VXS Processor Mesh Backplane The VXS Processor Mesh is a powerful architecture with bandwidth that can deliver 112 Gbps of aggregate throughput within the processing mesh in a single chassis. This is an improvement of 6x over standard Star or Dual Star VXS topologies. Developed to enable a switch/processor mesh technology for applications which require multiple boards for application processing, this hybrid backplane implements two VME64x slots, three VME64x/VXS payload slots, one collected rear I/O slot, and six VXS switch slots. Each switch slot implements twenty x 4 links for a total of 25 Gbps per switch slot. The system architecture supports up to 7.5 Gbps of throughput between the I/O front end and the processing mesh. The addition of a fat-pipe processor fabric segment further enhances the other key features of the VITA 41 architecture such as I C management bus, backwards compatibility to VME64x, PMC/XMC mezzanine sockets, rear transition slots, and the rugged VME64x mechanical architecture. The backplane can still utilize standard VME64x and VXS Payload slots. Therefore, backwards-compatibility is still an important element of the solution. As Processor Mesh VXS is a point-to-point mesh with out the complexity of a switched fabric, it is natural to use SERDES-based point-to-point signaling protocols. Because bandwidth-intense processes can now be offloaded to the VXS Processor Mesh, the conventional two-channel central switch fabric defined by VITA 41.0 may be used primarily for a control and IO pathway. This means less expensive switch cards may be used to support the Star or Dual Star switched fabric and the more powerfully equipped processor-switch engines may be reserved for signal processing and computational tasks. Also, the number of these cards used can depend upon the application process and not the switch fabric topology. |
 |
