Fernando Rodriguez <frodriguez.developer <at> outlook.com> writes:
> > albeit in it's infancy. Naturally it's going to take a while to > > become mainstream useful; but that more like a year or 2, at most. > > The value I see on that technology for desktop computing is that we get the > GPUs for what they're made (graphics processing) but their resources go unused > by most applications, not in buying powerful GPUs for the purpose of offloading > general purpose code, if that's the goal you're better off investing in more > general purpose cores that are more suited for the task. I think most folks when purchasing a workstation include a graphics card on the list of items to include. So my suggestions where geared towards informing folks about some of the new features of gcc that may intice them to consider the graphics card resources in an expanded vision of general resources for their workstation. > To trully take advantage of the GPU the actual algorithms need to be rewritten > to use features like SIMD and other advanced parallelization features, most > desktop workloads don't lend themselves for that kind of parallelization. Not true if what openacc hopes to achived indeed does become a reality. Currently, you are most correct. Things change; I'm an optimist because I see what is occuring in embedded devices, arm64, and cluster codes. ymmv. > That > is why despite similar predictions about how OpenMP-like parallel models would > obsolete the current threads model since they where first proposed, it hasn't > happened yet. Yes it's still new technology, controversial, just like systemd, clusters, and Software Defined Networks. > Even for the purpose of offloading general purpose code, it seems with all the > limitations on OpanACC kernels few desktop applications can take advantage of > it (and noticeably benefit from it) without major rewrites. Off the top of my > head audio, video/graphics encoders, and a few other things that max out the > cpu and can be broken into independent execution units. You are taking a very conservative view of things. Codes being worked out now for clusters, will find their way to expand the use of the video card resources, for general purpose things. Most of this will occur as compiler enhancements, not rewriting by hand or modifying algorithmic designs of existing codes. Granted they are going to mostly apply to multi-threaded application codes. When folks buy new hardware, it is often a good time to look at what is on the horizon for computers they use. All I have pointed out is a very active area that benefits folks to review for themselves. I not pushing expenditures of any kind on any hardware. Caveat Emptor. James
