High speed data transfer between two Real-time target machines or third-party nodes
Shared memory technology, also called reflective or replicated memory, is the most efficient way to transfer small to large amounts of data in real-time and lowest latency across two or more target machines or third-party nodes. For this each target machine or node has to be equipped with a shared memory I/O module from the same family. They are interconnected via fiber-optical cables to form various types of network topologies including simple node-to-node setups up to redundant ring or star topologies connecting up to 256 nodes and more. All modules are available with 3.3V or 5V signaling and 128-256MB of RAM.
The supported Shared Memory I/O modules support data transfer rates from 2.1 to 2.5Gbit/s and data throughputs exceeding 170-210MB/s compared to 1/10MB/s using Ethernet and 100MB/s using Gigabit Ethernet connections.
Speedgoat supports and installs shared memory I/O modules from various suppliers given the varying requirements. All supported shared memory I/O modules include xPC Target driver support and comprehensive Simulink blocksets to define memory partitions, read and write data from/into these partitions, retrieve statistical information on the running I/O module, and certain locations in the partitions to issue interrupts on a local node.
IO901
|
High-speed, low latency reflecteive memory I/O module from Curtiss-Wright with single fiber port (SCRAMNet GT200)
|
IO902
|
High-speed, low latency reflecteive memory I/O module from Curtiss-Wright with dual fiber port (SCRAMNet GT200)
|
5565PIORC
|
High-speed, low latency reflective memory I/O module from GE Intelligent Platforms (5565PIORC product family)
|