The development of a high-performance parallel system or application is an evolutionary process -both the code and the environment go through many changes during a program's lifetime -and at each change, a key question for developers is: how and how much did the performance change? No existing performance tool provides the necessary functionality to answer this question. This paper reports on t...

Information about a software system’s execution can help a developer with many tasks, including software testing, performance tuning, and program understanding. In almost all cases, this dynamic information is reported in terms of source-level constructs, such as procedures and methods. For some software engineering tasks, source-level information is not optimal because there is a wide gap betw...

Systems should be self-predicting. They should continuously monitor themselves and provide quantitative answers to What...if questions about hypothetical workload or resource changes. Self-prediction would significantly simplify administrators’ decision making, such as acquisition planning and performance tuning, by reducing the detailed workload and internal system knowledge required. This pap...

Network performance in tightly-coupled multiprocessors typically degrades rapidly beyond network saturation. Consequently, designers must keep a network below its saturation point by reducing the load on the network. Congestion control via source throttling—a common technique to reduce the network load—prevents new packets from entering the network in the presence of congestion. Unfortunately, ...

We present a framework which uses POET, an interpreted code transformation language, to effectively combine programmable control from developers, advanced optimizations by compilers, and flexible empirical tuning of optimizations to achieve portable high performance for scientific computing. We have extended ROSE, a C/C++/Fortran source-to-source compiler, to automatically analyze scientific co...

The nascent and rapidly evolving state of parallel systems often leaves parallel application developers at the mercy of ineecient, innexible operating system software. Given the relatively primitive state of parallel systems software, maximizing the performance of parallel applications not only requires judicious tuning of the application software, but occasionally, the replacement of speciic s...

Performance tuning in modern database systems requires a lot of expertise, is very time consuming and often misdirected. Tuning attempts often lack a methodology that has a holistic view of the database. The absence of historical diagnostic information to investigate performance issues at first occurrence exacerbates the whole tuning process often requiring that problems be reproduced before th...

Recent conversion from a Mainframe environment to AIX/UNIX presented many interesting challenges. Learning the AIX/UNIX operating system from manuals alone was not enough. Fortunately, many of the fundamental performance tuning principles learned in the mainframe environment applied well in the sense that you are still balancing CPU, memory, and I/O. Encountering real world system performance i...

This paper discusses the suitability of the Design of Experiments (DOE) methodology for storage performance tuning. The DOE estimates the main and interaction effects of three major storage factors viz, Cache Partition Size (CPS),LUN Strip Size (LSS) and Cache Segment Size (CSS) settings of an enterprise level storage system on the overall storage performance. Other factors influencing performa...

With the growing technology, most of the focus is on spatial data mining. Spatial data mining refers to the extraction of unknown and unexpected information from spatial data sets of massive, high dimensionality and complex spatial databases. Spatial data mining is a process to discover related knowledge, potentially constructive and high utility patterns embedded in geographic Information. App...