leonard@arizona.edu (08/31/90)
In article <63454@bu.edu.bu.edu>, kwe@buit13.bu.edu (Kent England) writes: > In article <1990Aug27.143144.62@arizona.edu> leonard@arizona.edu writes: >> >>The question is, what exactly does this Kalpana EtherSwitch thingy do >>when the segment it wants to transmit onto has a collision in effect? >> >>I think the answer is that it buffers. I got some literature from >>Kalpana and it says that there is a 256-packet buffer per Ethernet >>interface. > > I think you're right, but that sounds like a lot of buffer. > The Kalpana has to have very fast table lookup for forwarding. > Perhaps the 256 refers to the size of the MAC address table per > interface? At any rate, I would be interested to know how big the MAC > address table is, per interface, since that would affect where these > things could be used. I can't imagine that Kalpana could accept > arbitrary topologies with thousands of potential MAC addresses on some > interfaces and still maintain claimed forwarding rates. > OK, well let's back up a moment and refresh everybody on what this Kalpana EtherSwitch ("EPS-700") is supposed to be. It claims to be a switch that connects multiple Ethernets with an aggregate inter-Ethernet bandwidth of (n/2)*10Mb, where n is the number of Ethernets, up to a total bandwidth of 30Mb. (It can support up to 7 Ethernets.) Rather than storing and forwarding frames, it repeats them with minimal propagation delay onto the target segment, creating a virtual circuit that joins the communicating segments thru the "cross-point switch matrix". Here's a picture of what a 4-Ethernet EtherSwitch would look like. Such a box would supposedly have an aggregate thruput of 20Mb/sec. ---------------- AUI cable --- [epp] --| "cross-point |-- [epp] --- AUI cable | switch | AUI cable --- [epp] --| matrix" |-- [epp] --- AUI cable ---------------- | [ system module ] ("epp" stands for "Ethernet Packet Processor".) Here is what Kalpana claims from its literature: 100% filtering rate (15000 packets/sec.) 100% forwarding rate (15000 packets/sec.) Per-packet delay thru the switch: 40 microsec. Size of MAC address table per epp: 1750 addresses. Buffering per epp: 256 full-size packets. All for a list price of $9995 for a fully configured (7-ethernet) box. So the answer to Kent is: no, the Kalpana does not support "arbitrary topologies" the way a DEC LANbridge 200 does (which can handle 16,000 MAC addresses.) But it does claim full wire speed forwarding with 1750 addresses on a segment So the initial question stands: has anyone seen one of these? Does it live up to its billing? Aaron