Readme History/Event-based documentation is unclear

[jtramm]

Currently the history and event-based documentation in the readme is very terse. It is currently describing the differences in history vs. event-based for a full MC application, but does not describe how this is represented in XSBench. Need to add a pseudo code for event-based XSBench itself, and how this relates to the full application.

[williamfgc]

@jtramm thanks, I came across this. In particular, it's clear how history computation workloads will scale up with -p particles, but not in the event-based case. Is there an equivalent way to do the same in event-based? Asking as I'm studying workloads on GPU and event-based is the only supported mode due to its SIMD formulation. Any help is appreciated.

[jtramm]

Thanks for the reminder on this! Yes, we still need better documentation on this front. For the time being, you can increase the program runtime in event-based mode by increasing the number of lookups with the -l argument.

[williamfgc]

@jtramm thanks for the response. I can confirm I can increase the program runtime on Crusher, up to 2.2B lookups after that it dies. I'll open a new issue.