From hancock4@bbs.cpcn.com Mon Jul 21 10:14:45 1997 Path: stc06.ctd.ornl.gov!news.er.usgs.gov!mcmcnews.er.usgs.gov!news.indiana.edu!vixen.cso.uiuc.edu!howland.erols.net!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!news-xfer.netaxs.com!netaxs.com!bbs.cpcn.com!root From: hancock4@bbs.cpcn.com (Lisa or Jeff) Newsgroups: alt.folklore.computers Subject: Re: Why Mainframes? Date: 16 Jul 1997 01:17:10 GMT Organization: Net Access BBS Lines: 66 Message-ID: <5qh7em$hpv@netaxs.com> NNTP-Posting-Host: 205.164.134.210 Originator: root@bbs.cpcn.com Xref: stc06.ctd.ornl.gov alt.folklore.computers:163118 > Joe Morris wrote in article > > IM(NS)HO one source of many problems in today's computing environment > > is that too many people, having (correctly) found it easy to tune > > their personal computer to fit their needs, don't understand that an > > enterprise-level computer system isn't something that can be created > > by scaling up a dinky little desktop system. The above is an excellent point. I think it's obvious, but a lot of PC oriented gurus don't realize how hard it is to scale stuff up. One wrong mouse click and a PC is mis-configured and out of sync with the others, causing confusion with novice or non-technical users. Responding that "well, the users ought to be technically proficient" is not an answer. They didn't need to be with a dumb terminal. > I find it absolutely incredible that an organization will install all kinds > of PC's then do their absolutely best to turn them into dumb terminals. "Dumb terminals" appear to be very much out of fashion. I was expanding a mainframe application to new users and recommend simple cheap terminals for them (I think they're about $700 now.) That's all they needed to do their jobs. They rejected that in favor of full powered PCs, loaded with software--about $2,500 for the PC and software, plus $250 for the 3270 emulation card and software, plus installing the emulator cards and software on the PCs. 3270 type terminals plug and go. As far as I know, the users never took advantage of the PC or its software (Excell, Word, etc.) > What kinda thought processes are going on? My organization is actually > installing PC's where the A drives are locked shut!... Given all the ads in the trade press for this sort of stuff, or tools to lock out and monitor PCs tightly, I wonder too what's the deal. I think it's just because PCs are in fashion, as opposed to what they can do for a user. Most people I know who do word processing would do fine with an early simple W/P, like IBM's Writing Assistant. M/S Word is very nice and fun to play with, but I don't need to adjust font sizes and set up a table to tell my boss which week I'm taking for vacation and where I can be reached in an emergency. At home, I don't need to write to the phone company about an incorrect charge. > There are actual padlocks on the backs of the PC's so > that the end-users can't open the box and possibly by-pass the software > blocks. I'm glad I don't work for that company. > By the way, the total buy is 55,000 machines, not counting file > servers, routers, bridges, etc..... Simply incredible. Every day I hear Bill Gates accumulated another billion. Now I see why. The man's a genuis, I give him credit. He's convinced each and every one of us we MUST have a powerful PC with Windows and current Windows software. Most of us could get along with DOS 3.3, but that would be boring. Of course, we could all drive Chevy Novas or VW Beetles, but that would be boring too. Gates learned well from the auto industry. From stephen@stevef.demon.co.uk Mon Jul 21 10:16:55 1997 Path: stc06.ctd.ornl.gov!novia!news-feed.inet.tele.dk!europa.clark.net!dispatch.news.demon.net!demon!stevef.demon.co.uk!stephen From: Stephen John Farthing Newsgroups: alt.folklore.computers,alt.sys.pdp10 Subject: Re: Why Mainframes? Date: Tue, 15 Jul 1997 19:28:42 +0100 Organization: Ministry of Administrative Affairs Distribution: world Message-ID: References: <01bc8523$84da9b80$e12185c2@rashid1> <5pfvrg$goj$1@decius.ultra.net> <33bbcafd.3174005@news.vip.net> <5pimve$es$2@decius.ultra.net> <5pkj08$b2c@ss1.digex.net> <0q4ta6r06p.fsf@rorschach.hks.net> <5pu79l$vha$1@ausnews.austin.ibm.com> <5q2t1v$leo$1@flood.weeg.uiowa.edu> <33C5B51D.70E3@ibm.net> <868975272.5512.1.nnrp-3.c29fa409@news.demon.co.uk> NNTP-Posting-Host: