Slashdot Mirror
The Last Multics System Decommissioned
Bell Would? writes: "A key feature of the brief news item, 'The end of the Multics era,' in the latest issue of the The Risks Digest is the 'list of goals' Multics had fulfilled which, as the author describes them, are as relevant today as they were 35 years ago." Odd -- I assumed these were all long since junked or put into museums, since my first exposure to the name Multics was in books which spoke mostly in the past tense. That list of goals is one that I hope architecture designers consult frequently.
Yeah, we actually maintained multics for a few years. It was still there when i was done, so i have know idea what happened to it.
I guess there won't be any Real Men around any more. After all, it is a well known fact that Real Men use Multics.
Tarsnap: Online backups for the truly paranoid
"Fortunately for us, Dennis Richie and Ken Thompson decided to pare down some of the features and create a version of "Multics without the balls." Thus Unix was born (the name being a pun on "Multics")."
... need I go on?
And fortunately for us, an army of people have put every one of those features BACK [into Unix]... I don't think anybody would be wanting to run a process swapping OS with a16 bit address space these days... The list of features Richie and Thompson removed include demand paging, dynamic linking, shared memory, memory mapped files,
The big problem with Multics was that is was 20 years ahead of the hardware.
In addition to Unix, I know a significant portion of the Multics staff was instrumental in developing the proprietary OS run by Stratus Computer (www.stratus.com) and I believe Multicians also had a major hand in VMS and later in NT development.
An engineer who ran for Congress. http://herbrobinson.us
I know for a fact that the ABB (Asea Brown Boveri) offices in Columbus, Ohio still have a functional Multics. At least they still did a few weeks ago when my father was there on business...
ICQ# : 30269588
"I used to be an idealist, but I got mugged by reality."
ICQ# : 30269588
"I used to be an idealist, but I got mugged by reality."
That's pretty much what we do now in Linux - when you write it doesn't go to disk, it goes onto memory pages. When you read you're reading from memory pages and if they're not there, they get 'swapped in' from your file using the same mechanism we use for virtual memory, though we bypass the paging hardware in this case (it's faster that way).
Neat idea - but imagine the 32-bit address space crunch happening 20 years ago instead of now :)
We get around that by using disjoint pages of virtual memory mapped into the file's address space with a hash table, so the file has a 44 bit address space - that should be enough for a while. This works well, and doesn't cause virtual memory fragmentation. We'll probably start mapping the files in chunks larger than one page pretty soon.
--
Life's a bitch but somebody's gotta do it.
Perhaps it went onto one of our naval vessels, and collapsed some poor rusted out deck. What a sad end to a noble operating system!
If you are modding me down because you disagree with me, use the "Flamebait" category, not the "Troll" one.
Well my first AC Flame of the day. wow. I think to better state my question (considering the topic of the story seemed to be centered on the longevity of the Multics systems), perhaps I should have asked ... What are the possibilities for developing a *nix, BeOS whatever system with that type of longevity? Everything *nix wise is booming so fast that some distros are skipping version numbers. And BeOS has gone from "The Multimedia OS of the future" to an IA OS! But now that you mention it.. hmm just imagine.. a beowulf cluster of Multics-based Dr Dos machines, wow.... the SETI data I could process!
Are YOU listed?
Ahhh, isn't it nice to see that while the dominant OS may change, some things always remain the same?
"They do not preach that their god will rouse them, a little before the Nuts work loose." Kipling, 'The Sons of Martha'
What's interesting is that Bull (?), the company that owns the rights to the source code, never wanted to release the source code because they claimed they had to "continue to support the few remaining Multics systems in existence" -- they can't possibly be doing this now, so give us the damn source! ;-)
--TheOrangeSquid Is it any wonder things seem so awry? We swim in a sea of confusion and don't have to think to survive
Mmmmm, Apollo. My trusty DN3000 is still sitting about five feet away. I do not fire it up as much as I used to since the disk started making unhappy noises. It is not dead yet, though. :)
I always wondered why MultiCS was still in its server type choices, I didn't even know any were around to this day?
Read my plan to save the Bengals
1) Was this the longest lasting O/S thus far? Anyone know of a production O/S that pre-dates mid-1965 that is still running?
