Testbed hardware status 03/07/2000

• Have all components of 40 PCs with the exception of processors, DNARD racks, and floppy drives. Expect the remaining hardware by, at the earliest, by ??? with the exception of the DNARD racks which may not be in by then. The DNARDS are not critical initially.
• Experiencing CPU backorder problems. Steve came up with another processor idea: order 30 466MHz PPGA celerons w/ adapters, put 10 in group's desktop machines and take out the 20 350s currently in those machines. (Jay wants 40.)
• Floppies probably won't hold up initial testbed as they're not critical. It will cause more work on Mike and Steve's part, though: about 1.5 days estimating 20 minutes installation per machine. Floppies are on order from solutions4sure..
• Management hubs/switches. Ordered NetGear DS516 dual speed hubs: rack mountable, 16 ports per, full 10/100 connectivity. Ordered from solutions4sure (with floppies).
• Rack configuration. Mike is going to make up some xfig drawings. Once they settle on a rack and wiring config, I volunteered to call the wiring folks for info.
• Wiring.While we could wire a limited number of nodes ourselves (10?), wiring of the full testbed has to happen before the testbed is maximally useful. Jay has references from Dave for people who will, at a minimum, make cables. Not sure if they will do a big installation job. Some thought given to telecommunications but concern about (1) backlog and (2) cost. If wiring is not done correctly, we will be paying the price for a long time to come. As an alternative, perhaps we could hire a professional to come out and show us the best way to string the wires.

  Steve has found that the rails on the racks are very difficult to work with because they're easily misaligned. Note that this adds time to testbed setup.

• DNARD racks. Might be shipping from the manufacturer to our vendor (buy.com) on the 6th, but this is unconfirmed.