|
|
| High-performance embedded systems are familiar to those who design within telecommunications, wireless, defense-related communications and signal processing fields. While each of these systems has its own unique requirements, they also share much commonality in their internal interconnect requirements.
The very high degree of connectivity required in the systems is an extremely important attribute of high-performance embedded systems. These systems are typically based around backplane architectures with resource boards providing specialized functionality. In the past system OEMs used proprietary in-house interconnects for these applications. Recently the lower-end segments of the market have made use of the PCI interconnect architecture as a solution for backplane and perpheral connectivity.
PCI was initially developed as a desktop computer technology. CompactPCI has become the popular industry-standard form-factor for the embedded use of PCI. Despite this popularity, CompactPCI suffers from limits on backplane performance, scalability and reliability. As a result, CompactPCI’s PCI bus interconnect has been supplemented by a variety of switch-based interconnects such as Ethernet and StarFabric.
|
|
|
|
|
|
|
|
| High-performance embedded systems usually consist of a variety of processing elements operating in a distributed computing fashion. The processing elements move data between each other using direct peer-to-peer communications. This is different from the typical personal computer where there is a grouping of peripherals that communicate with a common host-processor through a shared memory.
As the industry moves away from shared broadcast buses to point-to-point links, processing elements should be able to communicate directly with each other without the need for intervening bridges or host controllers.
|
FPGA Processing Platform
|
| StarFabric, originally developed by StarGen, Inc. with the assistance of the StarFabric Working Group, is a powerful switched interconnect technology for backplane and chassis-to-chassis applications. It enables next generation Embedded Distributed Processing applications to meet market demands. The technology provides for unprecedented levels of scalability, performance and availability in a wide range of systems. It supports multiple classes of traffic adding further flexibility and enhancements to systems while leveraging existing standards-based software and hardware investments.
StarFabric enables you to expand your PCI topology. By deploying next-generation products such as Parsec’s PM410 StarFabric based CompactPCI PMC carrier board, you are provided with the additional expansion slots you need as well as the increased bandwidth, scalability and reliability required by your application.
PMC modules may be placed anywhere in the StarFabric network and systems may be spread across different enclosures. The PM410’s PCI to StarFabric bridges provide point-to-point, high-speed links (2.5 to 5.0 Gbits/sec) back and forth across the switched fabric between the local host and remote PMC slots, increasing transmission speeds and reducing latency. It is possible to obtain up to 420Mbytes/sec total sustained bandwidth on any point to point StarFabric interface.
The PM410’s StarFabric switches rapidly route traffic to different network segments. Failures are easily contained on local buses. This means reliability is improved and troubleshooting is made simpler. More PMC slots can easily join the topology by simply adding PM410 cards that are connected together by common CAT5 cables or over PICMG 2.17 backplanes.
|
Why use Parsec for your next embedded FPGA system?
|
| Parsec develops and produces customized electronic sub-systems for the defence, industrial and telecommunication sectors. Our CompactPCI boards and PMC modules have been incorporated into sub-systems for clients operating in the radar, electronic counter-measure and communication intelligence fields. These include:
|
|
|
|
|
|
| These boards and processing modules enable us to enhance the competitiveness of our clients by providing them with complementary capabilities and to ensure timeous, low risk, competitively priced solutions. The flexibility of the interconnect architecture and the high bandwidth enabled designs of our processing modules allow us to adapt, innovate and regenerate according to the needs of our clients.
|
FPGA Processing Platform Training
|
|
|
FPGA Processing Platform Training
As part of the delivery of the Complete FPGA Processing Platform, we will facilitate a five day training session on the demonstration platform. During the training, the customer will learn to use the platform and to access and modify the firmware reference design.
|
|
|
Outsourcing: Start-to-finish FPGA Solutions
|
|
|
Outsourcing: Start-to-finish Solutions
Using the most advanced Mentor™ tools, our technology partner Parsec designs FPGAs using VHDL. Their experience includes the design of large scale FPGAs in excess of 1 million gates. Parsec has completed FPGA designs using ALTERA, XILINX, and ACTEL devices. A well-proven FPGA development process focuses on the development of modular and maintainable designs. Extensive use of synchronous design techniques and simulation ensures that timing and routing requirements are met before hardware testing is attempted. Parsec's specialty includes the design of custom and standard bus interfaces, DSP and specialized processing functions into FPGAs.
|
|
|
  |
