Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uunet!aplcen!samsung!brutus.cs.uiuc.edu!psuvax1!psuvm!BRAZOS!SCHAFER From: schafer@BRAZOS.RICE.EDU (Richard A. Schafer) Newsgroups: bit.listserv.nodmgt-l Subject: Re: BITNET Restructuring Proposal Message-ID: <4865@brazos.Rice.edu> Date: 9 Feb 90 16:51:11 GMT References: Sender: Node Management Discussion Reply-To: "Richard A. Schafer" Organization: Rice University Lines: 27 Approved: NETNEWS@PSUVM Gateway Comments: Warning -- original Sender: tag was root@RICE.EDU Comments: To: bit-listserv-nodmgt-l@cs.psu.edu While VAXen running JNET 3.4 and JNET TCP NJE can indeed participate in a VMNET link, the VM VMNET sites with RSCS V2 have a particular major advantage: it is possible to support multiple data streams. As far as I know, and we have a connection to such a JNET site, the JNET support does not include multiple streams. I don't know enough about the Wollongong or TGV versions to say whether they do or not, but I believe that one of the very important pieces of the backbone arrangement will be the ability to support multiple data streams between backbone sites. While I know that the MVS sites can support multiple streams, none of them that I know of support anything like the VMNET code. Another point that needs to be made is the amount of load those backbone sites are going to be committing to carrying; I believe that's why the careful notation was made that each of the suggested sites was a fairly good size IBM machine, not a 4361 or something like that which I sincerely doubt would be able to support the traffic. I have heard doubts expressed in the past about VAXen, but am not qualified to say how valid those are, except to doubt that any serious person would believe that a MicroVax would be acceptable. :-) Richard