1* Private Network 2* Fast server 3* QoS (Solaris Resource Manager, et al) (if you can't trust your developers not to be stupid and suck down the resources...) 4* Limit the number of Sun Ray appliances per server to something reasonable...
... only to have DIGITAL purchased by Compaq, which was later melded with HP.... to have Tru64 (nee DIGITAL Unix) cannibalized into a new version of HPUX.
You really should check the Seagate 18.2 GB FC-AL disks. They're crap. The firmware is crap, the drive is crap, and the failure rate is WAY WAY WAY too high.
I can't tell you how many times I've seen an entire loop on an A5200 go offline because a single disk was failing.
Piece of..... drives. Not to mention the multi-initiator bug, where the drive locks up both the A and B bus. That does WONDERS for clustered environments.
The active system can't access the disks, so it attempts a failover. Can't deport the disk group, so it shoots itself in the head.... Then the secondary node can't import the disk group either, so it shoots ITSELF in the head, and now you have 2 downed nodes.
Lovely.
Yeah, Sun's storage sucks. It's sucked since the A3500 hit the market. The only thing they sell that isn't garbage is the D1000 (how hard was that to build? It was outsourced anyway...) and the HDS storage (9970/9980)...
Try running a backup server, or anything I/O intensive, across those things.... it's not much better than slowbus (SBus)!
I've actually run one of our 6800's out of I/O... we're being constrained because of the slowwwww PCI slots. Time for us to move it to either an HP rp8420 or an IBM p5 570...
It's not on life support -- not even close. You don't port to a completely new architecture (Superdome/Integrity) for life support purposes.
As for customers -- VMS customers are *extremely* loyal.
We got a call from a major financial institution on 9/11 asking if we had any spare VAXes around. They needed to replace a few hundred(!!) VAX7000s that were lost in the Trade Center.
We had a few in our storage facility (We're on GS1280's right now... bless our developers who have managed to keep their code upgradeable!), but by the time we got back to them, they had found a warehouse in Texas that had a bunch of them...
And of course, HP is making service revenue off each and every one of those VAXes -- as well as the AlphaServers they're migrating to as we speak.
VMS is going to be here for a long, long time. It's stable, it's robust, it's secure. Why not?
Yep, gotta love that DEC support. It's amazing to me how many DECies are still around, now working for HP. In fact, we had a Superdome install about a month ago, and our onsite Alpha support guy ran into someone during the install that he hasn't seen in 10 years -- it was like a reunion!
I'm a UNIX admin, and right now, I'm playing Sun and HP-UX. We've got the Alpha/VMS "Platinum" support model right now on HP-UX, and I have to tell you, it couldn't be better. That "NOTHING is too much trouble" thing is still true. You have to be a Big Fish, and you have to pay for it, but I'm still getting that "Whatever we can do to help means literally WHATEVER" support.
Fantastic stuff; I'm happy to see that HP's willing to adopt that support model (at least for us, anyway) on the Superdome/HPUX side. I'm sure it's causing some infighting between ingrained HPers and DECies, but as the customer, that's not my concern.:-)
1) The "legendary" DEC service & support models. Nothing -- and I mean NOTHING, not even IBM -- can compare. Nobody's support is like DEC's. Their support is SO good, it's absurd. I can really consider the dedicated support team I've got as an extension of my admin staff.
2) Two profitable businesses: Alpha/OpenVMS and NonStop (a/k/a Himalaya). As fashionable as it is to bash VMS, guess what, it's still around, and it's still VERY profitable.
VMS shops will continue to use VMS for a long, long time. In fact, as I recall, DEC/Compaq/HP is obligated to continue support through at least 2017. Cool stuff. (Isn't that when the lights go out on Broadway? Ba-dum-bum.)
NonStop is what runs, well, everything. Most SS7 networks are *highly* dependant on Nonstop. Yeah, sure, it's ridiculously expensive -- but it works. If you need 99.999%+ uptime, nothing else provides it --- not even the mainframe.
If you look at this merger through PC eyeglasses, yeah, it probably doesn't make much sense. But then if you look at it with the enterprise market in mind, it makes LOTS of sense.
