Not everything with parallelism is suitable for GPUs [LWN.net]

Not everything with parallelism is suitable for GPUs

Posted Aug 10, 2023 22:01 UTC (Thu) by DemiMarie (subscriber, #164188)
In reply to: Using all of those cores by farnz
Parent article: Another round of speculative-execution vulnerabilities

GPUs have other limitations, though. For instance, the SIMT model means that GPUs are terrible at workloads with lots of non-uniform control flow. That isn’t a huge limitation for math or graphics, but it is a serious limitation for what I call “business logic” workloads, where a significant part of the problem is figuring out what to do next. This includes e.g. web applications, which have a huge amount of parallelism but lots of conditional branches and non-uniform memory accesses.

to post comments

Not everything with parallelism is suitable for GPUs

Posted Aug 10, 2023 22:08 UTC (Thu) by farnz (subscriber, #17727) [Link]

They're no more terrible at non-uniform control flow than CPUs are - in the worst case, you just use one SIMD lane per GPU core, get a much lower throughput, but still have the large number of threads. It's just that we look at GPUs differently to CPUs, so we see the slowdown from using only one SIMD lane as a big deal on a GPU, but we don't see it as a big deal that we only use scalar instructions on CPU cores with the ability to process 8 (AVX2) or 16 (AVX-512) 32-bit values in parallel, despite the fact that this is the same class of slowdown.