GPU Memory Hierarchy: Optimization

In course

Description

This lecture covers GPU memory hierarchy, including global, local, shared memory, and caches. It explains the CUDA processing flow, GPU optimizations, and control-flow divergence. The instructor discusses strategies to optimize algorithms for GPUs, exploit shared memory, and coalesce memory accesses. Various techniques to efficiently use parallelism and resources on GPUs are explored, such as reduction operations and addressing bank conflicts. The lecture concludes with a focus on scalability with array size and a summary of optimizing code for GPUs.

Instructor

Official source

https://mediaspace.epfl.ch/media/0_dtnxjksv

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Related lectures (68)

Related concepts (173)

Thread (computing)

In computer science, a thread of execution is the smallest sequence of programmed instructions that can be managed independently by a scheduler, which is typically a part of the operating system. The implementation of threads and processes differs between operating systems. In Modern Operating Systems, Tanenbaum shows that many distinct models of process organization are possible. In many cases, a thread is a component of a process.

Green thread

In computer programming, a green thread (virtual thread) is a thread that is scheduled by a runtime library or virtual machine (VM) instead of natively by the underlying operating system (OS). Green threads emulate multithreaded environments without relying on any native OS abilities, and they are managed in user space instead of kernel space, enabling them to work in environments that do not have native thread support. Green threads refers to the name of the original thread library for the programming language Java (that was released in version 1.

Memory access pattern

In computing, a memory access pattern or IO access pattern is the pattern with which a system or program reads and writes memory on secondary storage. These patterns differ in the level of locality of reference and drastically affect cache performance, and also have implications for the approach to parallelism and distribution of workload in shared memory systems. Further, cache coherency issues can affect multiprocessor performance, which means that certain memory access patterns place a ceiling on parallelism (which manycore approaches seek to break).

Screw thread

A screw thread, often shortened to thread, is a helical structure used to convert between rotational and linear movement or force. A screw thread is a ridge wrapped around a cylinder or cone in the form of a helix, with the former being called a straight thread and the latter called a tapered thread. A screw thread is the essential feature of the screw as a simple machine and also as a threaded fastener. The mechanical advantage of a screw thread depends on its lead, which is the linear distance the screw travels in one revolution.

General-purpose computing on graphics processing units

General-purpose computing on graphics processing units (GPGPU, or less often GPGP) is the use of a graphics processing unit (GPU), which typically handles computation only for computer graphics, to perform computation in applications traditionally handled by the central processing unit (CPU). The use of multiple video cards in one computer, or large numbers of graphics chips, further parallelizes the already parallel nature of graphics processing.