Will try to do updates soon, and I owe someone two articles (sorry!). Add to this fighting off a cold … not a happy camper.
Basically we are building an 8x JackRabbit JR5 storage cluster right now. I’ve caught a problem in Tiburon, our OS loader, in the process, and am fixing it. Tiburon is all about providing a very simple platform to enable PXE (and/or iSCSI) booting OSes to make installation/support simple. It uses our finishing scripts, which take a basic OS load and finish it, or polish it for the task at hand.
The philosophy for Tiburon is to install just enough OS so that you can post-install configure the machine(s) programmaticly. We don’t force this into the base OS installer as some others do, as, to be quite frank, most of the OS installers out there are pretty darned brittle. Think how forgiving anaconda is. Which is why you absolutely want to minimize your exposure to anaconda. Get what you need down on the disk, and then start from a known working node, and do all the configs you need from there.
We’ve done stateful and stateless (diskless) loads, and it can handle iSCSI loads as well (really demonstrates the power of the storage with this). This particular system is better setup stateful, so we will do it this way.
More very soon. I expect first baseline performance tests Tuesday-ish time period.
And paraphrasing my European colleagues, then we will test it in anger. Io-bm is part of this testing regimen, as is fio, bonnie, iozone, etc.