Computation Spreading: Employing Hardware Migration to Specialize CMP Cores On-the-Fly

Provided by: Association for Computing Machinery
Topic: Hardware
Format: PDF
In canonical parallel processing, the Operating System (OS) assigns a processing core to a single thread from a multithreaded server application. Since different threads from the same application often carry out similar computation, albeit at different times, the authors observe extensive code reuse among different processors, causing redundancy. Moreover, largely independent fragments of computation compete for the same private resources causing destructive interference. Together, this redundancy and interference lead to poor utilization of private microarchitecture resources such as caches and branch predictors.

Find By Topic