flee@shire.cs.psu.edu (Felix Lee) (10/09/89)
Getting rid of this. Didn't spend as much time as I would have liked doing the survey. Someone else's turn to survey now. Responses are lightly edited. -- Felix Lee flee@shire.cs.psu.edu *!psuvax1!flee <werner@nihefk.nikhef.nl> Werner H.P. Vogels We at the Haagse Hogeschool, Intersector Informatica have 16 6150's all running IBM/4.3 with X11R3. We are experimenting with AFS but we do use ATK. We do not use the IBM displays but a third party display with the apa16 controller. <karish@forel.stanford.edu> Chuck Karish The Stanford School of Earth Sciences has six 6150s running IBM/4.3. All of them have 6153s as their consoles, so they don't run X. <mbrookov@nike.cair.du.edu> Matthew B. Brookover We use IBM 4.3! We have 3 RTs, 2 6150's, and a 6151. RTTYPE OS disk space 6150 AOS 930MB (3 310H disk drives) 6150 AIX 2.21(yuck!) 620MB (2 310H disk drives) 6151 AOS 310MB (1 310H disk drive) We are using them to replace 3 VAX 11/170s that are running BSD 4.3. We also have 15 or so 6152's, but the powers that be wanted to run MS-DOS (yuck) on them. Just as soon as I get an ethernet connection in my office it will be running AOS and X11. AOS seems to be pretty good except for the C compilers, what I need is the GNU C compiler. Another annoying problem is the three RTs have IBM 6154 color monitors wich are not supported by X11. If you know of a driver for the 6154 please let me know. <wittmann@engr.wisc.edu> Art Wittman Well, here at the University of Wisconsin Computer Aided Engineering Cntr, we've got 3 RTs running IBM/4.3. They don't use any of the Andrew extensions either. Two of them have a lot of disk (6X310 Meg) and the other acts as a mail/news router for our network. The two systems with a lot of disk are used as file servers for our network (not the best choice, but a politically prudent one). Not too many applications are run on these things, although we have been able to port spice and few other berkeley-type public domain programs. We were letting users remotely log into the systems to use compilers but that was too much of a load along with the file serving duties. <henryc@mcs213k.cs.umr.edu> Here at University of MO - Rolla, we are stuck with 15 6152's and a 6150/135 in our department (Comp Sci). We run AOS/4.3, Dec. 88 Release, X11, but *NOT* Andrew or AFS (although I believe there's some people in IA who actually got AFS to work after *MUCH* grief, as I recall). Another dept. on campus got 25 6152's and a 6150 server and *NEVER* got AFS to work and so now have 25 DOS machines and an RT (they got sick of dealing with IBM, as we are also). 10 of our 6152's are on Token-Ring for student use - the others are in faculty offices. Overall, we've been real *UNHAPPY* with IBM. They just *DON'T* have their act together in the small system business - this is also true of the PC side of their operations. What more do I need to say about a system than the c compiler doesn't work, the pascal compiler doesn't work, dbx doesn't work, yp doesn't work, name service (DNS) doesn't work, some of the networking hardware doesn't work, dos <-> unix transitions are flaky, os/2 <-> unix interface is flaky, the thing changes permissions on our filesystem when it restores the filesystem on a fsck, X doesn't work correctly, and so forth, and so forth, and so forth.... One other story...I mentioned above that another dept. tried to run AFS... I'll flame for a bit. IBM promised the world to this non-computer oriented department so they got 25 6152's, a RT, and a 3812 printer. Well, we both got our machines at the same time and while we were beating on our machines, they had IBM *reps* trying to get their's running (this is at the first of the year). At the end of Feb., IBM sent a guy from CA to fix some problems we were having and to try and get their network up and running. The guy *TRASHED* our system (by the time he left, nothing on our system worked - no NFS, no yp, no nothing). At any rate, he couldn't get the other dept's machines to work either, so they converted them to DOS machines and let them run the rest of the semester. During the summer and continuing presently, they're trying very quietly to *DUMP* the things on other people and they're going to buy some other computers. I believe I can say with some authority that when we go out poking around for more workstations, we'll look at the other three letter workstation manufacturer (the one that starts with "S"). <tengi@princeton.edu> Christopher Tengi Here at Princeton we have a bunch (7 at the moment) of 6150s in the computer center running AOS and X11R3. There are also some RTs of various flavors out on campus running both AOS and AIX. We chose AOS for a few reasons, I guess. For one thing, you can hack the sources