Download Now Free registration required
Manycore processors with wide SIMD cores are becoming a popular choice for the next generation of throughput oriented architectures. The authors introduce a hardware technique called "Diverge on miss" that allows SIMD cores to better tolerate memory latency for workloads with non-contiguous memory access patterns. Individual threads within a SIMD "Warp" are allowed to slip behind other threads in the same warp, letting the warp continue execution even if a subset of threads are waiting on memory.
- Format: PDF
- Size: 230.2 KB