stevef.demon.co.uk X-NNTP-Posting-Host: stevef.demon.co.uk [158.152.33.117] MIME-Version: 1.0 X-Newsreader: Turnpike Version 3.03a <0tbEg$MHZK+M9wdwakxIDRGahN> Lines: 80 Xref: stc06.ctd.ornl.gov alt.folklore.computers:163137 alt.sys.pdp10:3446 Having followed this thread for a while I thought I'd post a contribution :- Mainframe - The ICL 3980 in my computer room is an excellent example of a mainframe. It can support over 800 concurrent interactive users at one time each one getting sub second response times from the system (depending on the applications system software), it runs a significant batch workload at the same time. The hardware is ultra reliable. I know of systems which have run 24 hours a day for several years without any downtime. It is self monitoring - a system within it looks for problems and warns the engineer if preventative maintenance is needed. The hardware is mostly unique to ICL using custom chips and other components. It can support huge disc arrays. Each node (a sort of CPU) is connected to its peripherals by a high speed fibre optic token ring and nodes can be separated by some kilometers. Adding more nodes to a system increases its power and the OS can support either multi or single node processing. (My first VME mainframe was a 2957 which ran a 250 user system plus the Operating system and a development user in 8 MBytes of store) The Operating System is similarly robust and also unique to ICL. It is called VME and is written in a 3 GL rather than assembler. It is flexible and easy to use. A crash in one Virtual Machine will seldom crash the system. I have known it to run for months without a crash or a systems re-load. Later versions support Open Systems but alas we cannot run them on our 9 year old hardware. There is a fairly limited range of applications packages available so we write our own using COBOL and IDMS, C, FORTRAN and other languages are available as are some RDBMS. The software is leased rather than purchased. The comms software can connect, via an ethernet to dumb terminals, PCs and just about anything you may wish to connect to. There are sophisticated performance monitoring tools available. The manufacturers support service is very good - in the past I have spoken to the development staff to resolve issues - try doing that with Microsoft. Backup and recovery systems for discs and file systems are part of the OS and work well. Whilst the hardware and software is very costly the support costs are very low due to the reliability of the software and hardware - I guess that two people full time is all it takes. Minicomputers seem to be called servers now. I believe that any computer attached to a network which can deliver a service to another is a server. So my mainframe qualifies. However ignoring that. The servers we use are based on hardware available from a number of sources. The CPUs are either Intel or Sparc, they run UNIX or Windows NT, we buy them from a number of vendors depending mostly on price. They tend to run packages for Office Automation, for example MS Office and any applications run on RDBMS such as Oracle and are written in Visual Basic or the package equvalent. Whilst the cost of buying these beasts is low the ownership costs are high. I guess we have 30 or so and require 5 or 6 people to support them. Systems integration issues are always a problem. The systems crash fairly frequently although we do push the hardware quite hard. Lots of GPFs and ill behaved software. Diagnostic information, especially on the NT Systems is poor at the best, still a re-load cures most problems - shame about the other users of the system. The software is purchased rather than leased, support when available, is poor. Hardware problems are fairly frequent. Users tend to want the latest version of software which means that we are constantly involved in purchasing, testing, de-bugging , implementing and training. Performance can be dismal when more than a few people are using the system for complex tasks. On the positive side they allow our users access to software using a Windows GUI which they much prefer to the dumb terminal stuff they use to get. We still run some UNIPLEX for those who wish to use it. Also implementing intranets, E mail and anything involving the non-trivial use of graphics would be impossible using the mainframe and PCs