000135587 001__ 135587
000135587 005__ 20190331192648.0
000135587 0247_ $$2doi$$a10.1145/1064212.1064278
000135587 022__ $$a01635999
000135587 037__ $$aCONF
000135587 245__ $$aTurboSMARTS: Accurate microarchitecture simulation sampling in minutes
000135587 269__ $$a2005
000135587 260__ $$c2005
000135587 336__ $$aConference Papers
000135587 520__ $$aRecent research proposes accelerating processor microarchitecture simulation through statistical sampling. Prior simulation sampling approaches construct accurate model state for each measurement by continuously warming large microarchitectural structures (e.g., caches and the branch predictor) while emulating the billions of instructions between measurements. This approach, called functional warming, occupies hours of runtime while the detailed simulation that is measured requires mere minutes. To eliminate the functional warming bottleneck, we propose TurboSMARTS, a simulation framework that stores functionally-warmed state in a library of small, reusable checkpoints. TurboSMARTS enables the creation of the thousands of checkpoints necessary for accurate sampling by storing only the subset of warmed state accessed during simulation of each brief execution window. TurboSMARTS matches the accuracy of prior simulation sampling techniques (i.e., ±3% error with 99.7% confidence), while estimating the performance of an 8-way out-of-order superscalar processor running SPEC CPU2000 in 91 seconds per benchmark, on average, using a 12 GB checkpoint library.
000135587 700__ $$aWenisch, Thomas F.
000135587 700__ $$aWunderlich, Roland E.
000135587 700__ $$0243553$$g177958$$aFalsafi, Babak
000135587 700__ $$aHoe, James C.
000135587 773__ $$j33$$tPerformance Evaluation Review$$k1$$q408-409
000135587 8564_ $$uhttps://infoscience.epfl.ch/record/135587/files/sigmetrics05.pdf$$zn/a$$s234162
000135587 909C0 $$xU11837$$0252231$$pPARSA
000135587 909CO $$ooai:infoscience.tind.io:135587$$qGLOBAL_SET$$pconf$$pIC
000135587 937__ $$aPARSA-CONF-2005-003
000135587 970__ $$a2006099727517/PARSA
000135587 973__ $$rREVIEWED$$sPUBLISHED$$aOTHER
000135587 980__ $$aCONF