AMD-based GPU programmable platforms are centered around the Heterogeneous System Architecture (HSA). HSA is a cross-vendor set of specifications that allows the integration of CPUs and GPUs on the same bus, with shared memory and tasks. The HSA Foundation continues to be developed by AMD and many other members. AMD was one of the founding members of HSA.
HSA was developed with a programmer's perspective, to take care of the issue of prolonged data transfer between memories, especially when both CPUs and GPUs are involved (as is the norm when programming on CUDA or OpenCL).