alone. I guess that we are all familiar with single user PCs - the P200 on my desk has a far higher MIPS rating than the 3980 but I know which one I would use to deliver services to the 500 users we have. Slightly off topic is my preference for the MAC plus I used as a terminal/OA machine 12 years ago. The GUI and packages were simpler and more usable that Windows 95 and networking the beast was simplicity itself. -- Stephen John Farthing MBCS G0XAR Melksham, Wiltshire UK RSGB G-QRP 7766 From eapallat@orion-data.com Mon Jul 21 10:17:35 1997 Date: 14 Jul 1997 18:09:00 GMT From: "Eugene A. Pallat" Newsgroups: alt.folklore.computers, alt.sys.pdp10 Subject: Re: Why Mainframes? Henry W. Miller wrote in article <33C89C39.7F92@mp.usbr.gov>... > Mark H. Wood wrote: > > > > In article <5pkj08$b2c@ss1.digex.net>, doug@ss1.digex.net (Doug Humphrey) writes: > > [deletia] > > > I think that a lot of good knowledge has been lost in terms of > > > systems design. The 10 series left things on the floor when the > > > VAX people refused, and then only grudgingly, took some of the > > > righteous things from the 10, and in general I think that the > > > PC world would benefit form looking at some of the ways that it > > > was done. Faster CPUs are not everything... > > > > I feel the same way, but do you think they would listen? The PC world seems to > > go out of its way to avoid learning anything from the 30 years of ADP that > > preceded it. I've long felt that the whole market has a big, BIG case of NIH > > Syndrome, an attitude that anything and everything the mainframers did was > > wrong. > > OTOH, I adhere to the belief that if unix does it a certain way, > then it is almost certainly the wrong way to do it... > Every time I ask a PC hardware or software company rep why they don't have certain features, I invariably get an answer like "Give us a break - PCs haven't been around that long." My response? Competent people have. Why do you refuse to hire them? I usually get a blank stare in return. There are even times when the companys' reps haven't the slightest idea what the features are. Remove the '-' from orion-data for sending email to me. Gene eapallat@orion-data.com Orion Data Systems Solicitations to me must be pre-approved in writing by me after soliciitor pays $1,000 US per incident. Solicitations sent to me are proof you accept this notice and will send a certified check forthwith. From peter@taronga.com Mon Jul 21 10:18:54 1997 Path: stc06.ctd.ornl.gov!news.er.usgs.gov!mcmcnews.er.usgs.gov!news.indiana.edu!vixen.cso.uiuc.edu!news-peer.sprintlink.net!news.sprintlink.net!Sprint!cs.utexas.edu!news.uh.edu!bonkers!not-for-mail From: peter@taronga.com (Peter da Silva) Newsgroups: alt.folklore.computers,alt.sys.pdp10 Subject: Re: Why Mainframes? Date: 16 Jul 1997 08:07:26 -0500 Organization: none Lines: 50 Message-ID: <5qih2e$2fr@bonkers.taronga.com> References: <01bc8523$84da9b80$e12185c2@rashid1> <33C5B51D.70E3@ibm.net> <868975272.5512.1.nnrp-3.c29fa409@news.demon.co.uk> NNTP-Posting-Host: localhost.taronga.com Xref: stc06.ctd.ornl.gov alt.folklore.computers:163156 alt.sys.pdp10:3451 In article , Stephen John Farthing wrote: >Mainframe - The ICL 3980 in my computer room is an excellent example of >a mainframe. It can support over 800 concurrent interactive users at one >time each one getting sub second response times from the system >(depending on the applications system software), it runs a significant >batch workload at the same time. This is quite impressive, though we have a couple of minicomputers (or are they micros? They're Alpha-based) that support 100 X-based users with response times that are as good as a local PC, along with significant amounts of software development and database activity. The state of the art has advanced somewhat, and minicomputers are breathing hard on the low end of the mainframe's range. I wouldn't expect a mini to equal the workload of a big mainframe based installation, like an airline ticketing system, but I suspect that there's a lot of mainframes in use in smaller installations that aren't really justifiable any more. >The hardware is ultra reliable. I know of systems which have run 24 >hours a day for several years without any downtime. Again, the state of the art has improved. I have PCs with uptimes in the 2 year range. There are redundant