2) Multics died mostly from being proprietary and running on proprietary hardware. The first, the Multicians thought could be surmounted by a gift from the current code owners. The second, alas, was fatal.
The industry's fixation (mostly because of the volume curve) on VonNeuman architechtures that lack any real new features cause us all to not have things we could really really use -- like the ring security that Multics offered which had direct hardware support. Too bad ASICs are not yet dense enough... maybe soon. :-)
--Multics
Something is quite wrong with the Canadian military. First they de-commission Multics. Now I hear that they are getting new helicopters to replace the Sea-Kings. What's next? Our diesel submarine getting replaced? I am starting to see a conspiracy.
Actually, I don't begrudge the millitary for getting new helicopters. They are really needed.
-- Spammers: My E-mail server is in California. Consider yourself warned.
We do that too, it's called mmap. The nice thing is that the same primitives used in mmap (and now swap too... and soon, shm) are also used in read and write. All this in a nice, compact efficient package. Um, as long as you close your eyes and forget that buffer.c exists. :-)
--
Life's a bitch but somebody's gotta do it.
Yes, good old Domain/OS.
That was a nice OS. And the apollo's amazingly quick for what they were (68030's, as I recall).
-- And let there be light... so he fluffed the light spell
I believed the "Hackers" (the book) hype against Multics and heard many an anti-Multics joke throughout the years. But I ended up meeting (at a short-lived contract) a wonderful gentleman who had worked on Multics and was very proud of the accomplishments he'd achieved and what ground Multics had broken.
I gained a new respect for the achievements of the Multics team, and I know today my former coworker and friend would be very unhappy to learn of this news. It really is the end of an era, and we have a lot to thank for what was learned as a result of it.
--- Jump!! Fire!! Bullet time!! - Lego version of the Matrix
Some of the Multicians should do this, just so it's not forgotten. It's still one of the most secure operating systems around.
I was intending to do something like this, but since its already done I will just add a couple of things...
A wide range of system configurations, changeable without system or user program reorganization. Windows: Only three reboots to install a sound card! Linux: Exchange anything but the kernel without rebooting Microkernels: 8-D
Well I take this to mean 'hot plug' (since dorking with the system/modules is 'system reorganization') which as far as I'm aware linux doesn't support but Win 2k if your HW supports it, AIX, and Solaris do.
Hierarchical structures of information for system administration and decentralization of user activities. Not entirely sure what they mean by this...
I think they mean NDS, Active Directory (which is basically LDAP with a bunch of support) and of course LDAP if you are willing to spend the time to get it to support all the cool stuff NDS does .
It was actually the first computer I ever used. My father, a professor of computer science at MIT who did some development on the Multics virtual memory, got me an account.
Unfortunately, he agreed to pay for it. The Multics billing system was the most elabourate I've ever seen, before or since. You were billed for CPU minutes, connect time minutes and I think even a whole bunch of other minutes. As a result, I ran through $150 of computer time in three days, which is not exactly cost-effective, so he wound up getting me a free ITS tourist account.
I don't remember much about it anymore, since it's been years since I've had an account, but I do feel a little nostalgic now that it's gone. Pity no renegade hobbyist could put one together, as some individuals have with ancient PDP-10 systems. I have to assume that the cost of wiping classified data from the systems is sufficiently high that the recycler is the only realistic destination for these ancient systems.
D
----
Sid from User Friendly will be devastated!
The interesting part about the article was that the last Multics machine was being run by the Canadian military! Only in Canada, eh?
BroadbandPig
where can i download this mystical linxu ?
The NSA used to use multics for their mail server. till it got cracked. using old fashion proprietary systems with secure classifications is no justification of lax security.
http://www.iretro.com
http://www.iretro.com
Empeg Kicks Ass
Well, all modern operating systems can do this in theory at least
If you're talking "analogous to power & telephone services", that means to be even resistant to hardware failures. Which means hot-swap disks and CPUs. Certainly Sun systems can do this, except maybe if you lose the drive with the root partition, but I'm not aware of Winderz machines which allow CPU swapping.
