Name
Title | ||
|---|---|---|
Dynamic task mapping of graphics processing applications on many-core architectures through stream rewriting |
Abstract | ||
|---|---|---|
Although modern graphics processing units (GPU) contain a large number of programmable shader cores, the focus on data parallelism and also the lack of efficient on-chip communication hinder the creation of custom graphics pipelines with arbitrary topologies. Based on the concept of stream rewriting, we propose a novel many-core architecture for graphics processing, which supports dynamic scheduling of recursively expandable task graphs and graphics pipelines. In particular, the tasks and their dependencies are encoded as a token stream, which is iteratively rewritten via pattern matching on multiple cores in parallel. The scalability of the proposed hardware architecture has been evaluated using an FPGA prototype. |
Year | DOI | Venue |
|---|---|---|
2015 | 10.1109/ESTIMedia.2015.7351763 | 2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia) |
Keywords | Field | DocType |
dynamic task mapping,graphics processing applications,many-core architectures,stream rewriting,GPU,programmable shader cores,data parallelism,on-chip communication,graphics pipelines,many-core architecture,dynamic scheduling,recursive expandable task graphs,graphics pipelines,token stream,pattern matching,scalability evaluation,hardware architecture,FPGA prototype | Graphics,Computer architecture,Graphics pipeline,Computer science,Graphics address remapping table,Parallel computing,Real-time computing,Data parallelism,Real-time computer graphics,General-purpose computing on graphics processing units,Shader,Texture mapping unit | Conference |
ISSN | Citations | PageRank |
2325-1271 | 0 | 0.34 |
References | Authors | |
9 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Lars Middendorf | 1 | 16 | 4.07 |
| Christian Haubelt | 2 | 796 | 68.77 |