i.e.
I want to evaluate a thread scheduler framework,
mainly the performance in different scheduler policy(FIFO,RR,others) pluse some other algorithms
which events I should monitor?
or just the thread cost time? (maybe the Cache miss,CPI, ..will have some changes but I don't think they are impacted
by the scheduler policy?
mainly the performance in different scheduler policy(FIFO,RR,others) pluse some other algorithms
which events I should monitor?
or just the thread cost time? (maybe the Cache miss,CPI, ..will have some changes but I don't think they are impacted
by the scheduler policy?
I guess it depends on what exactly you want to get by changing the scheduler policy. Which benchmark you will be using matters as well.
I guess it depends on what exactly you want to get by changing the scheduler policy. Which benchmark you will be using matters as well.
such as wakeup latency(switch from a thread to another thread),not sure if it can measured by vtune?
and how much time spent on a certain thread(or the percentage) compared with thread use scheduler policy.
can CPI reflect theprogram effeciency at this situation?(same code bu different thread scheduler policy)
such as wakeup latency(switch from a thread to another thread),not sure if it can measured by vtune?
and how much time spent on a certain thread(or the percentage) compared with thread use scheduler policy.
can CPI reflect theprogram effeciency at this situation?(same code bu different thread scheduler policy)
I'm not an expert in the Linux kernel, but in my opinion you'd need to write your own driver for accurate measuring thread latencies.
As for the time spent in each thread, any tool based on instrumentation might help.
>can CPI reflect the program effeciency at this situation?(same code bu different thread scheduler policy)
If you are measuring efficiency of the real project, VTune sampling might help, but you will need to investigate to understand why CPI appeared better or worse depending on scheduler mode.
If you are measuring with small benchmarks (preferred way), usually it's better to count clock ticks within the program using appropriate API calls.