Automatic Fusion and Threading

IBM Streams 4.2 introduces two new features designed to make it easier to attain high performance: automatic submission time fusion and the automatic threading model. Automatic submission time fusion means that you can achieve a reasonable number of PEs for your system without having to manually fuse operators, and the automatic threading model means that you can take advantage of multiple cores per host, without having to manually place threaded ports in your application. Continue reading Automatic Fusion and Threading

Parallelized File Processing with the Parse Operator

Reading from an external source—such as the network or filesystem—is often a performance bottleneck. When source operators are the performance bottleneck for a streaming application, we have a tendency to blame the reading from the external source. But, that is not always the case. Particularly for large tuples which have many attributes, the actual performance bottleneck can be parsing. Continue reading Parallelized File Processing with the Parse Operator