GPU Performance

1. An adaptive performance modeling tool for GPU architectures §

Sara S. Baghsorkhi et al., PPoPP 2010

• progress: 40%, link

1.1. 主要工作 §

• 设计了一个模型，用于提供performance information to an auto-tuning compiler（这个compiler是什么？），assist it in narrowing down the search （辅助在搜索空间中进行剪枝）。 “We introduce a abstract interpretation of a GPU kernel, work flow graph, based on which we estimate the execution time of a GPU kernel”

• 如何做到不依赖于GPU architecture/high-level prgramming interface

• PDG: 用于性能评估。for performance evaluation, framework to represent control and data dependences for each program operation

• work flow graph

• key factors

1.2. Term §

• SPMD: Single-Program Multiple-Data

• SIMD: Single-Instruction, Multiple-Data

• thread granularities

  NVIDIA	thread-block
  ATI	groups
  OpenCL	work-groups

  Threads within a thread-block are grouped into warps

• SM: streaming multiprocessor

• PDG: program dependence graph

• WLP: Warp-level parallelism (threads within a thread-block are grouped into warps)

• DLP: data-level parallelims

• TLP: Thread-level parallelism

• ILP: instruction-level parallelism

• NUM_blocks: the number of active thread-blocks on a streaming multiprocessor

1.3. Performance Model §

• 多个层面上的并行
  • kernel level -> warp level -> thread level

graph TD;

	kernel[kernel]
	warp[warp level:<br />GPUs attempt to reduce memory latency by exploiting the data-level parallelism]
	thread[thread level:<br />instruction-level parallelism can still improve performance by partially covering intra-warp stalls]
	
	kernel-->warp
	warp-->thread
	

1.4. Work flow graph §

• What to take into consideration:
  • SIMD pipeline latency, global memory latency

1.5. Measurement §

• 执行时间

---