Report of Discussion

Tuesday, 14 May 2013

A Discussion was held in the Senate-House. Senior Pro-Vice-Chancellor Professor Steve Young was presiding, with the Registrary, the Senior Pro-Proctor, a Deputy Proctor, and five other persons present.

The following Reports were discussed:

Second-stage Report of the Council, dated 22 April 2013, on the construction of a Data Centre on the West Cambridge site (Reporter, 6304, 2012–13, p. 471).

Mr J. P. King (University Computing Service):

Deputy Vice-Chancellor, I am Julian King, an elected member of the Information System and Services Syndicate, employee of the University Computing Service (UCS), and member of the Regent House. However today I am here again as the Data Centre manager for the University Computing Service.

It seems there is an on-going failure to communicate. This Report is almost certainly positive spin on information which already had positive spin applied, resulting in a Report full of misleading statements and half-truths.

In July 2012, the First-stage Report on the Data Centre was published. It is probable that many of the details were correct when the Report was proposed, but they were untrue by the time the Discussion was held. This seemed reasonable; projects change and, especially in the early stages, tend to be fluid. However I had assumed that someone would give a speech introducing that Report and clarifying aspects. Instead there was silence.

This time I will make no assumptions. There is virtually no element in this short Report that isn’t misleading to some extent or other. I would like to highlight some of them.

Whilst it is true that the current Data Centre for the UCS and High Performance Computing Service (HPCS) can no longer provide the infrastructure required, the statement is misleading because the reason is that both organizations are being ejected from the Arup Building1 and the current facilities will be stripped out. Indeed any mention of the Arup Building has been dropped from the Second-stage Report, presumably because this Data Centre arrives at least a year too late to be useful, a fact that only seems to have been communicated to Council after we were committed to the new Data Centre.

It is misleading to suggest that this will provide ‘a secure, robust, and energy-efficient location for University servers’, because there is only a fraction of the space required by the University and many of the elements which might have made this truly energy efficient have been stripped from the project to save up-front costs.

It is misleading to say that this will ‘provide sufficient space for medium-term expansion’. The UCS will have less physical space in the new Data Centre than it currently has in the Arup Building Data Centre. Along with Chemical Engineering and Biotechnology, the UCS hall is liable to be full from day one. If there is significant migration of other departments to the Data Centre, then the ‘fallow hall’ will need to be fitted out almost immediately.

The 40% reduction in carbon emission is, as far as I can tell, grossly misleading. To achieve that would require a 40% reduction in total power usage (or to make better use of the waste heat). This is not going to happen without making changes beyond anything implicit in building a new Data Centre. I am sure the new space will be more efficient, but accurate and credible information would be useful.

This is just the Report. There are additional failings outside of the Report. Despite my role, I have heard nothing formally about this Data Centre since before the First-stage Report. I recently found out that the Representative User has changed, but this new User hasn’t contacted me to understand the needs of one of the three constituencies.

The design and layout of the UCS space hasn’t been communicated to me. This isn’t a matter of idle curiosity; it would affect the equipment purchases we are making right now.

The UCS is in the process of an emergency move out of the Arup Building and into the Roger Needham Building. Getting information out of, and come to that into, this process has been hard work. The same is true for this Data Centre. The small Data Centre provided is only just large enough for the UCS to keep running its current services. We need more space to provide any of the major new services we are being asked to deliver.

The HPCS are in temporary accommodation and will have to move out, at the very least to satisfy the planning permission approval they have been given.

There are significant failings with the new proposed Data Centre, some of which I have highlighted above. There may be many more that I don’t know about. The University has put itself into the position where we have to build it or else close down the HPCS and cripple the UCS.

It would be helpful if we could have some foresight in our management. There is no cure-all, but communicating plans down the chain earlier, and accepting feedback up the chain would seem to be a step in the right direction.

It would be nice to stop sailing between Scylla and Charybdis all the time.

Report of the General Board, dated 29 April 2013, on the establishment of two Readerships in the School of Clinical Medicine (Reporter, 6305, 2012–13, p. 477).

No remarks were made on this Report.

Report of the General Board, dated 29 April 2013, on the establishment of a Readership in Neuroradiology in the Department of Radiology (Reporter, 6305, 2012–13, p. 478).

No remarks were made on this Report.