On Fri, Feb 19, 2016 at 05:01:19PM +0700, Tinker wrote: > On 2016-02-17 01:01, [email protected] wrote: > .. > >Why do you think you need to build such a device? Why don't you buy it? > > > >(Dell PowerEdge VRTX, HP hyper converged, etc) > > Colocation requires rack servers, but thanks for thinking about it. > > >Some important things: > > > > - what is the purpose of this collection of clients, servers, > >networks and software? > > - who will judge, and how will they judge, the effectiveness of it? > >How fast/correctly it performs not just reliability > > - what is their budget? > > - how much time will they give you? > > - how will you spend your time? > > - how will you prove to yourself that you have finished? How can you > >prove to your users/customer that it works? > > > I feel the general takehome from this conversation and Nick Holland's > suggestions, is that really anything might break and everything needs to be > set up to handle that. > > So like, all of > > * data integrity verifications, > * checksumming of everything, > * automated routines to take a node out of use in case of IO slowdown or > failure, or any other error, and > * live syncing of important stuff to another datacenter for the case o > power failures > > need to be in place for there to be any real data integrity + QoS > guarantees. > > > Thanks!
You keep forgetting "planning for the manual stuff that will need to be done when all automated stuff fails".
