This paper describes an algorithm for procedure placement, one type of code placement, that signicantly differs from previous approaches in the type of information used to drive the placement algorithm, and gathers temporal-ordering information that summarizes the interleaving of procedures in a program trace.
This paper focuses on the operating system support that is required to collect and manage profile information on an end-user's workstation in an automatic, continuous, and transparent manner and shows that Morph can use statistical profiles to improve application performance.
Proceedings 22nd Annual International Symposium…
1 May 1995
TLDR
A framework is presented that categorizes branch prediction schemes by the way in which they partition dynamic branches and by the kind of predictor that they use, to show how a static correlated branch prediction scheme increases branch bias and thus improves overall branch prediction accuracy.
This paper describes the notation and the results of the application of genetic programming to the design of branch and indirect jump predictors, and provides a simple algebraic-style notation that allows one to describe a wide variety of predictors in a uniform way.
23rd Annual International Symposium on Computer…
15 May 1996
TLDR
It is found that user-only traces yield accurate prediction results only when the kernel accounts for less than 5% of the total executed instructions, and that flushing the branch history state at fixed intervals does not accurately model the true effects of user/kernel interaction.
An algorithm for procedure placement, one type of code-placement algorithm, that significantly differs from previous approaches in the type of information used to drive the placement algorithm is described, that gathers temporal ordering information that summarizes the interleaving of procedures in a program trace.
This paper describes the experience with a prototype system for continuous and automatic generation of profile data, and finds that the overhead of this prototype monitor is negligible, making it practical to collect profile information continuously for all activity on the system.