BROWSE BY TECHNOLOGY



RTC SUPPLEMENTS


SOLUTIONS ENGINEERING

VME 25th Anniversary

VME: Moving from Evolution to Revolution

VME has undergone a profound and compatible evolution over its 25-year life. Today that evolution is undergoing a deliberate and orderly development as new serial interconnects are adopted. The result is not a break but an extended ecosystem that promises to unfold for many years to come.

DAVE EVANS-HUGHES, CONCURRENT TECHNOLOGIES

  • Page 1 of 1
    Bookmark and Share

Since its adoption in the early 1980s, the VME architecture has evolved continuously, and remarkably successfully, to keep pace with the improvements in microprocessor and communications technology. In the early 1980s the world was learning about PCs that used 5 MHz CPUs, 640 Kbyte RAM and the ISA bus. Now we are growing used to dual core 2 GHz CPUs with 4 Gbyte RAM and PCI Express. As the CPU capabilities have grown, so must the communications between them, particularly in backplane systems utilizing several intelligent boards. Parallel bus technologies have been stretched so far, but now the future is plain—serial interconnects with ever-increasing bandwidth are the choice for the future. VME has willingly embraced this change with the introduction of three major standards: VITA 31, VITA 41 and now VITA 46.

VME has moved from the 20 Mbytes/s backplane speed where it began, through 40 and 80 to 320 Mbytes/s offered by 2eSST technology. Each step has pushed the limits a little more, and stretched the backplane signal timings so that skew or signal quality effects are increasingly causing concern for further enhancements. Serial interconnects reduce the scale of these problems significantly, and allow not just improved speeds now (in excess of 1 Gbyte/s), but the promise of significant increases to come. This is countered to some extent by increased latencies and overheads associated with the serial protocols (see sidebar “The Good and Bad Sides of Serial Interconnects,” p.36).

The boards in VME systems, as in other types of systems, have become more and more intelligent over time, and with this intelligence the level of inter-board communications has changed. As an example, consider an advanced radar system using several VME CPU boards to collect and process imaging satellite data via multiple DSP-based acquisition boards. Figure 1 shows the way these boards used to be connected for this real-world application. The DSP boards acquire data from the satellite receivers, then process the data to filter noise and to create digitized information that is passed via VME bus to one or more CPUs, normally located in the same VME chassis. The user interface is provided by an external PC connected via a LAN. In this situation, the VME bus provides both the control and data planes for the acquisition boards. As a result, in higher-end applications, the number of VME slots used by the DSP boards, together with the total VME bus bandwidth requirements, limits the total processing capacity of the system.

Now consider Figure 2, where the DSP boards have been replaced by FPGA PMC boards on VITA 41 baseboards, and the main CPU board has been replaced by a much more powerful unit using a very fast dual-core processor. The data processed by the FPGA boards is now passed via backplane Gigabit Ethernet connections to a CPU board in the same chassis. The VME bus remains in use as the control plane while the VITA 41 serial interconnect is used as the data plane. The system connects via the chassis switch boards to the LAN, which hosts the PCs providing the user interface. In this system, flexibility is greatly improved by utilizing the in-system LAN to provide both the internal and external connectivity, allowing CPUs or additional FPGA engines to be added either inside the chassis or even in an entirely separate chassis. Splitting the control and data planes for the FPGA boards also improves responsiveness to control functions by minimizing interference from data transfers.

In adopting a serial interconnect, the traditional backplane “bus” is no longer present. Several interconnection topologies are possible, but the most common are the star and dual star configurations shown in Figure 3. The obvious complication with these topologies is the need to include one or two additional switch boards, which adds to the costs and of course lowers the system MTBF. However, a dual star configuration also provides multiple interconnection paths between boards, potentially improving overall system reliability. An alternative topology, also shown in Figure 3, is the mesh, which allows boards to directly connect to each other without using switch boards. This improves the cost but makes the implementation of multiple interconnection paths much more complex.

VME – The Serial Interconnect Story

For VME, the modern serial fabric evolution started with VITA 31, which adopts for P0 the connector style and pinout used by the newer CompactPCI standard for its Packet Switched Backplane (PSB). By using Ethernet technology for the fabric, speeds up to 2 Gbits/s full-duplex are possible, with dual fabric support for redundancy or bandwidth improvements. One big virtue of this standard is the ease with which it can be adopted. Although new backplanes are needed, existing CompactPCI switch boards can be used and the remaining VME64x features can be retained, including the ability to deliver more power to each backplane slot. VITA 31 supports only star and dual star fabric topologies.

