> > > > We didn't want to exec external python programs because that certainly > *does* have bad scalability, terrible error reporting facilities and > need to parse ill defined data formats from stdout, etc. It doesn't > magically solve the complexity, just moves it elsewhere where we have > less ability to tailor it to fit into libvirt's model. > > > BTW, to clarify, RMD is not wroten by python, it's golang, and it's not just a tool, it's a running service(agent) on the host, provided RESTful API by unix/TCP socket. and much more smarter (policy based) then static allocation. Support re-enforcement based on monitoring data (cache usage).
It aimed to be the only one interface for all who want to operation /sys/fs/resctrl, or even early kernel (4.10) which has no /sys/fs/resctrl (thought MSR). It not only provide VM but for all kinds of workload/cpu/containers. BR - Eli > Regards, > Daniel > -- > |: https://berrange.com -o- https://www.flickr.com/photos/ > dberrange :| > |: https://libvirt.org -o- > https://fstop138.berrange.com :| > |: https://entangle-photo.org -o- https://www.instagram.com/ > dberrange :| >
-- libvir-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/libvir-list
