VME obsolescence prompts thoughts of alternatives
A GE Intelligent Platforms perspective on embedded military electronics trends
A supplier of a small but important VME component recently announced that the part will go obsolete this year. In all probability, the manufacturer plans to move on from VME to something in greater demand from the commercial world. Although the move was hardly surprising in the larger scheme of things, it made news in the corner of the electronics market devoted to the embedded-computing systems used in military and aerospace platforms.
Yet even in that sector, the news of a key component’s imminent demise was far from a death knell for VME. Although the bus-based technology peaked at hundreds of megabits per second on the backplane, it remains an attractive alternative for systems like command and control, which stress board-level compute power more than board-to-board bandwidth. The latest VME designs, moreover, can compete successfully for slots in more demanding signal-processing applications such as radar and sonar. VME can claim all this, despite being directly descended from technology invented – and adopted by military programs – long before the official advent of the commercial off-the-shelf (COTS) era in 1994.
Military programs have used VME systems for decades; these programs have notoriously long gestation periods and even longer service lives. The investment has produced handsome dividends, though. As VME technology has evolved over the years, suppliers of embedded systems have developed products using the latest families of integrated circuits (ICs) while retaining the same card form factors and pin-outs and the same chassis sizes. When the time comes to upgrade a legacy VME-based system, there are always suitable VME candidates.
Reason to hedge
Over time, VME has eclipsed other technologies in the military and aerospace embedded-processing market. By the same token, however, most of the demand for VME today is from military users. This niche status is reason enough for customers to hedge their bets on new system designs. VPX, a point-to-point fabric-based architecture that offers a similar form factor (with a very different backplane), is well positioned for forward-looking programs with higher board-to-board bandwidth requirements.
VME currently has a far larger share of the board-level market than does VPX, with projected revenues of more than $200 million in 2017, according to IHS. Suppliers are still designing VME products and have extended the technology’s reach into more demanding applications. Board manufacturers fully expect to be shipping VME systems five years from now.
An example of the latest in VME is the GE Intelligent Platforms XVR16 Rugged 6U VME single board computer (SBC), a sixth-generation, Core i7 board that aims at signal processing applications such as intelligence, surveillance, and reconnaissance (ISR); sonar and radar; and command and control.
Life cycle management
Nevertheless, the end-of-life (EOL) announcement for the VME part should not and will not be ignored. Managers understand that obsolescence is a fact of life in the COTS world. Life cycle management contingency planning is mandatory in military procurements, given their dependence on the commercial market.
In this particular instance there are at least four options. Military customers, integrators, or board suppliers could make EOL buys to ensure the availability of the expiring part for the remaining life of their programs. These components could be stored indefinitely in specially controlled chambers until needed. Alternatively, other VME parts could be substituted, perhaps lowering performance but containing costs. Third, an equally performing part could be designed, using a technology such as a field-programmable gate array (FPGA).
A fourth alternative, more attractive to new programs than legacy efforts, would be to transition to a higher-performance architecture such as VPX. While the VPX route would be more expensive than the other alternatives, it might be an acceptable investment for an ISR application requiring significantly greater bandwidth than VME can provide.
To sum up, the news of the imminent demise of a key VME component – the bridge that translates chip language to the lingua franca of the bus – presents military customers with a host of choices and tradeoffs, not with a crisis. Through careful advance planning, a range of alternatives is available to meet industry need and maintain continuity at an acceptable price.