Hi Ben, In my opinion, put the scale test in a separate repo under https://github.com/openvswitch/ is a better choice. As you said OVS is quickly evolving and the review process is heavyweight and slow, workload is already high, put scale test into it can cause more overhead.
The review & merge process can be same as ovs repo, but not interleave with ovs's. BR Huang Lei On Tue, Apr 12, 2016 at 6:57 AM, Ben Pfaff <b...@ovn.org> wrote: > [dropping some CCs for people I know to be on ovs-dev] > > On Sun, Apr 10, 2016 at 08:45:38PM -0700, Han Zhou wrote: > > As requested by several folks and also mentioned in last week's ovn > > meeting, we would like this repo to be hosted publicly under > > https://github.com/openvswitch/ so that more people in the community can > > contribute and benefit from it. Or it could be a folder under ovs repo, > > e.g. ovn/scale-test. The goal is to be able to enable regular regression > > test for scalability of ovn, so that we are aware of impact introduced by > > changes. > > Which possibility do you like better? As I see it, here are some > advantages and disadvantages of each approach. > > Probably, it is easier to quickly evolve the scale-test in a separate > repo, because for something that is quickly evolving OVS (as you know) > has a relatively heavyweight and slow review process. I don't know > whether this is important. > > Probably, it will be easier to keep the scale test in sync with OVS and > OVN if it is in the same repository, because devs will be more likely to > see it as they work on OVN. I don't know whether this is important > either. > _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