when you need to. For another, there is more familiarity with BSD than SysV among those of us who use UNIX around here. There may be other reasons that I am not aware of as the decision was made before I came on board. <ROB@UNB.bitnet> Robert Robson We have one 6150 running AOS4.3 8meg/apa16 X11R1. We use it for research in the CS department. The OS release is one less than the latest since we installed the latest and found it could not handle our async card. I would very much like to be running X11R3 but it does not seem to be coming from IBM. It's not a bad port although there are some problems. Still, it is usable and beats AIX all to pieces (at least the PS/2 version). I suspect they will be dropping support for this product in favour of AIX (we have it on order to get the latest X). <edler@cmcl2.nyu.edu> Jan Edler Here at the NYU Ultracomputer Research Laboratory, we use 10 RTs. All the machines have apa16 displays, and a mixture of IBM and Micropolis disk drives. We use the December 1988 aos release, but we still use X10. <orachat@edison.seas.ucla.edu> Orachat Choedamphai Here at University of California at Los Angeles, department of SEASnet (School of Engineering and Applied Science), we use IBM/4.3. We have 6150s and 6151s. All running IBM/4.3 with X11R2, NFS, but no Andrew whatsoever. I have one 6151 running X11R3. <jnford@jay.weeg.uiowa.edu> Jay Ford We have one 6150, two 6151s, and four 6152s. They are all running the December release of IBM/4.3, with X11R3, SNMP, and gated. We do a little NFS, but no AFS or Andrew. The 6150 is a network server (dns, news, mail, white pages, etc) and network operations monitor (via SNMP); it will soon also be a BITNET/Internet mail gateway. Two of the 6152s are IP routers (and potential name servers), and others are pretty much X terminals plus local mail & news (via nntp). One of the 6151s will be a router & name server; the other will be a personal workstation for a CS faculty member. <lichter@cs.ucla.edu> Here at UCLA Computer Science we have about 35 6150s and 4 6151s, all except one running AOS, and most of these running either AOS 11.0 or 12.0. We have X11R3 and X11R2 running on them, but both show standard RT bugginess. We run Magic on our RTs with 5081s (Megapels), of which we have about eight. They are not very heavily used because their software tends to be unreliable, and because our main systems are Suns and therefore not binary compatable. We use NFS and not any of the other remote filesystem packages. <moore@utkcs2.cs.utk.edu> Keith Moore We have 18 model 115's with a 70 Mb disk each. We use 4.3 with NFS, and using a couple of old VAXen and some fast Suns as file servers. We even install 4.3RT on the RTs from the VAXen, with a hacked-up installation floppy. <davidra@helios.tn.cornell.edu> David Rabson I am a very part-time system manager of two PC/RT's running AOS. The campus supposedly has a person who is in charge of support; IBM refuses to talk to anyone but him, and no one stays in the job very long. Our support, then, is non-existent. For the first two years, the software was so broken that the machines were hardly used at all. Now a few people use them as X-terminals to real machines (Suns and mainframes). I also have a prototype preprint-title database up on one. This is a SUN shop, so there is little enthusiasm for IBM. <fetrow@bones.biostat.washington.edu> Dave Fetrow University of Washington has something like 40 AOS RTs. We're reasonably happy but then we had 2 (later 1) really GOOD systems programmers working on them at least half time for over a year (I think the IBM AOS NFS was based on the port done here). Note that we already had a pretty hefty investment in BSD Vaxen but only a few workstations (in math/stat/biostat anyway) so the RTs are very welcome. They've migrated mostly into the Statistics and Math departments over time. Having a pile of them makes life quite a bit easier plus the support of a VAX behind them. We run X10 rather than X11 but that's livable. All things considered it wasn't too bad a deal for us. We've had other smaller donations of equal (or even better) equipment that didn't work out becasue we lacked the "critical mass" that makes fixes and workarounds worth doing. Having that many machines meant we could spend a lot of time on them and still come out ahead. <jab@dukee.egr.duke.edu> John A. Board We run AOS on a couple of model 125's in the Dept. of Electrical engineering here at Duke University - it's absurd that it took AOS so long to get things to work correctly, but I would agree with the IBM Yorktown comments that it has improved greatly in 1989. With a megapel we find them to be very good X11 platforms for VLSI CAD work and other applications. <henry@garp.mit.edu> Henry Mensch we are