VMIVME-5565 VMIVME-2128 modular

Description

VMIVME-5565involvement of the processors on any node. Using this system,all nodes on the network have a local copy of shared dataavailable for immediate access.Product Overview: The Reflective Memory concept provides avery fast and efficient way of sharing data across distributedcomputer systems.GE Fanuc Embedded Systems’ VMIVME-5565 Reflective Memoryinterface allows data to be shared between up to 256independent systems (nodes) at rates up to 174 Mbyte/s. EachReflective Memory board can be configured with either64 Mbyte or 128 Mbyte of onboard SDRAM. The local SDRAMprovides fast Read access times to stored data. Writes arestored in local SDRAM and broadcast over a high speed fiber-optic data path to other Reflective Memory nodes. The transferof data between nodes is software transparent, so no I/Ooverhead is required. Transmit and Receive FIFOs buffer dataduring peak data rates to optimize processor and busperformance to maintain high data throughput.The Reflective Memory also allows interrupts to one or morenodes by writing to a byte register. These interrupt (four levels,each user definable) signals may be used to synchronize asystem process, or used to follow any data. The interruptalways follows the data to ensure the reception of the databefore the interrupt is acknowledged.Each node on the system has a unique identification numberbetween 0 and 255. The node number is established duringhardware system integration by a series of onboard switches.This node number can be read by software by accessing anonboard register. In some applications, this node number wouldbe useful in establishing the function of the node.Link Arbitration: The VMIVME-5565 system is a fiber-optic daisychain ring as shown in Figure 1. Each transfer is passed fromnode-to-node until it has gone all the way around the ring andreaches the originating node. Each node retransmits alltransfers that it receives except those that it originated. Nodesare allowed to insert transfers between transfers passingthrough.Interrupt Transfers: The VMIVME-5565 provides four networkinterrupts. Any processor can generate an interrupt on anyother node on the network. In addition, any processor cangenerate an interrupt on all nodes on the network with a singleregister write.In response to this interrupt register write, the sendingVMIVME-5565 issues a special packet over the network, whichcontains the command strobe, the sender node ID, thedestination node ID, and 32 bits of data. When a receiving nodedetects the proper combination of destination node ID andcommand strobe, it stores the sender note ID and the data inone of four 127 location-deep FIFOs. The four FIFOs correspondto the four interrupts. Upon storing this information in a FIFO,the receiving node issues an interrupt to the local processor if ithas been software-enabled. The 32 bits of data stored in theFIFO is user-definable and typically is treated as an interruptvector. As part of an interrupt service routine, the localprocessor reads this information out of the FIFO and actsaccordingly.

VMIVME-5565