HVM - A Massively Parallel Architecture
summary
Introduces HVM, a massively parallel computing architecture model intended to redefine how computers are designed. Describes the computational model’s goals and principles, emphasizing its parallelism-centric architecture and mathematical elegance as the guiding design heuristic.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:07Von Neumann limitations
- 2:12Hardware parallelism history
- 3:58Software parallelism challenges
- 7:32What is HVM
- 9:19Performance demo
- 13:30Lambda calculus basics
- 21:52Combinators and parallelism
- 29:19Full practical example
- 34:34HVM architecture optimizations
- 37:43Future language support
- 40:32Community next steps
- 41:38Q&A
related talks
01h 15mThe Evolution of Memory Technologies
14mFoundational HPC Systems for 2020 and Beyond
19mNuclei System Technology Releases UX1030H with Full Support for RVA23
01h 27m“Patterns and Techniques Used in the Houdini 3D Graphics Application ”
33mManaging High-Bandwidth Memory is a Parallel Scheduling Problem
21mVentana’s Second Generation RISC-V Processor for Data Center and Other High Performance Applications
01h 28mProcessing-in-Memory: Theory and Practice
21mEnabling the next level of scientific exploration