1 Dec
1999
1 Dec
'99
6:45 a.m.
--- CLASSICCMP(a)trailing-edge.com wrote:
If you want to see slow, remember the process of
building standalone
backup on TU58's on a 11/750 or 11/730? We're talking most of a *day*!
No wonder we've tried to erase those times from our memories :-).
Ah... you needed the scripts I wrote eleven years ago... they would
build a console TU58 with EXCHANGE and sling the files out to the tape
in the right order. The benefit was that apparently the microprocessor
in the TU58 (at least on the 11/730, don't know about the 11/750) would
buffer the directory so that the only seeks were to *read* files, not to
*locate* the files.
Our 11/730 reboot time went from 18 minutes to under six with properly
constructed tapes. It became a function of the serial line speed and
tape velocity, not of random seek times on a 250-foot tape. You heard the
drive go and go and go with occasional pauses at each new file. Every
time I rebooted I was glad I'd taken an afternoon to automate the building
process (which I had to re-do with every VMS major update).
Boy... lots of VAX stuff flying the past few days. I'm going to have to
fire something up and remind myself why I don't use 1980-vintage hardware
on a daily basis. It was fun, but *damn* it was slow. I got a lot of reading
done back in those days... seven hour offline backups, five hour upgrades...
oh, yeah... the "good" old days. ;-) At least the stuff stayed up for more
than ten minutes. Our record was 45 days between VAX reboots, but only because
we were developing software and would have periods of several reboots per day
to clear device drivers. Since then, I've seen uptimes on the cluster at
Lucent
measured in months.
-ethan
=====
Infinet has been sold. The domain is going away in February.
Please send all replies to
erd(a)iname.com
__________________________________________________
Do You Yahoo!?
Thousands of Stores. Millions of Products. All in one place.
Yahoo! Shopping: http://shopping.yahoo.com
1 Dec
1 Dec
7:37 a.m.
Boy... lots of VAX stuff flying the past few days.
I'm going to have to
fire something up and remind myself why I don't use 1980-vintage hardware
on a daily basis. It was fun, but *damn* it was slow. I got a lot of reading
done back in those days... seven hour offline backups, five hour upgrades...
oh, yeah... the "good" old days. ;-) At least the stuff stayed up for more
Remember the positive things you're saying are more about the OS, than
about the hardware. Although you still have to have stable hardware. Get
a modern Alpha, and you can have both OpenVMS and Speed. Of course I'm
wondering how fast a VAX in the 20-30 VUPs range feels compared to a slower
(ie affordable) Alpha.
than ten minutes. Our record was 45 days between VAX
reboots, but only
because
we were developing software and would have periods of several reboots per day
to clear device drivers. Since then, I've seen uptimes on the cluster at
Lucent
measured in months.
-ethan
I've heard rumors of at least one cluster with an uptime measured in years.
Me, I'm only at 18 days, but then I just put the system into production,
and most of the time there is only one machine in the cluster (yeh, that
doens't make a lot of sense).
Zane
| Zane H. Healy | UNIX Systems Adminstrator |
| healyzh(a)aracnet.com (primary) | Linux Enthusiast |
| healyzh(a)holonet.net (alternate) | Classic Computer Collector |
+----------------------------------+----------------------------+
| Empire of the Petal Throne and Traveller Role Playing, |
| and Zane's Computer Museum. |
| http://www.aracnet.com/~healyzh/ |
8:39 a.m.
On Wed, 01 Dec 1999, Zane H. Healy wrote:
Remember the positive things you're saying are more
about the OS, than
about the hardware. Although you still have to have stable hardware. Get
a modern Alpha, and you can have both OpenVMS and Speed. Of course I'm
wondering how fast a VAX in the 20-30 VUPs range feels compared to a slower
(ie affordable) Alpha.
I haven't run any direct timings, but my 3100/80 seems much more responsive
than an Alpha 3000/300LX running VMS v7.2.
I've heard rumors of at least one cluster with an
uptime measured in years.
Me, I'm only at 18 days, but then I just put the system into production,
and most of the time there is only one machine in the cluster (yeh, that
doens't make a lot of sense).
A former place of employment had uptimes in excess of 2 years on a pair of
very heavily used 11/750s running VMS v4.7. Ridiculously long uptimes aren't
rare with VMS systems.
-Dave McGuire
2 Dec
2 Dec
12:07 a.m.
we were developing software and would have periods of
several reboots per day
to clear device drivers. Since then, I've seen uptimes on the cluster at
Lucent
measured in months.
-ethan
Nice to see Lucent still has at least one Vax. Wish I could get an
account or had the time to get my Vaxstation at home up.
Bell Labs is pretty much Suns, PC's an occasional SGI or HP and
no visible DEC. Boy the times have changed. I was the DEC install guy
who installed 11/780's at the labs back in '81.
Most Vax sites I worked on measured uptimes in months to years except
for scheduled downtimes (most often system reconfigurations to add more
hardware) or software upgrades.
Bill
---
bpechter@shell.monmouth.com|pechter@pechter.dyndns.org
Three things never anger: First, the one who runs your DEC,
The one who does Field Service and the one who signs your check.
9156
days inactive
9157
days old
3 comments
4 participants
participants (4)
-
ethan_dicks＠yahoo.com -
healyzh＠aracnet.com -
mcguire＠neurotica.com -
pechter＠pechter.dyndns.org