Changes

Jump to: navigation, search

GPU621/False Sharing

776 bytes added, 02:18, 26 November 2021
Cache Coherence and Cache Line
[[File:cache_coherence.png]]
 
Each processor has their own local cache. When data is needed, a fixed block of memory is transferred to the cache; this block is known as a cache line.
 
However, if they each modify their own copy, how do we know which one correct?
 
We can’t store all the different versions of the data and or just pick a random one to be the correct version, this would create havoc in our computer systems.
The answer is we need cache coherence which is the uniformity of shared resource data that ends up stored in multiple local caches. In other words, we must keep all the local caches synchronized. The challenge of doing so is known as the Cache Coherence Problem. To solve this problem, multiprocessor systems rely on coherence protocols to manage and maintain cache coherency.
== False Sharing ==
83
edits

Navigation menu