By John Keller
SCOTTSDALE, Ariz. - Computer scientists are joining hands to push the speed of the venerable VME64 backplane data bus from its current maximum of 80 megabytes per second to 320 megabytes per second perhaps by the end of the year.
Designers of military systems - particularly who are involved in avionics designs - also are showing interest in VME320, yet are not yet willing to lend their wholehearted support. Military advocates need to see further demonstrations that this new technology not only will work as promised, but will also provide the reliability and long-term logistics support that military operators need.
The proposed VME320 standard from Arizona Digital of Scottsdale, Ariz., and Bustronic Corp. of Fremont, Calif., would be backward compatible with existing VME technology by retaining the VME64 96-pin DIN connector, and would accept older 6U VME printed circuit cards, says Ray Alderman, executive director of the VME Industry Trade Association (VITA) in Scottsdale, Ariz.
VME board suppliers such as DY 4 Systems Inc. in Nepean, Ontario, Themis Computer of Fremont, Calif., and CSPI in Billerica, Mass., are expressing guarded interest in VME320, but are waiting to see how the new technology progresses before committing to it.
Executives of VME interface specialist Tundra Semiconductor Corp. of Kanata, Ontario, are joining in the push to develop VME320, yet point out that more work needs to be done before anyone in the industry can declare it a success.
"There needs to be a new standard protocol that can take advantage of VME320," says Richard O`Connor, Tundra`s director of marketing and business development. "Tundra is eager to investigate and support new technologies such as VME320."
But the VME community "needs to standardize protocols so we can put them into our Universe 2.0 interface chip by the end of 1997," he cautions.
Avionics experts say VME320 would give designers a powerful new tool if the data bus is as good as its supporters say it is, particularly for computer-intensive tasks such as data fusion, situational awareness, and graphical mission planning.
"Just going faster doesn`t help," says a Washington-area avionics consultant who does not want to be named. "What would you bring forward from VME64 that would minimize risk and minimize disruption? If there is to be a radical departure from VME64 and you discard the past, you have really made things obsolete. But if they can make it backward compatible with VME64, that is a big plus. That is a big mitigating risk factor of going to VME320 if they have continuity between VME64 and the new things."
U.S. Navy Cmdr. Chris Evans, mission systems integrated product team leader for avionics and weapons integration in the Joint Strike Fighter (JSF) program office in Arlington, Va., calls VME320 "very intriguing," adding, however, that "I would not presuppose a technological solution at this point of the JSF program."
Although Evans and his deputies are looking at the Scaleable Coherent Interface 1-gigabit-per-second data bus as their leading candidate for the JSF, he says the future aircraft`s avionics architecture is far from being set, and that VME320 could offer much to JSF designers, particularly if it can be made to be backward compatible with existing VME technology.
"We are looking at other things than SCI, including VME, Fibrechannel, and about seven or eight other alternatives," Evans says. "The big JSF pillar is affordability, and VME320 has the potential of affordability. There is a lot of VME out there."
Although VME320 might offer an important solution to JSF avionics designers, Evans points out that "it offers a great deal more for legacy systems. It looks like it will work with old cards, and that offers benefits with legacy implementations."