Right now, my main Linux server is whining and rumbling like a banshee on testosterone, and it's not the power supply fan, so it must be the old 17 gig hard drive. So it looks like I'll have a few hours of downtime to get a new one in there.
Hierarchical structures of information for system administration and decentralization of user activities. Not entirely sure what they mean by this...
Sounds like NIS on steroids. Or maybe the Windoze Registry without the suckage, and distributed over multiple machines. Or better yet, NetInfo from NeXT/OS-X.
#naabhaprzrag, #sverubfr-000, #agi-fcbafberq, negvpyr[pynff*=' negvpyr-ary-'] { qvfcynl: abar !vzcbegnag; }
Nope, more likely something like the ACLs of NT...
Simple things should be simple, complex things should be possible.
have been holding off a US invasion all this time with MULTICS??? How Embarassing....
Wondering what is replaceing it.... Suddenly the invitation to Microsoft to move north of the border makes sense.
try { do() || do_not(); } catch (JediException err) { yoda(err); }
I've never heard of one at U of C, but that's only since 1995.
However, speaking of old systems, when I was in 2nd year engineering (1996), we had to do C programming on old DECstations running OSF. They finally replaced them in 1998 with RedHat 4.2 or something on P-233's, then switched to WinNT 4 in 1999 fall, then a year later switched back to Debian 2.2.whatever (same hardware all along).
--
Yes, the University of Calgary housed two Multics systems in the 80's: a 6-CPU system and a 1-CPU test system. The company that supported Multics after Honeywell (ACTC) was a spin-off from U of C.
... all very fascinating to a tech-hungry teenager.
:-)
Multics (at UofC) was the first large system I used, and I have many fond memories of it. I attended the Shad Valley (technology + entrepreneurship) summer program in 1984 and spent hours absorbing 'everything Multics'. On-line manuals, pathnames, processes, e-mail, chatting, windowing systems (character-based)
It's interesting to note that Multics underwent a development surge in the early 80's and despite the aging hardware design still had a number of sites at that point (Ford, Canadian defense, US DOD).
I'm sad to see it go, though its time has come (without portability, it was doomed to die with the hardware). I remember touring the U of C computer room when a tech was on site, reportedly doubling the cache *width* while the system remained on-line (I presume he was taking one CPU offline at a time). The LED bargraph pads showing CPU utilization for each processor that were scattered around the room were quite impressive too
...has upgraded it's 5-processor Multics system to multiple handheld devices to "stay ahead, technologically.
Info about said devices is available here.
Karma karma karma karma karmeleon: it comes and goes, it comes and goes.
From the Multics Glossary entry on virtual memory:
I take this to mean that Multics had no read(2) or write(2)... from the application-writer's point of view, the equivalent to these system calls was simple memory access.
Presenting this analogy to programmers as their primary means of file access is different from using such tricks down at a level where (theoretically) no one except kernel programmers needs to know ahout them.
It shows a completely different point of view... instead of "everything is a file", the MULTICS way seems to be "everything is core".
where there's fish, there's cats
I had the "opportunity" to work as a systems operator on *6* Multics systems, from 1986 to 1988. (Yes, I'm listed with Multicians.org.) Your interpretations of some of the goals of the Multics project is somewhat colored by modern technology. Let me explain what some of those goals meant to the Multicians, and why they still aren't met by modern operating systems:
This meant that the entire system was hot swappable: disk drives, CPUs, Memory units, IO units. Of course, your odds of the system surviving the addition or subtraction of any one of these were
This is the hot-swappable hardware thang again. You could add a CPU to a system without interrupting the processing on the rest of the system. System software updates were quite a different matter -- that generally required a system restart, and there were still "system" drives whose failure could cause the entire system to crash.
This refers to classifying information, not filesystems. Multics could run with Classified, Secret, and Top-Secret information (and programs) all co-resident, and without a lower-classification program being able to access higher-classification information. No modern operating system works this way; the set of systems that replaced the Multics group that I worked on was *3* separate Unix networks, one for each security classification.
This refers to the traditional hierarchical file structure, with hierarchical user management thrown in for good measure. What CP/M and MS-DOS stole from Unix, Unix in turn stole from Multics.
