On 4/12/20, Leo Gaspard <[email protected]> wrote: > Hello, > > I come wondering: the FAQ claims that seL4 is “the world's fastest > microkernel”. However, I can only find benchmarks of just seL4, without > comparison to other operating systems running on the same machine. > > Would you happen to have such comparative benchmarks? I'd be > particularly interested by something against L4Re, as it'd probably > allow me to guess at the performance of something similar to L4Linux > ported to seL4. > Kernel performance is not the only factor when it comes to performance of microkernel OSes. An OS with a heavyweight structured RPC transport layer with mandatory dynamic marshaling and lots of vertical separation of subsystems into multiple processes is likely to be slower than an OS with an unstructured stream transport layer and subsystems that are mostly single processes, even if both are built on the same high-performance kernel. It seems that most microkernel OSes follow the former model for some reason, and I'm not sure why.
_______________________________________________ Devel mailing list [email protected] https://sel4.systems/lists/listinfo/devel
