Cost reduction and agile enterprises to force change on IT infrastructures

Key drivers for manufacturing businesses in 2004 will continue to centre on reducing costs, becoming more responsive and agile and ensuring business continuity. So says Simon Gay, consultancy practice manager with European integrated infrastructure services provider Computacenter. And he adds that those issues will translate into demands on companies’ network and IT foundations that facilitate moves towards the goals of the ‘real time enterprise’ – with greater application availability, managed services, consolidation at every level, and re-use of assets moving up the business agenda. Manufacturers, he says, will need their entire IT and network infrastructures to become more joined-up and flexible so that, for example, sales, pricing and production can be more dynamically driven against demand, capacity and stockholding. Gay: “Supporting systems have to become more dynamic and automated – better at empowering companies and their users for good, fast decision-making. So the emphasis in 2004 will need to be, for example, on business intelligence, workflow and business process automation, but also on integration and consolidation.” He sees that in turn driving more demand for managed services – already up around 10% year on year according to Computacenter – and with that consolidation. Gay believes there’s a growing understanding of the need to take advantage of best practice and the economies of scale that come from outsourcing aspects of infrastructure provision and associated services. “Service level agreements can be set up for implementing whole infrastructures, as well as managing and remote hosting – particularly of newer initiatives like e-commerce and supply chain management. We call it ‘selective outsourcing’ and the point is it’s safe.” As for consolidation, he says it needs to happen at several levels – data centres, people, processes, the entire network and infrastructure stack, voice over IP (VOIP) ... the whole bit. “Mergers and acquisitions are one driver. But the fact is there are efficiencies and economies of scale to be had from today’s better management tools, single user interfaces and network, server, application and desktop management systems.” And again, centralised managed services are looking more attractive – all the way from break/fix contracts upwards. Looking specifically at VOIP, he says: “There certainly can be merit in reconsidering the sense, cost and complexity of running two entirely separate networks to the same desk points. Voice over IP in this context is looking increasingly attractive as prices continue to fall and the technology and methodologies mature.” However, he believes there are more pressing issues, and he cites ‘re-use of IT assets’. “Most companies have got disparate technologies – different servers, applications, storage, the network layer and so on. It all needs consolidation and integrating together to cut costs and get to the real time enterprise.” It’s the argument for harnessing our existing legacy network and IT infrastructure and adding overarching ‘best of breed’ systems, instead of ripping and replacing with pre-integrated enterprise packages. However, he makes the point that making that work requires vendor-independent integration services, and makes the point that that’s precisely Computacenter’s role – also underwriting and potentially managing resulting systems. Moving on to business continuity, he says that post 9/11, industry hasn’t mobilised to the extent that it should to improve its disaster recovery policies and systems – but adds: “Organisations are investing seriously now.” His concern, however, is that most manufacturers’ disaster recovery plans are a “triumph of hope over experience”. He points out that it’s all very well conducting resilience and stress testing under controlled conditions, but: “when real life disaster strikes, it isn’t like that. “Many companies wildly underestimate the scale of problems and the time it takes to recover data. There are plenty of examples where it takes many days, not hours. And for some that will put them out of business. Whatever else they do, they should have proper data replication procedures and systems in place. We can help them with everything else.” His key message: change on the big picture scale is inevitably disruptive, so manufacturing businesses have to attend seriously to proof of concept and real and intensive stress testing. “There’s a business continuity issue here: how do you know your systems are going to continue working when you make these business-justified changes?” He urges companies to employ risk mitigation services like those available from Computacenter by running full infrastructures at high loading in its data centres and laboratories so that implementations and operations go smoothly and safely. A final thought: who’s concerned about open source. “There’s activity but a lot of talk too,” says Gay. “Companies are interested in the potential savings, but it’s not likely to be high on the agenda for 2004.” Although, on the face of it, companies could reduce costs on licences and maintenance contracts by using more open source at least at the server level – and there are good examples in the government sector – manufacturers need to look at the total cost of ownership and resilience of alternative solutions. Messing with mission critical systems away from the tried and tested platforms is a risky business. For those that do want to investigate open source, Computacenter is offering a vendor-independent business case service to establish holistic cost/benefits and risks, and make justification more watertight.