In general, Multics achieved its goals, though the cost was too high. More recent operating environments have judged the cost of some of those goals (primarily security) to be so unrealistic as to be completely undesirable. While I think that Multics aimed too high on some goals, I think that too many operating systems (including Linux) aim too low.
Are you moderating this down because you disagree with it,
We call it art because we have names for the things we understand.
Right now, my main Linux server is whining and rumbling like a banshee on testosterone, and it's not the power supply fan, so it must be the old 17 gig hard drive. So it looks like I'll have a few hours of downtime to get a new one in there.
I have a problem with OLD beging used for the label of a 17 gig hard drive. Possibly you could describe an old CDC Wren III 300 Meg drive as old (I still have one in service), but by no stretch of the imagination could I imagine a 17 gig drive as old. Especially interesting considering the topic we are under.
You mention Museums... Where can i find a computer museum or is there one? Besides my closets of course.
"Bye bye the mainframe has died". It's alwawys sad to see a history thing as a live Multics system dying. It did inspire *nix, didn't it? So, we got a lot to thank these machines for, *waves sadfully goodbye*
-Stskeeps, http://unrealircd.com
..for adapting some of this into the *nix world? Or to BeOS? etc.?
Are YOU listed?
You've got it turned around -- Unix is similar to Multics, since Multics came first -- but I'll take you seriously anyways:
On a more humorous note ...
And, last but not least ...
On a historical note, Primos (the Pr1me Operating System), was a much more direct steal from Multics, down to implementing CPL exactly. I learned Primos years before I used Multics, and Multics was merely more difficult to administrate.
Are you moderating this down because you disagree with it,
We call it art because we have names for the things we understand.
Just goes to show how much longevity the original systems had. I can't imagine that the systems of today will still be serving their purposes 35 years into the future. This is a pretty cool testimonial to the time-tested power of the *NIX'es which are mostly based on Multics.
i also had only heard about Multics through incidental mention on other subjects.it could be interesting to study the application of the goals to current systems and see what type of improvements would be possible.
Multics was ahead of it's time. Now It's at the end of it's time. I hope that before I reach the end of my time, I read an article about the last Windows system reaching it's final blue screen.
134340: I am not a number. I am a free planet!
I may be wrong, but I seem to recall a Multics system at U of Calgary (Alberta, Canada) when I was there around 1986. Can anyone confirm or Deny this?
BroadbandPig
Neat idea - but imagine the 32-bit address space crunch happening 20 years ago instead of now :)
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
heh, was Chris Walpole around in those days?
Linux/BSD/UNIX: Check! telnet/ssh and X can make nearly everything network transparent
Windows: Need an extra program like PCanywhere, and even then it's single user. (but isn't m$ fixing this in win2k?)
Well, all modern operating systems can do this in theory at least
Windows: Only three reboots to install a sound card!
Linux: Exchange anything but the kernel without rebooting
Microkernels: 8-D
Windows: NTFS seems to be close enough for most people Linux: ext3 and reiserfs
Windows: Network Neighborhood ought to be enough for anybody!
Linux/UNIX/BSD: NFS, Coda, FTP, scp, etc...
Not entirely sure what they mean by this...
Check.
Windows: IDEs, IDEs and more IDEs.
Linux: Your choice of gcc,emacs,kdevelop,vi, or whatever else you find on freshmeat
Open source!
-- 2 + 2 = 5, for very large values of 2
For those of you who don't know, Multics (Multiplexed Information and Computing Service) is a comprehensive, general-purpose programming system which is being developed as a research project. The initial Multics system will be implemented on the GE 645 computer. One of the overall design goals is to create a computing system which is capable of meeting almost all of the present and near-future requirements of a large computer utility. Such systems must run continuously and reliably 7 days a week, 24 hours a day in a way similar to telephone or power systems, and must be capable of meeting wide service demands: from multiple man-machine interaction to the sequential processing of absentee-user jobs; from the use of the system with dedicated languages and subsystems to the programming of the system itself; and from centralized bulk card, tape, and printer facilities to remotely located terminals. Such information processing and communication systems are believed to be essential for the future growth of computer use in business, in industry, in government and in scientific laboratories as well as stimulating applications which would be otherwise undone.
