Conference paper Open Access

Global optimization of operand transfer fusion in heterogeneous computing

Kessler, Christoph

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.

Files (5.4 MB)
Name Size
global_optimization.pdf
md5:f658ca9e00a2c91479568f39439dc743
5.4 MB Download
36
5
views
downloads
Views 36
Downloads 5
Data volume 27.2 MB
Unique views 35
Unique downloads 5

Share

Cite as