Revision as of 16:03, 13 June 2016 edit Fmadd (talk \| contribs) Extended confirmed users 10,468 edits →gather with sequential writes ← Previous edit		Revision as of 16:03, 13 June 2016 edit undo Fmadd (talk \| contribs) Extended confirmed users 10,468 edits →gather with sequential writes Next edit →
Line 47: ==== gather with sequential writes ==== An algorithm may be re-worked such that the reads are random [[gather (vector addressing)\|gather]] , whilst the writes are sequential. The disadvantage is that caching (and bypassing latencies) is now essential for reads, however it is easier to parallelise since the writes are guaranteed to not overlap. As such the gather approach is more common for [[gpgpu]] programming<ref name="gpgpu gems"/>, where the massive threading is used to hide read latencies.

Memory access pattern: Difference between revisions