Rich Kenny is from Techbuyer.
Q. Let’s look at the ICT landscape first. What developments and trends should universities be aware of over the coming few years, and how should they make sure they are responding to them?
Moore’s Law [the principle that the speed and capability of computers can be expected to double every two years] is now slowing down. This has an implication on buying new, because the efficiency gains could only be incremental, rather than exponential, going forward.
The other major issue is the edge and inherent security risks involved. The market is becoming increasingly aware of the need to release patch tests but these may not be available on all equipment, particularly from other less security-conscious markets. With the number of students potentially bringing smart electronics such as electronic assistants into their halls of residence, IT managers need to keep abreast of the best security solutions, education and segmentation to protect all parts of the system.
Q. How important – and how feasible – is it for universities to have a fully scalable ICT framework, capable of growing in capacity to suit demand either over time and/or for specific projects?
I’d say flexibility is just as important as scalability, if not more so. For networking, the best plan is to design an architecture that can be modified as often and as cost-effectively as possible to cope with fluctuating demand. It’s an approach we increasingly see at Techbuyer – companies and organisations save money on component parts like storage and switches, and save the big spend for the stuff that matters.
We see this approach a lot when it comes to our configure-to-order server service. Some organisations choose all new parts, some refurbished, some a combination of both. This helps them save money for high-tech projects, such as crypto mining research.
Q. Are there ways in which ICT provision can be monetised (eg offering students different internet capacity/bandwidth at different tariffs)?
It’s possible, but likely to make the institution unpopular with its students. For online learners, who are already sacrificing face-to-face teaching while still paying sizeable fees, adding more to their bill in the form of higher network charges would not be a good business decision.
A smarter approach would be to save money overall on the system design, using third-party or refurbished components where possible. A 2017 report called How to Avoid The Biggest Ripoff in Networking pointed out that transceivers account for 10–15% of network capital spending, and that significant savings can be made using non-branded versions of these: essentially the same, but with out-the-door prices typically 80–90% lower than list prices from major manufacturers.
There are also monetisation opportunities at the decommissioning stage. The top-end IT asset disposal companies will offer cash returns on IT equipment that they are able to refurbish and redeliver to the secondary market. Security is an issue, so you need to look for companies that have relevant ISO qualifications 9001, 14001 and 27001 and are also able to demonstrate permanent data erasure.
Q. How should universities balance budgets while also future-proofing their ICT? Does more durable always mean more expensive?
I don’t think that durability and cost-effectiveness are necessarily incompatible. A big reason why larger data centres buy new servers is around the customer support, rather than the inherent reliability of the equipment. New equipment comes with standard three-year warranty and service contracts that guarantee delivery and installation of spare parts within a very limited timeframe.
However, you tend to pay a premium for this, and it’s a cost that many experienced IT managers avoid wherever they can. Buying refurbished parts, or spares from a secondary supplier willing to provide a three-year warranty, fast delivery on spares and sensible return policies essentially gives customers the same peace of mind at a vastly reduced price.
There is a limited lifespan on mechanical parts – however, a lot of components coming onto the second-hand market have never been used. An example might be an SSD that has been used as part of a failover system that never got switched on. Choosing this option over factory-sealed means you can buy effectively new equipment from your preferred brand such as HP or Dell at a much lower price.
Q. What is the minimum that a university should be offering its students in terms of its ICT provision?
A service that is secure and fit for purpose. It should be designed so that students have the capability they need to carry out their studies and the peace of mind of knowing the university systems are secure.
Q. Should ICT systems be updated on a piecemeal basis, or via more radical overhauls every few years?
Universities are under pressure to meet multiple IT demands with a restricted IT budget. One way around this is to carry out a rolling upgrade process over three or four years. It spreads the cost out in a more predictable way and is less hassle for the IT department. It also means that there are options for carrying out an in-house ‘refurbishment’ process, stripping equipment down to component parts, reusing where possible, and only buying what the institution actually needs. The only slight bump in the road is data erasure, which many institutions are not set up to carry out safely.
Universities are also under pressure to have sustainability in mind with every decision they make, and IT is no exception. Only 15% of e-waste is recycled and so throwing away your server or component parts after a few years is not sustainable. We at Techbuyer are members of EAUC, the sustainability champion for UK universities and colleges, and are working closely with them to drive IT sustainability to the heart of higher education.