Peter: Please file a bug/rfe in bugzilla to track the problem with AI performance improvement with your data.
Peter Tribble wrote: > My initial impressions of AI performance were that it was significantly > slower than jumpstart. Given that my major criticism of jumpstart is > how slow it is, that's a problem. > > (And I know how to speed jumpstart up quite a bit - simply use less > computationally intensive compression schemes for the packages. > Really, to get it up to scratch would need 2-3X improvement in what > Jumpstart can currently give.) > > So an actual test: > > Test client is a SunBlade 1500 with 1G RAM. > > AI, with the repo image lofi mounted: 71 minutes to install packages. > SXCE Jumpstart, off my usual (busy) jumpstart server: 44 minutes to > install packages > AI, repo copied to disk: 70 minutes to install packages. > > That's not brilliant, but the SXCE install is quite a lot bigger (7.5G with > 300K files compared to the 2.9G in 110K files that the AI gives me). > Calculating rates: > > AI: 750K/s, 27 files/s > Jumpstart: 2.9M/s, 110 files/s > > That's a real problem. I imagine this is IPS, but we really need a factor > 5-10 improvement. > > The overall time in addition to that is approximately the same in both cases > - about 15 minutes covering the two boots required. Although WANboot > does occasionally throw a fit and take an extra 30-45 minutes. > >
