Evaluating the Importance of User-Specific Profiling


This paper examines common assumptions about user-specific profiling in profile-based optimization. We study execution profiles of interactive applications on Windows NT to understand how different users use the same program. The profiles were generated by the DIGITAL FX!32 emulator/binary translator system, which automatically runs the x86 version of Windows NT programs on NT/Alpha computers. We have found that people use the benchmark programs in different ways. These differences in program usage can have impact on the performance of profile-based FX!32 program translation and optimization, up to 9%.