This week: Regional Network budget modeling, end-of-year expenditure delay, CATalyst fault-tolerance testing, Dell contract pricing, VACC Director interviews, University Policy web site review, policy on email as a standard method for communicating with students, F&A division reorganization
It was a quiet week in Lake Woebegon. [No, I haven't seen the movie yet, but I'm glad it got great reviews. Sounds fun.]
I spent a good deal of time this week writing and editing spreadsheets and memos. [But now that I've written the rest of this and am back reviewing, I see that I spent a lot of time in meetings, too! ]
We’re trying to finish up the contract terms and analysis for several critical network connections. We have a number of point-to-point connections in place now (CATalyst office, Colchester Research, Rubenstein Lab) and some critical ones we need to develop (Tech Park, new office space — site not yet determined). Taking that longer-term view, the University would be best and most economically served by contracting with TelJet for dark fiber on the ring they’re building around Burlington. We usually refer to that as the “fiber ring” or “ring of fiber” (the latter a variation that Johnny Cash fans will appreciate), but I’ve been trying to convert the nomenclature to “regional network” in anticipation that we might develop this into something broader over the next few years (to include regional wireless, for example, and we would have included connectivity to Montpelier had the Vermont College project gone forward). So a good deal of time this week was devoted to analyzing the proposals and creating a 10-year cost-of-use analysis for various scenarios, then documenting that for the VP and the Budget Director. Not yet finished, but should be within a few days.
In May, VP Gower directed that a summary of unexpended end-of-year divisional funds be reported and that planned expenditures against them be held until the Division’s end-of-year financial status could be determined. Those summaries were submitted last week, and this week his senior staff met to discuss disposition of the funds. Several offices had major unplanned and unbudgeted expenditures, and the fund balances from the rest of us (CIT was a major positive contributor) will be used to cover those debts. While there appear to be funds remaining to be allocated, for now, at least, the planned purchases of server and public lab equipment from “budget dust” will be deferred until July, when the actual balances are known. We (CIT) have some critical purchases that we really need to make now, though, and I will continue to emphasize that we need to make those purchases now and not defer until July.
In the course of that discussion, we realized that we probably don’t want to replace the equipment in Waterman Lab this year, even if funds are available. With the anticipated move of CIT out of Waterman and uncertainty about where the Lab might be located in the future, it makes more sense to extend the lifetime of the equipment in the Lab for one more year (four years — longer than I’d like for a public lab) rather than develop new images, replace equipment, and then have to move it all or store it within 12 months.
The CATalyst team continued to work on the fault-tolerance and fail-over testing for the Peoplesoft implementation. Most of you know that the hardware and software system was designed to be highly redundant so that single failures wouldn’t disrupt work and catastrophic failures would result in transfer of work to a live, secondary system. We’ve been testing that for about 2 months now, and it’s not as easy to make this work as it might seem. This week, the DBA and systems teams that have been working on this made good progress in resolving some of the problems they’ve been having, and the systems now survive the UPS failure tests. We will soon have — if we don’t already — the world’s most expert team in Oracle RAC/DataGuard fault-tolerance and fail-over testing. Kudos to Norman, Nancy, Ed, Michael Grundhauser, Kent, Mike Austin, and Marianne Incerpi for their persistence and expertise. This is the sort of unsung heroic effort that typifies what CIT does: if we do it right, no one else in the UVM community ever knows just how hard we had to work to make the infrastructure rock-solid. But we know. Thanks, gang!
This work is doubly important since we plan to implement similar kinds of fault-tolerance support into the Banner system. Our experience with Peoplesoft now should make that implementation much easier.
Prompted by Andy Gingras’ and Sue Parizo’s very helpful analysis, we’ve had some discussions with Dell about the levels of their discounts to UVM relative to both their historical discounts and their prices to other customers. We expect to see lower costs as a result of those discussions. Again, the UVM community won’t know what had to be done to get the favorable pricing they’ll see, but you should all know. Thanks, Andy & Sue!
We had on-campus interviews with two candidates for the VACC (Vermont Advanced Computing Center, http://www.uvm.edu/cit/IT-news/2005summer/?Page=announce/vacc.html&SM=announce/annmenu.html ) Director position. We have two more within the next week, then it’ll go to the VP of Research for decision. The interviews included very technical discussions about architectures and strategies for high-performance computing, and I was reminded, yet again, of just how well UVM has been served by CIT. It’s clear that the architecture and systems decisions made here are comparable to those made by international leaders in the field. Kudos to Jim Lawson and Mike Austin for their work on the VACC cluster — the results are a credit to the University and an attraction for the VACC Director candidates.
The CATalyst Finance System implementation continues to progress toward its July 1 go-live. There is much work to be done yet, particularly testing of interfaces with other components at UVM, but we anticipate that we will cut over promptly according to schedule. A number of training sessions are being offered now to prepare staff to use the new system. Like all ERP implementations, we should expect some surprises as we start up; and like all ERP implementations, we’ll find short-term work-arounds and long-term process changes to deal with those surprises. The CATalyst staff is now working long hours during the week and on weekends to make sure the startup transition is as smooth as possible. This is the phase in which ERP implementation staff experience the greatest stress, so words of encouragement and support would most likely be very much appreciated! (These are my words of encouragement and support: Thanks Gang! Hang in there!)
The President has directed that UVM review and revise its documented policies (see President Fogel’s memo linked from the PPG site, http://www.uvm.edu/~uvmppg/ppg/ ). We in CIT (Keith, Bess, Dean, Sharon, me) have reviewed the policies for which we are responsible and will be submitting just those that fit as University policies. We’ll need to link our departmental policies, procedures, and forms pages to the University pages that set the institutional policy framework. This is a Good Thing in that it provides us an opportunity to review our policies, update them, discard the ones that don’t make any sense any longer, and fill in gaps that might have developed. Along the way, I’m hoping we’ll rethink the way the PPG site is organized and present it in a couple of different ways for ease of navigation.
On a related issue, I met with representatives of Communications, the Provost, and the Dean of Students to discuss electronic communication with students. We will be drafting a policy stating that the University will conduct its official business with students via email and that it will be the students’ responsibility to maintain that communication path should they choose to use something other than their UVM email account as their email home. This will be a companion policy to a similar one for faculty and staff that had been previously approved in principle by the Provost’s Senior Leadership committee.
The F&A Associate VPs have continued to meet to discuss division-wide reorganization. Our next step is to identify customers and stakeholders of each functional area so that we can consolidate questions about service needs and expectations for focus group interviews. I think the organizational model I’m developing for CIT will be consistent with the evolving divisional model.