An interesting topic to research, but a shame they botched the statistics (and benchmarking in general). These results are not believable.
For information on system calls see: POSIX Abstractions in Modern Operating Systems
The issue of hardware variability was not covered: http://shape-of-code.coding-guidelines.com/2015/02/24/hardware-variability-may-be-greater-than-algorithmic-improvement/
and there are lots of OS specific variability issues; they should have at least rebooted multiple times and rerun the benchmark.
As for picking the top 5 out of 10,000 measurements. Shoot that man!
Thank you for this. I saw this link on another site and only skimmed it so I was kind of baffled that the (claimed) results could be the case.
An interesting topic to research, but a shame they botched the statistics (and benchmarking in general). These results are not believable.
For information on system calls see: POSIX Abstractions in Modern Operating Systems
The issue of hardware variability was not covered: http://shape-of-code.coding-guidelines.com/2015/02/24/hardware-variability-may-be-greater-than-algorithmic-improvement/
and there are lots of OS specific variability issues; they should have at least rebooted multiple times and rerun the benchmark.
As for picking the top 5 out of 10,000 measurements. Shoot that man!
Thank you for this. I saw this link on another site and only skimmed it so I was kind of baffled that the (claimed) results could be the case.