Global optimization of operand transfer fusion in heterogeneous computing
作者: Christoph Kessler
作者单位: Linköping University
论文集英文名称: Software and Compilers for Embedded Systems
来源数据库: Association for Computing Machinery
DOI: 10.1145/3323439.3323981
关键词: Computation In MemoryMemristorNon-Volatile MemoryNon-Von Neumann ArchitectureSimulator
原始语种摘要: We consider the problem of minimizing, for a dataflow graph of kernel calls, the overall number of operand data transfers, and thus, the accumulated transfer startup overhead, in heterogeneous systems with non-shared memory. Our approach analyzes the kernel-operand dependence graph and reorders the operand arrays in memory such that transfers and memory allocations of multiple operands adjacent in memory can be merged, saving transfer startup costs and memory allocation overheads.
全文获取路径: ACM  (合作)

  • 极小化 
  • transfer 转换
  • operand 运算对象
  • allocation 分配
  • saving 节省
  • costs 费用
  • problem 题目
  • kernel 
  • minimizing 
  • approach 
  • consider 仔细考虑