Currently, each timed iteration includes a single reader/writer instantiation, followed by a read/write of the input, followed by a reader/writer close. Add an option to allow the user to tell the benchmark to execute those steps repetitively within a single timed iteration. This would allow, for example, measurements of reader/writer instantiation performance.
Currently, each timed iteration includes a single reader/writer instantiation, followed by a read/write of the input, followed by a reader/writer close. Add an option to allow the user to tell the benchmark to execute those steps repetitively within a single timed iteration. This would allow, for example, measurements of reader/writer instantiation performance.