Benchmarks regarding the Sather 1.1 compiler

Send questions and remarks to Claudio Fleiner.

On this page you will find links to several benchmarks that show the impact of five different optimizations of the new 1.1 Sather compiler. Results for the 1.0.8 Compiler are also available.

For more information about Sather check out the Sather Home page at the International Computer Science Institute.

All benchmarks presented on this page were run on either Solaris 2.4 or Solaris 2.5, using GCC version 2.7.1 on a Sparc 10 with 4 Processors.

The Benchmarks

For greatest flexibility and to select different options on how to view the results, use the form based interface. If your computer does not support forms, you can use the following pages:

Note that the parallel version of the program were run on a 4 processor shared memory Sparc running Solaris 2.5. There are no benchmarks available for distributed pSather yet.

The Optimization Options:

Description of the Tables

Each line of the tables shows the result when using some of the optimization options of the compiler. The options that are turned on for each line are marked with a blue dot. A red dot (only used in the impact tables) signifies that this line shows the difference between two runs of the compiler, once with this option turned off, and once when it is turned on.

To the right are the different measurments made. Each column is divided into up to 4 subcolumns (the standard tables have just the first two columns, but with the form interface you can turn each column on and off individually).

Note that the impact tables will only show the last two subcolumns, and instead of comparing the values with the standard compiler, they show the difference introduced by using the option marked with a red dot.

The Measurements