project athena; we run it on approximately half of 1200 workstations deployed all over campus (the other 600 workstations are of the DEC variety). most of the workstations in the field are of a standard configuration (apa16-type display, 4 meg of memory, 'table-top' cabinet with the 'new' cpu board). we deliver software to them with NFS, RVD, and AFS; they all run X11R3 with local hacks. we have full source. we use these workstations for all kinds of tasks; the students use them as a general computing facility, the staff uses them to develop software, internet access, research, usenet, xtetris, etc., etc. in the early days, we weren't particularly pleased (to the point that software development was a problem because none of the developers wanted to use RTs). this is now no longer true; for the past year (since the APC upgrades happened) the RT became the hardware of choice for our developers (but not for long ... we just got a shipment of P?AXes waiting to be uncrated :>) one more addition to my comments: we aren't real pleased that there aren't real workable compilers for the RT. high c washes out when you need it most. the f77 isn't much better ... <tim@cs.wisc.edu> Tim Theisen We have 43 RT's in the Computer Sciences Department. 20 are used for instructional purposes (In a lab for student use.) The rest are in graduate students offices. The Faculty Support Center has 2 RT's running AOS. There are 3 in the College of Engineering, and 1 in the School of Pharmacy. <jparnas@ibm.com> Jacob Parnas At IBM Research in Yorktown Heights NY, the math department maintains about 45 RTs using 4.3 BSD UNIX (AOS). We have been very happy with this operating system. Some features that we really appreciate are full 4.3 BSD, source code available for practically all utilities and the kernel, good X11R3, the andrew toolkit, NFS, good asynch support, complete online manuals and documentation, and easily available fixes distributed on usenet. The 800 USC technical support number is useful for quick help with system problems. Most public domain software built for BSD UNIX systems builds and runs on AOS with little or no modification. What I like most about AOS is that it is 4.3 BSD to the core and not System V with some BSD goodies patched on as an afterthought. Charlie Slater, Eric Brunner, and Jeff Weinstein at IBM Palo Alto have been very helpful in helping to fix the few problems that we have had. We use these RTs for almost all of our computing including operating systems research, networking tool development, networking kernel research, SLIP, usenet, mail, gateways, printing, editing, text formatting, internet access, VM terminal emulation, graphing... I think that AOS has improved greatly in 1989. Most of the problems that I've had with it in the past have been fixed (high speed serial line support (thanks to Charlie Slater), X11R3 (not perfect, but overall quite good with the Megapel, and the C compiler among other parts of the OS have been greatly improved. I highly recommend trying 4.3 AOS, if it is available to you. <scott@longs.lance.colostate.edu> Scott Douglas We have one RT, and it runs AOS. I threatened to quit if AIX wasn't removed from our realm. The RT is part of a 190 node UNIX/TCP-IP/NFS/YP/eieio network. It seems to work just fine. No complaints. Not much usage. We plan to add a large ESDI disk (600MB) very soon. The 105MB disk plus an NFS partition is too restrictive. <end>
brunner@bullhead.uucp (01/17/90)
Hello everyone, I've reserved a time slot for an IBM/4.3 (aka "ACIS") BoF, Wednesday from 8pm until 10pm. This will follow the two-hour USENIX Association reception. The room location will be announced on the various notice boards and in the daily newsletter. Bill Webb, Larry Breed, Conrad Minshall, Charlie Slater, and several others from IBM Palo Alto will be available for taking questions and suggestions. I expect that we'll cover the existing support system, the status of the various compilers, 3rd party drives, NSFnet's use of the platform for the national net, and take questions. I selected this date to avoid schedule confilicts with Len Tower's GNU BoF, the Transarc's Andrew File System BoF, and Peter Salus' report on the OSF, all scheduled for Tuesday, and the UUNET and CSRG BoFs scheduled on Thursday. See you in WDC! Eric Brunner, Consultant, IBM AWD Palo Alto (415) 855-4486 inet: brunner@monet.berkeley.edu uucp: uunet!ibmsupt!brunner
steven@pacific.csl.uiuc.edu (Steven Parkes) (01/20/90)
> I've reserved a time slot for an IBM/4.3 (aka "ACIS") BoF, Wednesday from > 8pm until 10pm. Can somebody who goes to this (or to IBM in general ...) ask about the future of AOS/ACIS ... the X11R4 config/ibm stuff has stuff all over it that says "this is the last release that will support 4.3 ..."