Computing Atlas

How Computing Was Built
Sign In
Text size
100%
Theme
Algorithm

Collective operation

Parallel and Distributed Algorithm

Collective operations are standard building blocks for structuring the communication and interaction patterns used inside SPMD, single program multiple data, algorithms in parallel programming. Rather than every process in a parallel computation exchanging data through ad hoc point to point messages, a collective operation defines a common pattern, such as distributing data from one process to many or combining data from many processes into one, that many parallel algorithms rely on repeatedly. Because these patterns recur so often, there is a strong incentive to implement them efficiently rather than leaving each program to reinvent them. The Message Passing Interface, MPI, is a widely used standard that provides working implementations of these collective operations for parallel programs to call directly.

Connections

In Field

Source Collective operation (Wikipedia)

Uses Design Technique

Source Collective operation (Wikipedia)
Sources
Collective operation (Wikipedia)
  • Lead section
    Collective operations are building blocks for interaction patterns, that are often used in SPMD algorithms in the parallel programming context.
  • Uses Design Technique: Divide and Conquer, Article body
    Since an implementation by means of a sequential for-loop with p − 1 {\displaystyle p-1} iterations becomes a bottleneck, divide-and-conquer approaches are common.
  • In Field: Algorithms and Complexity Theory, Lead sentence
    Collective operations are building blocks for interaction patterns, that are often used in SPMD algorithms in the parallel programming context.
View the Source
Comments (0)
No comments yet. Be the first to share a thought.
Reader Challenges (0)
No disputes yet. Spotted an error or a better source? Open the first one.