Address-Branch Correlation: A Novel Locality for Long-Latency Hard-to-Predict Branches

DMCA

Address-Branch Correlation: A Novel Locality for Long-Latency Hard-to-Predict Branches

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Hongliang Gao , Yi Ma , Martin Dimitrov , Huiyang Zhou

Citations:

11 - 0 self

BibTeX

@MISC{Gao_address-branchcorrelation:,
    author = {Hongliang Gao and Yi Ma and Martin Dimitrov and Huiyang Zhou},
    title = {Address-Branch Correlation: A Novel Locality for Long-Latency Hard-to-Predict Branches},
    year = {}
}

OpenURL

Abstract

Hard-to-predict branches depending on longlatency cache-misses have been recognized as a major performance obstacle for modern microprocessors. With the widening speed gap between memory and microprocessors, such long-latency branch mispredictions also waste substantial power/energy in executing instructions on wrong paths, especially for large instruction window processors. This paper presents a novel program locality that can be exploited to handle long-latency hard-topredict branches. The locality is a result of an interesting program execution behavior: for some applications, major data structures or key components of the data structures tend to remain stable for a long time. If a hard-to-predict branch depends on such stable data, the address of the data rather than the data value is sufficient to determine the branch outcome. This way, a misprediction can be resolved much more promptly when the data access results in a long-latency cache miss. We call such locality addressbranch correlation and we show that certain memoryintensive benchmarks, especially those with heavy pointer chasing, exhibit this locality. We then propose a low-cost auxiliary branch predictor to exploit address-branch correlation. Our experimental results show that the proposed scheme reduces the execution time by 6.3 % (up to 27%) and energy consumption by 5.2 % (up to 24%) for a set of memory-intensive benchmarks with a 9kB prediction table when used with a state-of-art 16kB TAGE predictor. 1.

Keyphrases

address-branch correlation long-latency hard-to-predict branch novel locality hard-to-predict branch speed gap longlatency cache-misses interesting program execution behavior data structure stable data modern microprocessor prediction table energy consumption key component data value locality addressbranch correlation long time novel program locality heavy pointer chasing substantial power energy memory-intensive benchmark long-latency cache miss branch outcome major performance obstacle execution time low-cost auxiliary branch predictor certain memoryintensive benchmark long-latency hard-topredict branch long-latency branch mispredictions large instruction window processor wrong path tage predictor major data structure experimental result data access result

DMCA