Intel Making Progress On Their "mOS" Modified Linux Kernel Running Lightweight Kernels

[jabl]

I don't see why you need to modify the kernel for that purpose?
Linux already provides near perfect CPU isolation.
Squeezing the last few cycles would mean that the kernel should be completely unaware of the cores?
Ie. Viewing them as accelerators? Sure. But cumbersome.

CPU isolation still gets all the normal functionality of userspace (If you need to use syscalls for opening files etc).
Seems like a lot of work for 0.0001% more performance?

OS jitter is a well known problem for certain tightly synchronized massively parallel applications. LWK's are one way around the problem, though I suspect you could achieve much of the same with the various cpu isolation etc. approaches + cgroups.

[milkylainen]

I don't see why you need to modify the kernel for that purpose?
Linux already provides near perfect CPU isolation.
Squeezing the last few cycles would mean that the kernel should be completely unaware of the cores?
Ie. Viewing them as accelerators? Sure. But cumbersome.

CPU isolation still gets all the normal functionality of userspace (If you need to use syscalls for opening files etc).
Seems like a lot of work for 0.0001% more performance?

[phoronix]

Phoronix: Intel Making Progress On Their "mOS" Modified Linux Kernel Running Lightweight Kernels

For a while now Intel has been quietly been working on "mOS" as the "multi-OS" that is a modified version of the Linux kernel that in turn is running lightweight kernels for high-performance computing purposes...

http://www.phoronix.com/scan.php?page=news_item&px=Intel-mOS-Multi-OS-Linux