minis you can get with hot-swappable CPUs and peripherals that shoudl have no trouble meeting this target at a significantly lower cost. Nine years ago you probably wouldn't have any alternative, but today you would probably want to look at minicomputers as viable alternatives if you were starting fresh. The servers you're describing sound more like micros than minis. While there are some very powerful Sparc-based systems they're on the low end of RISC hardware and the complexity of Solaris works against them. Sun's move from SunOS to Solaris might have been a good marketing move, but technically it's been a disaster. The less said about NT the better. Our NT boxes are mainly used as cheap X terminals that happen to run Microsoft Office and Netscape locally. The twenty or so NT boxes take up more of my time than all the X-terminals put together. The minis (older futurebus-based Alphas running Digital UNIX) are largely in a fire-and-forget mode. (If you don't have 64 bits you're not playing with a full DEC) -- Thought for the day: According to Eileen Harrington of the FTC, out of 283 pieces of spam received in FTC email in one week, only two were clearly not fraudulent. If you're a legitimate business, is this the company you want to keep? From stephen@stevef.demon.co.uk Mon Jul 21 10:25:36 1997 Date: Fri, 18 Jul 1997 07:58:31 +0100 From: Stephen John Farthing Newsgroups: alt.folklore.computers Subject: Re: Why Mainframes? In article <5qmgaj$eth@netaxs.com>, Lisa or Jeff writes >> >Mainframe - The ICL 3980 in my computer room is an excellent example of >> >a mainframe. It can support over 800 concurrent interactive users at one > >Isn't "ICL" the company that took over "Leo", the British baking company >that pioneered commercial computer development in England? > >When you use their computers, does a cupcake pop out at the end of a session? > I'm not sure if ICL ate Lyons (Lyons Electronic Office = LEO) but ICL, in turn, ate Ferranti, Intenational Computers and Tabulators before it did the current deal with Fujitsu. Many of us old timers - well 20 years in IT from programmer through to Assistant Director - consider ICL Mainframes to be much superior to other brands - notably IBM. Most of this is down to the VME Operating System, much easier to use and manage than OS/VS1 and MVS. Alas, to the best of my knowledge, a cupcake maker never did feature in their list of peripherals. However the previous model the 1900 Series did have a cullinary use :- Cast your minds back to the early 1970s, flared trousers were still fashionable, the Greatful Dead played the Bickershaw Festival in the UK, Punk was in its infancy. Computers were kept in huge, air conditioned rooms with at least one glass display wall so that visiting Important People could be shown the prestigous machine. The ICL 1900 series was in use throughout the UK as a General Purpose mainframe. In common with others of the time it used core store for memory which generated a fair amount of heat. An unnamed Government computer installation which had recently installed such a beast at huge expense together with a state of the art, bespoke, real time Transaction Processing System. A party of VIPs was being shown round by a serious suit who "knew a bit" about electronics. Having emphasised how much the project and Mainframe had cost he mentioned that the machine used core store. "Can we see the core store" said a visitor. "No problem" said the suit. They walked into the Computer Room. The suit took the back off of the cabinet which housed the store. Unbeknown to him the Operators had put their Pie and Chips Lunches on top of the store to keep warm. "Must be the most expensive Hostess Trolley in the world"* said a VIP. Visitors ROTFL, Operators flee, Suit with much egg on face. * A Hostess Trolley was very popular at the time for keeping food warm at dinner parties - like Flared trousers they seem to have dropped out of fashion. -- Stephen John Farthing MBCS G0XAR Melksham, Wiltshire UK RSGB G-QRP 7766 From bill_h@azstarnet.com Mon Jul 21 10:27:12 1997 Path: stc06.ctd.ornl.gov!utk.edu!newsfeed.usit.net!newsfeed.telalink.net!telalink!news-fw-6.sprintlink.net!news-fw-12.sprintlink.net!news-sea-19.sprintlink.net!news-in-west.sprintlink.net!news.sprintlink.net!Sprint!169.197.1.4!news.azstarnet.com!news From: bill_h Newsgroups: alt.folklore.computers Subject: Re: Why Mainframes? Date: Fri, 18 Jul 1997 07:10:09 -0700 Organization: Starnet Lines: 60 Message-ID: <33CF7941.BD4@azstarnet.com> References: <01bc8523$84da9b80$e12185c2@rashid1> <5pkj08$b2c@ss1.digex.net> <1997Jul10.122310.28825@indyvax.iupui.edu> <5qaed0$f0n$1@nntp2.ba.best.com> <5qn82p$h0k$1@nntp2.ba.best.com> NNTP-Posting-Host: usr17ip56.azstarnet.