: single precision, branchless, avoid register spilling, convert the problem - Reusable parallel design patterns with real-life applications - Under the hood: Warps, masked execution, floating point rounding

Practical vectorization

• Sebastien Ponce

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

Parallel and Optimised Scientific Software - exercise 1

• Sebastien Ponce

Optimisation of an existing, production grade large codebase - exercise 1

Parallel and Optimised Scientific Software - exercise 2

• Sebastien Ponce

Optimisation of an existing, production grade large codebase - exercise 2

Programming for Heterogeneous Architectures - exercise 3

• Daniel Campora (Nikhef National institute for subatomic physics (NL))

Programming for Heterogeneous Architectures - exercise 4

• Daniel Campora (Nikhef National institute for subatomic physics (NL))

Announcements

Summary and future technologies overview

• Andrzej Nowak

Technologies and Platforms - lecture 4 - Teaching program summary and wrap-up - Next-generation memory technologies and interconnect - Rack-sized data centres and future computing evolution

Exam

Closing Session