Performance measurements

The performance measurements in this information were made with COBOL 6.4 on a z16™ system except where otherwise noted. The programs used were batch-type (noninteractive) applications. Unless otherwise indicated, all performance comparisons that are made in this information are referencing CPU time performance and not elapsed time performance.

Note: Except where otherwise noted, performance comparisons in this document are measured using microbenchmarks. Each microbenchmark is designed to highlight the compiler's performance improvement in a specific area. Microbenchmarks are made to amplify the performance of instructions generated for a specific COBOL statement and help the compiler team choose the optimal set of instructions for a specific COBOL statement. They are not indicative of the overall performance of a real-world application and should not be used as a measure or indicator of the overall performance of a real-world COBOL application. Real programs typically use a mix of features. A program will only see a performance improvement from improvements in the compiler if the program spends a significant percentage of its time executing code that is affected by those improvements.
Note: Performance results for customer applications will vary, depending on the source code, the compiler options specified, and other factors.