AMD Heterogeneous System Architecture Powers High Compute Growth - Allows WorkStation Integration

HSA is one of the more interesting architecture implementations out at that place and holds great hope for the future of computing. Currently GPUs and CPUs are designed and implemented as completely split up processors which, though has its advantages, is but non future proof anymore - something HSA aims to modify. AMD recently held its Future of Compute conference and talked in length about HSA and how it can be used to combine workstation types into a single station. We received the consummate slide deck courtesy of DG Lee over at IYD.KR

HSA GPU Comparison A slide from Future of Compute Result showing the departure in growth of a CPU and an HSA based APU - AMD Public Domain

AMD talks about HSA in Hereafter of Compute - Workstation Integration and the Potential of Exponential Growth

HSA, is something that is a bit difficult to explicate, merely simply put it is the unification of the CPU and GPU with both processors sharing memory, bus and scheduling amongst other things. AMD's APUs are examples of HSA based processors (one of the major differences betwixt Intel CPUs with iGPUs and AMD APUs past the style). Nevertheless, the major trouble that currently exists is that while HSA is a stride frontward in the correct management, the ecosystem of the same is severely lacking. There is an almost complete lack of HSA compatible software on the (commercial) market and HSA hardware is mostly made in the low-mid end segment. Take for instance the AMD A10-7850K or Kaveri APU, which on HSA enabled work loads actually beats out the Intel i5 4670k (HSA benchmarks). Which is nothing short of astonishing, considering that steamroller doesn't compare at all to Intel Haswell in terms of architectural power; that ladies and gents is the power of HSA correct there.

AMD's talk about HSA in the Future of Compute Effect primarily explains how information technology can be used to merge three dissimilar workstation types into one single station. The three types that are mentioned (quoting marketing buzzwords hither) are Acquire, Simulate and Render. Basically, the Larn type means a workstation for storing and accessing vast amounts of raw data. Since they are talking about 4k and 8k RAW video, information technology basically ways that in such computers, yous have configurations designed for quick access and storage (SSD-HDD combinations). For Simulations you usually have TESLA enabled workstations coupled with RAID configurations and finally for rendering you lot unremarkably go with a render farm (for CPU rendering) or Nvidia Quadro/FirePro setup for GPU rendering. The reason for splitting is because these desktops are optimized for their own specific purpose which varies from high accurateness to double precision performance to high io etc. With HSA, the unified processor can do nearly everything that these tasks require.

With the GPU treatment parallel tasks and the CPU treatment serial tasks with a unified architecture in place, it'due south about processor nirvana. Another very big advantage of HSA is the fact that even though CPU engineering has slowed downwardly to a very stagnant pace, the fact that the unified processor has a GPU inside information technology ways that its growth curve volition exist exponentially bigger than a CPU. Once over again, something very desirable for the compute market. Now simply if nosotros had an bodily ecosystem in place, we could get some proper progress.