British Petroleum International Ltd has been throwing the spotlight on its IBM disaster recovery facilities, based in Peterborough and known as Fastbak. BP acquired Fastbak nine months ago from the Midland Bank’s Thomas Cook travel organisation, and integrated it into its Information Systems Services division. The ISS division was retained by BP after it sold Scicon to Systems Designers, and provides IBM, DEC, and Unisys recovery facilities both within BP and to third parties (CI No 1,223). ISS has recovery facilities in Antwerp, Harlow, London, Peterborough, and Glasgow, and the last was established as a result of BP’s controversial acquisition of the Scottish company Britoil Plc. The division as a whole employs 750 staff, achieved a turnover of UKP50m last year, and in addition to disaster recovery, provides processing and consultancy services. Most of Fastbak’s business, approximately 88%, is written outside BP, and it forecasts that the marketplace is set to grow at at rate of 25% per annum over the next five years. The dedicated centre at Peterborough offers a warm recovery service, covers 10,200 square feet, and has two IBM computer systems, a 4381 and 3081K. Plans are afoot to introduce a 3090-200E in the first quarter 1990, which will add 36 MIPS and ESA capability to the existing 18 MIPS processing power. Current disk capacity, around 55Gb, will also be upped by the introduction of a 15Gb 3380K at the end of 1989, increasing to 30Gb in the first half of next year. By 1992, Fastbak will have introduced 3990 controllers and disk storage according to customer demand. Communications capabilities include 3274 modems, X25, V24, V35, and X21 lines. During 1990, the 3705 will be replaced by a 3725, and 1991 will see the introduction of a 3745. Unlike some companies, Fastbak has not experienced the installation bottleneck associated with British Telecom’s Kilostream, and says that in the Peterborough area, it takes takes no more than one month to get one in. Like many recovery companies, Fastbak would face a dilemma if more than one of its 27 clients needed to use the 18 MIPS facilities. However, over the past five years the full range of services has only been needed once, and that was when Thomas Cook’s own data centre caught fire. If such a situation did arise, it’s a case of first come first served for the first 12 weeks, with BP’s other sites coming in as back-up.

IBM’s Entry

On IBM’s entry into disaster recovery, Fastbak professes little unease. As always with IBM, there is a suspicion that it may well step in and undermine third parties, but it also gives credibility to the view that recovery should be an important part of a company’s operations – something DEC did little to further when it withdrew from the UK market, although maintaining its continental facilities. Mike Bullions, sales and marketing manager for the DEC/Unisys activities of ISS, believes there is an increasing likelihood of legislation compelling companies to have recovery facilities, and he doesn’t rule out legislation that requires companies to report disasters. Both are standard practice in the US, and it’s one way of forcing company directors to regard disaster recovery as a financial issue, rather than a peripheral concern for computer room boffins. – Janice McGinn