Because the system must ultimately be comprehensive and able to adapt to unknown future requirements, its framework must be general, and capable of evolving with time. As brought out in the companion papers, this need for an evolutionary framework influences and contributes to much of the system design and is a major reason why most of the programming of the system will be done in the PL/I language. Because the PL/I language is largely machine-independent (e.g. data descriptions refer to logical items, not physical words), the system should also be. Specifically, it is hoped that future hardware improvements will not make system and user programs obsolete and that implementation of the entire system on other suitable computers will require only a moderate amount of additional programming.
Take a look at the way multics handled dynamic linking. Calling a non-existant symbol caused the process to suspend. Someone could write a replacement for the missing subroutine and resume the process.
Multics had a hierarchical administration system, unix has sudo(1).
Multics was designed for large systems. Unix was designed for small systems and grew large.
Personally I'm hoping for next Tuesday to be the dooms day for Windows.
How we know is more important than what we know.
Yes, I worked on Apollo workstations in the 80s at Birmingham University in England. They were effing fantastic for their time. The group I worked for would do large non-linear finite element analysis of plastic deformation (e.g. forged con-rods) using parallel fortran jobs spread across all the workstation cpus on the network. Although this was slow, it was still much faster than submitting the jobs to the University computer centre which was running, yes, you guessed it, MULTICS!
Last time I talked to my old supervisor they had transitioned OSes as follows
MULTICS->DOMAIN/OS->Irix->Linux
He seemed to be quite pleased they had skipped the Windows phase entirely.
48 bits? I recall 36 bits, split into 18 bit segment and 18 bit offset. Because of the memory mapping, standard files were limited to 2**18 words, approx 1 MB. Larger files were "MSF", or multi-segment files, which had a sort of builtin directory structure, which not all tools supported.
I will say, that the I/O speed impressed me!
(P.S. I used Multics at the U.S. Geological Survey arouund 1980. It was a fun machine.)
I have often wondered why we so stubbornly worked so hard to make the system survive. My own take on it is that we were young and wanted to make a dent in the psyche of the industry which in those bad old days was incredibly shortsighted. And I think we did.
Heh.
How we know is more important than what we know.
It had TONS and TONS of features (look here for a list), but unfortunately it took too long to implement, and when these features were finally implemented, the resulting OS was so damn slow nobody wanted to use it. Consequently it was canned.
Fortunately for us, Dennis Richie and Ken Thompson decided to pare down some of the features and create a version of "Multics without the balls." Thus Unix was born (the name being a pun on "Multics").
And we all lived happily ever after!!
Hopefully the legacy found in Unix and to a larger degree in Domain/OS (anyone else remember Apollo?) will live on.
Who, me?
I looked at the sorce code these guys wrote in PL/1 in the 70's and early 80's and to my surprise, they docuemented clear, I expected there code to be massive pile of shit with no comments but I was already, I am also very impressed that they wrote 3000 pages of specs before starting the implementation.
Checkout for some source. http://www.multicians.org/multics-source.html
------ Curiosity killed the cat. {satisfaction brought it back | it didn't die ignorant | lack of it is killing mankind
To think that people are exclaiming at Multics finally ceasing! There is quite an old OS out there still... UNIX. Of course not in its original AT&T code...
;)
I can draw three of morals from this:
1) A good overall design never grows old. Not to mention excellent foresight by its desginers. These ideas that the Multics archatecs have thought of are still the model of a good server/client OS.
2) A good overall design is built to last! If the Canadians just took out a Multics system last month, there must of been a reason why it was still in commission for this long of time. Most certainly it could be argued that it was because there wasn't funding... maybe, but that's aside from this point. If it works, why fix it?
3) If Bill Gates didn't drop out of college in the 1970s, and actually studied a Multics or an UNIX system... maybe his OS would be good? Or was it inevitable?
Karma whorin' since 1999
Following the links at Multicans.org took me to: The US National Security Agency's DOCKMASTER machine was shut down in March, 1998, after repeated extensions. The hardware from this site, except for the hard drives, was given to the National Cryptologic Museum, which in turn loaned it permanently to the Computer Museum History Center in Mountain View, California
hope this helps