An ancient Chinese symbol representing the rising sun, an electrical repair shop established in 1910, and 1989 revenues of $48,500m. It’s an increasingly familiar Japanese success story, and it’s the power that’s backing Hitachi Data Systems. National Advanced Systems put its European operations up for sale in 1988, and after Memorex failed to make the acquisition, it seemed a canny move for Hitachi to acquire one of its three European distributors. However, the perspicacity synonomous with Japanese business acumen made Hitachi stall until it found a partner in General Motor’s Electronic Data Systems Corp.
Paranoia
John Clark, Hitachi Data System’s director of European marketing, says Hitachi’s hardware skills combined with EDS’ expertise in consultancy and support, made the alliance attractive to both partners. Nonetheless, both Clark and Peter Waller, head of UK operations, acknowledge that the EDS’ 20% stake in Hitachi Data is a sound political move given US paranoia about Japan Inc. EDS doesn’t contribute to the day-to-day running of Hitachi Data, and according to Clark, the reality is that Hitachi Data is a subsidiary of Hitachi Ltd. Which must be cold comfort to Hitachi’s other European distributors, Ing C Olivetti, and Comparex Informationssysteme GmbH. The fact that Comparex failed to take over NAS when Hitachi has made it clear that rationalisation of distribution channels is on its list of priorities, has occasioned much speculation. The collapse of negotiations has never been clearly explained, and regardless of whether it was a issue of price or strategy, the prolonged uncertainty damaged Comparex’ business in the short term. In the long term, commentators are speculating that Hitachi Data and Comparex will engage in a price war and Hitachi Data, as a subsidiary of Hitachi, will inevitably win. The two companies insist that when they have met in the past, they have always competed aggressively. Hitachi Data says that will continue. However, Peter Waller categorically denies that Hitachi has approached or attempted to poach clients from Comparex. There is obviously a nasty after-taste from the failed takeover, and it seems to have become personalised. When the Comparex salesforce complains that Hitachi would rather lose an account to IBM or Amdahl, Hitachi employees echo the sentiment. When products are the same, it’s difficult to distinguish except on a personalised basis. Waller says he never believed that Comparex would acquire NAS, but given Hitachi’s desire to rationalise its distribution network, a merger of operations was more likely. The best way to do that would be for Hitachi Data to acquire Comparex. Yet, even if that were on the cards, Japanese caution and German nationalism make it unlikely that Hitachi, BASF, and Siemens could agree such a proposition. –
By Janice McGinn
Comparex says that it has seen little of Hitachi Data recently, and others are suggesting that the Hitachi salesforce is thin on the ground. Waller insists that perception is not reflected in sales revenue, which is 40% up on last year. He has increased both sales and system engineers over the past year, and says that Comparex and Hitachi Data don’t meet all that often since the split between systems and peripherals is different for each company. Half of Hitachi Data’s revenue comes from systems, but Comparex gets about 75% of revenue from peripherals. Waller would like to see Hitachi widen its base in the UK, targeting large corporations with increased levels of service and support, and a wide range of users that NAS didn’t deal with. Both Waller and Clark are adamant that Hitachi Data’s brief is to sell Hitachi equipment in Europe. The prime competitors are Amdahl and IBM. Like the other plug compatible suppliers, Hitachi Data has a peculiar relationship with IBM. A high percentage of the workforce are ex-IBMers, and Waller was with Big Blue for 18 years. The company wouldn’t exist without IBM, and says it intends to maintain total compatibility. Clark cites Amdahl’s approach to PR/SM, and claims that the plug-compatibles
are actually offering non-compatible routes. Well yes, Amdahl’s approach to partitioning is different from IBM’s, but only because Amdahl invented it, and IBM saw no need to pay Amdahl the compliment of following its approach exactly. As a rule, IBM is ungenerous towards the plug compatibles, and Hitachi is as just as vulnerable as the rest. It is forced to sell its disks at 60% of the IBM price, simply because the market is dominated by that one vendor. Clark believes the situation will change when clients are willing to pay for performance rather than name, and he’s proud that Hitachi’s new 7390 due in the fourth quarter of this year, will have an average seek time of 10.5mS as opposed to the 3390’s 12.5mS. He’s also happy that most of the worldwide supply of Hitachi’s 7900 semiconductor disks to Hitachi Data have gone to the European operation. France and Germany remain good performers for Hitachi, with the UK coming up a strong third, although Amdahl has maintained its head start. On new machines and the vexing question of Summit, or 3090 kickers, both Waller and Clark feel that there will be an IBM announcement this year.
Pre-empt
The longer the 3090 goes on, the better off is IBM, but assuming the new technology is ready, it’s simply a financial and marketing decision. Could Summit be a staging box? Waller feels that it has to be new if only because of the massive investment that goes into plants and manufacturing facilities. Will Hitachi pre-empt IBM’s announcement? Waller says he’d rather announce after IBM than reveal all six weeks before, only to be asked for the Hitachi response. The growth and power of Japanese companies has led to the suggestion that a Japanese operating system isn’t all that far away. A number of Japanese banks in the US, UK, and Australia already use Hitachi’s VOS rewrite of MVS, and some commentators feel that there will be a worldwide push within the next five years. Waller believes there is a trend, but he advocates caution and cites IMS, first announced in 1969, and the 10 year-old DB2, as examples of user reluctance to change. MVS, he believes, will still be dominant in 1995, and likely to remain so for a long time to come.
