Cell software development: Difference between revisions

Revision as of 01:34, 5 April 2008

Revision as of 19:21, 8 April 2008

Line 77:

==Local store exploitation==

Local stores can be aexploitedexploited using a variety of strategies.

Applications with high locality, such as dense matrix omputationscomputations represent an ideal workload class for the local stores in Cell BE.

<ref>{{cite web|url=http://www.research.ibm.com/people/m/mikeg/papers/2006_ieeemicro.pdf|title=Synergistic Processing in Cell's Multicore Architecture|date=[[2006-03]]}}</ref>

Streaming computations can be efficiently accomodatedaccommodated using software-pipelining of memory block transfers using a multi-buffering strategy.<ref>{{cite web|url=http://www.research.ibm.com/people/m/mikeg/papers/2007_ieeecomputer.pdf|title=An Open Source Environment for Cell Broadband Engine System Software|date=[[2007-06]]}}</ref>

The software cache offers a solution for random accesses.<ref>{{cite web|url=http://www.research.ibm.com/journal/sj/451/eichenberger.pdf|title=Using advanced compiler technology to exploit the performance of the Cell Broadband Engine architecture|date=[[2006-01]]}}</ref>

More sophisticated applications can use multiple stratwgiesstrategies for different data types.<ref>{{cite web|url=http://www.research.ibm.com/cell/papers/2008_vee_cellgc.pdf|title=Cell GC: Using the Cell Synergistic Processor as a Garbage Collection Coprocessor |date=[[2008-03]]}}</ref>

==Compiler-mediated parallelism==