Of Hair Loss—and VME

25 January 2017
richard_kirk_blog.jpg

When a new year dawns, if you’re like me, you probably find your thoughts are often drawn to the past as well as the future—reminiscing about how things used to be, and what might be in store for us in the coming years. 

I recently found myself looking back over the last 20 years and marveling at how quickly everything had changed—from mobile communications, to social media, to the pace of business, to how my kids had grown up, to how I’d lost a great deal of hair… 

I decided to stop at that point. However, I did find myself thinking about a pretty constant presence over the last 20 years: VME.

When I first became involved with embedded computing back in the 90s, VME was the dominant architecture in defense and aerospace applications—and in many ways, it still is. Today, there are many more alternatives, with lower power, smaller modules such as COM Express or high performance VPX cards—but VME is still widely used by many of our customers.

Every few years, a new architecture arrives which threatens to be VME’s replacement—Compact PCI, ATCA and so on—but VME just keeps steaming along, and is still getting valuable design-ins right now. In 2017! In my mind, this means that VME has a long future ahead of it, as many of these design-ins will require 10, 15, 20—perhaps more—years of support as they are integral to long term programs for defense organisations around the world.

Ageless?

So: is VME ageless? Well, not exactly. The VME bus itself has increased in throughput over the years with the introduction of various initiatives such as VME64 and 2eSST, but it is definitely on the slow side compared to fabric-connected schemes such as VPX. 

But bus speed isn’t everything, and the processing capability within a typical SBC has increased by a staggering amount in 20 years. In 1997, we were working with a single core PowerPC chip running at 300MHz—but just recently, Abaco released a single board computer with an 8-core CPU running at six times that speed. That’s a factor of 50x improvement—before taking into account built-in accelerators like AltiVec.

Here in the UK, you’d liken VME to an aging soccer player, competing against a new up-and-coming starlet. The legs may have gone—but the brain is a lot sharper, and he brings a ton of experience. The canny coach looking to win the championship often chooses the older guy he knows he can rely on, rather than the unproven teenager. 

And so it goes in the defense market: VME is reliable and proven and holding its own, but VPX offers a great alternative for those with a need for speed and the will to take on a new infrastructure.

Over the next few years, I guess I’ll continue to lose my hair. VME, on the other hand, will continue to thrive. Long live VME.

Richard Kirk

Richard graduated from the University of Manchester in 1984 with a BSc degree in Physics, and followed that in 1998 with an MBA from the Open Business School. In the interim, he’d joined Plessey Optoelectronics, part of one of the UK’s most venerable technology companies. He joined Radstone, located in Towcester, UK—subsequently acquired by GE—in 1999, and now has worldwide responsibility within Abaco's business as Director, Core Computing.