BSC Training Course: RISC-V principles for understanding how to freely develop new solutions

Course/Event Essentials

Event/Course Start
Event/Course End
Event/Course Format
In person

Venue Information

Country: Spain
Venue Details: Click here

Training Content and Scope

Scientific Domain
Technical Domain
Level of Instruction
Intermediate
Sector of the Target Audience
Research and Academia
Language of Instruction

Other Information

Supporting Project(s)
EuroCC2/CASTIEL2
Event/Course Description

RISC-V is an open instruction set standard which is experiencing extraordinary growth all over the world in numerous areas of focus ranging from HPC & ML to the data center to embedded computing. The standardization activities are community driven by expert members (from industry, academia and individuals). BSC is one of the community members that is contributing to the ecosystem. This course is an opportunity to get familiar with technical aspects of the standard through a combination of lectures and hands-on sessions.

This course identifies topics that are both fundamental to computer architecture and relevant to the design of RISC-V based solutions of the future. The emphasis is always on insights that will be useful to the (under)graduate student, whether he/she goes on for a PhD or joins a software or hardware development team. We will deal with principles, trade-offs, and implementation details related to the RISC-V standard. Along the course, the students will get familiar with some layers of the RISC-V software and hardware stacks using a learning-by-doing methodology. This will provide a mechanism to the students to understand the RISC-V ecosystem, BSC know-how and activities related to RISC-V, and explore the potential of the ISA to freely develop new technologies..

Topics will cover aspects like: how to boot an operating system, how to deploy a container and execute an application, how to tune an algorithm for improving its performance through the addition of custom instructions, and finally, how to improve the performance of an HPC application by taking advantage of the RISC-V Vector extension.