Surprisingly, the demand for VITA 31 systems does not seem to have been strong. Although several board and backplane vendors offer compatible products, the takeup has been relatively slow. Perhaps this is because VITA 31 is the first stage in this serial evolution, and has been the herald of change rather than being seen as the change itself.

VITA 41 (VXS) takes the serial fabric evolution further, supporting multiple alternative fabric technologies such as Gigabit Ethernet, RapidIO, InfiniBand or PCI Express. Like VITA 31, VXS retains the existing 160-way DIN connectors of VME64x, so standard VME bus connectivity and power delivery are unchanged. By using a newer style of P0 connector, more I/O pins are also available too. Even better, VXS boards fitted with P0 connectors will still fit in existing VME backplanes that do not have obstructions or connectors in the P0 region. Likewise, VXS boards without their P0 connectors and with no obstructions in that region will fit into full VME64x backplanes. Hybrid backplanes combining VXS and VME64x are available, and the fabrics can use different topologies, including a mesh as well as the star and dual star concepts offered by VITA 31.

The backplanes themselves allow for many different interconnect technologies, although only one can be used in any one backplane. Serial RapidIO, InfiniBand, Ethernet and PCI Express forms of the standard have either already been ratified or are on the point of ratification, so the options are increasing. Some vendors have developed boards that implement the interconnect protocols in an FPGA, allowing them (at least in theory) to be reconfigured for the fabric in use.

Although the use of VXS products is still in its infancy, there is a much stronger ground swell of interest than with VITA 31. The prospect of much greater interconnect bandwidth and the provision for many more I/O signals may be some of the reasons. Hot-swappable switch boards and the availability of system management options such as the Intelligent Platform Management Interface (IPMI) may also spark more interest from users concerned about 24/7 availability and lower maintenance costs. The Concurrent Technologies VX 405/04x is an example of a new high-performance board that has been designed to take advantage of the connector level compatibility between VXS backplanes and traditional VME64x backplanes, while also supporting the new features offered by VXS (Figure 4).

The next stage in this development is no longer an evolution. VITA 46 (VPX) and VITA 48 Ruggedized Enhanced Design Implementation (REDI) change the rules significantly, while offering both large improvements in outright speed and greatly improved options for power delivery, system management and cooling. Arguably, this is no longer VME—certainly it is not VME as we have known it. Although hybrid backplanes (VPX and VME64x) are an option, the VPX backplane looks and is completely different to what came before. Even though 6U and 3U form-factors are supported, there is no connector compatibility with VME64x, but the VME bus electrical and “software” interface is retained, though not in a simple way.

VPX is a new standard and is still being finalized, but several vendors, particularly those supplying to aerospace and defense customers, are extensively involved with the creation of the specifications, and have already announced products. Product costs are likely to be high, especially in the early stages, and of course there is the need to completely reconsider the backplane technology. Many VME customers have been using the traditional backplanes for some time, and are comfortable with them. To get them into the zone where they are equally comfortable with what inevitably must be seen as something other than VME may take some time.

System Management Improvements

Along with CPU speed, integration levels have improved, and the number of I/O signals available on a single VME board has also increased dramatically. Of course, as a result of all this improvement, the power demands have also increased, and despite the chip manufacturers efforts to reduce power dissipation while maintaining CPU speeds, the demands of the applications continue to push the envelope for backplanes and power supplies. Regardless, customers expect and demand higher levels of maintainability, and the concepts of hot-swapping and system management are moving into the limelight.

System Management features arrived on VME with the introduction of VITA 38, adding the option for an Intelligent Platform Management Interface (IPMI) on the 160-way P1 connector used for VME64 and VME64x, with dual Intelligent Platform Management Buses (IPMB) supported on VME64x. IPMI offers a way to connect with in-chassis sensors to detect and record significant operating events, as well as provide chassis inventory information. For example, power supply

voltages, temperature and system fans may be monitored, allowing improved system fault diagnosis or just enhancing routine maintenance. Some systems may support external connections to the IPMB interfaces, which allows for remote monitoring of an installed unit or subsystem via LAN, WAN or other connections. Figure 4 outlines some ways in which IPMI can be connected.

VME is once again changing. Despite its age, it continues to move forward to accommodate the changes in technology that are the key to its survival. Although the original designers could not have foreseen many of the developments to come in the 25 years since its birth, VME has proved remarkably durable. The introduction of switched fabric interconnects, the inclusion of new system management capabilities, and finally the jump to VPX and REDI, offer major improvements to the performance and flexibility for the VME systems of the future. VME will continue to be the bus of choice for many industrial, aerospace and defense applications for many years to come.

Concurrent Technologies
Ann Arbor, MI.
(734) 971-6309.
[www.gocct.com].