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 2.0 (Win16; I) Xref: stc06.ctd.ornl.gov alt.folklore.computers:163270 phil dawkins wrote: > ......PCs have different target audiences. For better or worse, > there is a new generation of people who look upon computers in > a quite different light. There are people who think computers are mainly tools to accomplish some task, something to do with manipulating and storing vast quantities of electronic representations of something in the REAL world. There are others who couldn't care less what's behind the data they're storing and manipulating.....or whether their computers even work reliably. For them, the line separating reality from fantasy doesn't even exist. They use a computer for entertainment, and to play games. Mainframes were/are used to play games sometimes, but it's probable that upwards of 90% of what they did was WORK..... and often for profit. 'personal computers' probably are used to play games, access 'adult' electronic sites, and feed their operators fantasies 90% of the time. > The situation is changing all the time, but I believe there will > remain a distinction between a personal PC and a "server", One would hope so. The coming revolution in personal computers will probably involve some sort of virtual reality......which simply means the difference between real and electronically created blurs further. > irrespective of whether the machine is at home or on the corporate > desktop. Whether a PC server will achieve mainframe performance and > accessibility is a contentious issue....... Since they are, in the main, put to different USES by their owners, the question is moot..... But it's apparent that PERSONAL USE is driving the evolution of the hardware.....selling millions of each new round of product. No reason non-entertainment users can't benefit from $200 multi-gigabyte hard drives, $50 for a hundred million bits of storage, $100 for 100,000 bits per second modems, etc...... it's driven the cost of computers capable of doing real work down by several orders of magnitude. But don't confuse what traditionally was meant by "mainframe computer" with what today is meant by "personal computer". The uses are different. What we need is word or words to differentiate computers based on the use to which they are put. "workstation" doesn't satisfy many. Maybe we need "playstations" in our vocabulary, too. For now, personal computers are a marketing failure, if counting homes equipped with one is the measure. While telephones, radios, vcr's, bathtubs, washing machines, toasters, automobiles all exceed 70%, some over 90%, of homes, personal computers are under 40%. More people have decided to do WITHOUT, than WITH. What it will take to 'succeed' isn't here yet, or people would be buying them. So a 'revolution' is inevitable. From larecom@fearthlink.net Mon Jul 21 10:27:55 1997 Path: stc06.ctd.ornl.gov!news.er.usgs.gov!news1.radix.net!newsgate.ptd.net!newsfeed00.btx.dtag.de!news-pen-4.sprintlink.net!news-east.sprintlink.net!news-dc-26.sprintlink.net!news-peer.sprintlink.net!news-pull.sprintlink.net!news-in-east.sprintlink.net!news.sprintlink.net!Sprint!206.250.118.17!nntp.earthlink.net!usenet From: Larry Allen Newsgroups: alt.folklore.computers Subject: Re: Why Mainframes? Date: Fri, 18 Jul 1997 22:10:33 -0400 Organization: EarthLink Network, Inc. Lines: 39 Message-ID: <33D02219.68F4@fearthlink.net> References: <5qbe4t$64r@netaxs.com> Reply-To: larecom@fearthlink.net NNTP-Posting-Host: 1cust122.max26.orlando.fl.ms.uu.