[Vm-dev] A fork in the road()
btc at openinworld.com
Fri Nov 29 13:28:22 UTC 2019
Sharing an interesting paper I bumped into..
> A fork() in the road
in particular Figure 1 - fork versus spawn performance
And something I didn't realise...
> a process using, DPDK with a kernel-bypass NIC , or OpenCL with a
> cannot safely fork since the OS cannot duplicate the process state on the
> This appears to have been a continuing source of bafflement
> among GPU programmers for a decade at least
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Vm-dev