Thanks Martin, You've done very important work and have reached the result of making real-world "multiprocess" software run on top of a cloud hypervisor. Not many people in the world can claim the same (at least not yet! ;)
To further dissect your results, seems like the "app-tools" toolchain continues to work, and your "rumprun" tool is proving itself for running the images as services. There were some minor hiccups, but the expectation of those will go away with a few more real world grade experiments -- if you take into account the lines of application code there were "practically" no platform bugs(!). One important thing we need to start teaching service writers is to not needlessly waste resources (e.g. 512kB stacks) given that low-resource environments are coming. Things have a natural tendency to consume ever resource available regardless of whether it is needed or not. Starting to send patches against this type of arbitrary resource consumption will hopefully eventually result in an attitude change. It's currently difficult to justify, given the "full OS" targets of the applications, multimegabyte stacks are not bugs per se, but we need to try. So, as we've been discussing on irc, we're a database short of a RAMP(*) stack ;) - antti *) my evil plan is to eventually have all software named after edibles ------------------------------------------------------------------------------ New Year. New Location. New Benefits. New Data Center in Ashburn, VA. GigeNET is offering a free month of service with a new server in Ashburn. Choose from 2 high performing configs, both with 100TB of bandwidth. Higher redundancy.Lower latency.Increased capacity.Completely compliant. http://p.sf.net/sfu/gigenet _______________________________________________ rumpkernel-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/rumpkernel-users