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 2.01 (Macintosh; I; PPC) Xref: stc06.ctd.ornl.gov alt.folklore.computers:163289 Lisa or Jeff wrote: > > Per Carl's post.. > > > Really the only need for compute > > speed in a typical commercial application was, and, to large extent > > still is, to keep up with the records coming in (from cards, tape, > > disk, &c) and going out (to printers, cards, tape, disk, &c) as there > > typically aren't a lot of operations carried out on any given record. > > The above is quite true. It wasn't until 1962 that IBM earned more > money from electronic computers than tab machines. A look at computer > magazines of the early 1960s finds a great many ads for punch card > machines and support services. Electronic equipment was still very > expensive in those days. > > It was IBM's 1401 computer that really made a dent on things. > That is, the > 1401 had much faster I/O capability than computers had to date. > This post comes close to identifying the real issues surrounding the title of the thread: why mainframes? Few seem to be aware that the ratio of CPU speed to I/O speed was more than 1000 to 1. So the game was to use all those wasted CPU cycles while waiting for ANY peripheral to do its thing. Philco had the Transac 2000 that was able to do all I/O to and from mag tape, and have all other secondary transfers off-line. This meant tape transports received card data and output to the printer and punch. This was 1961. Burroughs and Honeywell had multi-processors and multi-I/Oprocessors, which allowed many I/O's to occur concurrently and then to set a flag to tell the control progam that the I/O was done. The 1401 did increase I/O speed, but without multiple processes to use the wasted CPU power there was still a huge amount of time that the CPU sat idle. So the real reason for mainframes was that you could have hundreds of peripheral devices on-line concurrently, and tasking for transaction oriented applications. It was the ability to create program modules that could handle a variable load of on-line tasks, with security, breakout/restart/recovery, audit/load-management, and a host of other functions all in a single environment. The need still exists. From doug@ss1.digex.net Mon Jul 21 10:28:51 1997 Path: stc06.ctd.ornl.gov!novia!news-feed.inet.tele.dk!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!howland.erols.net!news2.digex.net!digex.net!not-for-mail From: doug@ss1.digex.net (Doug Humphrey) Newsgroups: alt.folklore.computers,alt.sys.pdp10 Subject: Re: Why Mainframes? Date: 19 Jul 1997 00:22:29 -0400 Organization: Express Access Online Communications USA: 800-969-9090 Lines: 57 Message-ID: <5qpfe5$7kv@ss1.digex.net> References: <01bc8523$84da9b80$e12185c2@rashid1> <5qnvnd$ocj@polo.demon.co.uk> NNTP-Posting-Host: ss1.digex.net Xref: stc06.ctd.ornl.gov alt.folklore.computers:163292 alt.sys.pdp10:3476 >Wasn't the all-time record (for the era) some airline transaction >system IBM built with 15,000 simultaneous interactive sessions? I am not sure of the numbers, but that would be TPF (Transaction Processing Facility) on large IBM iron. TPF is the modern name for it - anyone who has been around for a while would call it ACP for Airline Control Program, because as you mention, what it did (and still does) is reservation systems. TPF/ACP is a real story - it is a complete throwback, people here on this list who are old timers would feel right at home in it (of course, after 5 minutes you should start cursing the old IBM 360/370 artifacts, but I digress ;-) In fact, anyone who grew up with unix, relational databases, and modern operating systems of any kind (not to mention real languages) might not even recognize ACP stuff as code - this is the real deal, the real throwback to ancient times. The real code-bummed, bit- aligned, side-effect-code, 2k address blocked, put everything exactly where you want it, where will the heads on the disk be as we are executing *this* routine, custom nitro-fueled funnycar of a system, with only one purpose in life to get from here to there as fast as you can and if it takes a support team of thousands and costs millions of dollars that is small beans... Seriously, things have gotten a *little* more modern from what I am told (I know a few TPF programmers who work for amtrak, marriott, and other places) but this is computing the old fashioned way. Oh yeah, one other thing. Modern stuff can't TOUCH this performance. Relational Data Bases are a joke compared to this. They work on a different time base compared to this. It is, in a strange way, very satisfying to see this solidly 1970s stuff blow the doors off of the new stuff. It won't last forever, but it has lasted a LONG time. >Granted they were all logged into one application more or less, but >the sheer number (particularly since I think it went into production >around 1975?) was impressive, I doubt there are too many single >systems today which you could actually log 15,000 people into >(something would run out or go screwy), let alone give them all a >prompt back quickly even if they were merely hitting RETURN randomly. > >There are papers around describing the engineering of the system >which, as I remember them, describe a completely different era in >systems design and implementation. For example, they knew exactly how >many machine instructions every possible code path took and if you >fixed a bug or whatever which changed that it had to be cleared by >someone(s) who fussed about such things. Imagine filling out paperwork >on the performance impact of adding a test for a null pointer and >waiting for approval and you get the idea. Yep. It is a different world. Doug From jcewing@acm.org Mon Jul 21 10:32:14 1997 Path: stc06.ctd.ornl.gov!news.er.usgs.gov!news1.radix.net!nac!news.maxwell.syr.edu!news-peer.sprintlink.net!news-pull.sprintlink.net!news-in-east.sprintlink.net!news.sprintlink.net!Sprint!206.250.118.17!nntp.earthlink.net!usenet From: jcewing@acm.org Newsgroups: alt.folklore.computers,alt.sys.pdp10 Subject: Re: Why Mainframes? Date: Sun, 20 Jul 1997 13:49:30 -0500 Organization: EarthLink Network, Inc. Lines: 13 Message-ID: <33D25DBA.A38@acm.org> References: <01bc8523$84da9b80$e12185c2@rashid1> <5pkj08$b2c@ss1.digex.net> <1997Jul10.122310.28825@indyvax.iupui.edu> <5qaed0$f0n$1@nntp2.ba.best.com> <5qn82p$h0k$1@nntp2.ba.best.com> Reply-To: jcewing@acm.org NNTP-Posting-Host: 1cust90.max1.irving.tx.ms.uu.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 2.02 (OS/2; I) Xref: stc06.ctd.ornl.gov alt.folklore.computers:163373 alt.sys.pdp10:3486 phil dawkins wrote: > .... Whether a PC server will achieve mainframe performance and > accessibility is a contentious issue. You want my opinion? Of course > it will. PC performance has historically outperformed mainframes of one or two decades earlier, but mainframe technology doesn't stand still either. No doubt PC servers 10 years down the road will rival or surpass mainframes of today, but not the new mainframes of that day. One of the surprising results of past infatuations with C/S and workstations is the the total MIP's in mainframes has continued to increase, and recently the total number of systems has started to increase again as well. -- Joel C. Ewing, Fort Smith, AR jcewing@acm.org From morpheas@seattleu.edu Mon Jul 21 10:36:51 1997 Path: stc06.ctd.ornl.gov!fnnews.fnal.gov!newsfeed.pnl.gov!netnews6.nwnet.net!netnews.nwnet.net!seattleu.edu!mozart-2!morpheas From: Joel Newsgroups: alt.2600,comp.sys.mac.advocacy,alt.folklore.computers,comp.os.linux.advocacy,comp.os.msdos.misc Subject: Re: Windows 95, what a joke. Date: Thu, 17 Jul 1997 17:35:29 -0700 Organization: Seattle University Lines: 27 Message-ID: References: <5gt5c9$3ce4@elmo.cadvision.com> <5gu66d$f0v@mojo.europe.dg.com> <5hevd3$gnn@usenet.srv.cis.pitt.edu> <333c44c4.68306872@news.bright.net> <5htr2i$9d8@yama.mcc.ac.uk> <3346aead.277851239@news> <5ibhgb$piv@gcsin3.geccs.gecm.com> <5inren$cal$1@gail.ripco.com> <5iono9$mfn@crcnis3.unl.edu> <01bc486c$435764c0$f32fbccc@jeffh> <336E280D.191D@pwgsc.gc.ca1> <33739DFF.2985@tcc.tacoma.ctc.edu> <33CCB97A.6B78@spammers.die> <33CD68C1.A42F4CAE@sgi.net> NNTP-Posting-Host: mozart-2.seattleu.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Sender: morpheas@mozart-2 To: Adam Stouffer In-Reply-To: <33CD68C1.A42F4CAE@sgi.net> Xref: stc06.ctd.ornl.gov alt.2600hz:356483 comp.sys.mac.advocacy:255674 alt.folklore.computers:163227 comp.os.linux.advocacy:131431 comp.os.msdos.misc:80257 On Wed, 16 Jul 1997, Adam Stouffer wrote: > Brian Knox wrote: > > > Jerry Cornelius wrote: > > > > > I mean to say. WIN95 is solid. It's the bullshit apps that bomb a > > > system everytime! > > > > > > Jhary-a-Conel > > > http://www.seanet.com/~sizemore/enter.html > > > Mr Freeze > > > > If your OS was solid, bullshit apps wouldn't be able to bomb it. The > > system would bomb the bullshit apps instead, and then continue merrily > > > > on its way. > > Yep, got a point I've seen win95 bluescreen without any apps up. It just's as unstable as system 7 Joel C Laughman morpheas@seattleu.edu http//seattleu.edu/~morpheas/carnival.html