Now, I'm not wild about the prospect of using the Itanium chips, but I have to say, the idea of running OpenVMS on the same systems with HP-UX, along with Linux, is definitely cool. Even nicer is that HP-UX (which is arcane in a lot of ways) will get some "real" features like TruClustering. Can't wait to see that!
Interesting times are ahead with HP.... I think they're a real powerhouse, and especially now that the integration of both companies is really rolling along, they're going to be a Big Force in the enterprise space.
I think it's going to come down to IBM and HP. Sun's just dropping the ball on SO many fronts lately (Bring back the Blueprints Engineers!!) that it's hard for me to count them as real players in the market right now.... It's a shame too, I really liked Sun equipment, and *especially* Solaris. But 33mhz PCI buses on your high-end SF25k servers? Give me a break!
Nothing -- and I mean NOTHING -- even comes close to the Holy Trio of Enterprise Systems (HTES)...
VxVM, VxFS, VCS.
Volume Manager, File System, Cluster Server.
They're the gold standard. Can Linux *really* do what VxVM can do?
Don't think so. LVM isn't even close. And that's not a slap against Linux -- AIX's LVM and HP's LVM aren't close either. (Sun's Online Disksuite is just a joke)
VCS? What's better than VCS? Sun Cluster? Yeah, right. MC/SG? Try again. HACMP? No way. All 3 try to do too much. VCS has the right balance of simplicity with flexibility, that it just works.
Seriously, nobody's been able to compete with Veritas. Nothing's even close.
Slashdot Mirror
distinguished name?
dn=via.com,c=US
SecureID.
e cu reid.pdf
Whoo.
Been there, done that.
All it does is make an attack "more" difficult, but nowhere near impossible:
http://www.tux.org/pub/security/secnet/papers/s
... only if it's implemented poorly.
...
...
1* Private Network
2* Fast server
3* QoS (Solaris Resource Manager, et al) (if you can't trust your developers not to be stupid and suck down the resources...)
4* Limit the number of Sun Ray appliances per server to something reasonable
Add those 3 together, and you'll be fine
It's called Sun Ray, and it's here now.
And it ROCKS.
http://wwws.sun.com/sunray/sunray150/index.html
.... except that there is no "." after a NO CARRIER according to the Hayes command set.
</pedantic>
Yeah, it was called "Return to the Minors" or something like that...
:)
As GREAT as Major League was, that really ranks as one of the worst-attempts-at-milking-a-franchise-dry ever.
No, but they should understand the difference between a *real* assembler and one that a textbook author made up!
Yes, they are.
I was horrified when one of the folks I graduated with actually put "SIC/XE Assembler" on their resume:
http://www.cs.siu.edu/~mcglinn/sic/sic.html
Egads. Just proves what they didn't learn.
http://www.apple.com/appleworks/
....
Funny, there seems to be a blurb on the right about "AppleWorks 6.2.2 for Windows"
There ya go.
The spectrograph is what failed; the optics are fine and dandy.
...
We're still going to get nice pretty pictures out of Hubble, just no UV/wavelength pictures
Hubble's hobbled, but still alive and kicking.
TheDraw.
'nuff said.
... only to have DIGITAL purchased by Compaq, which was later melded with HP. ... to have Tru64 (nee DIGITAL Unix) cannibalized into a new version of HPUX.
:(
Lovely.
1 highway, 0 city.
</Wolfcastle>
I call BS.
..... drives. Not to mention the multi-initiator bug, where the drive locks up both the A and B bus. That does WONDERS for clustered environments.
.... Then the secondary node can't import the disk group either, so it shoots ITSELF in the head, and now you have 2 downed nodes.
You really should check the Seagate 18.2 GB FC-AL disks. They're crap. The firmware is crap, the drive is crap, and the failure rate is WAY WAY WAY too high.
I can't tell you how many times I've seen an entire loop on an A5200 go offline because a single disk was failing.
Piece of
The active system can't access the disks, so it attempts a failover. Can't deport the disk group, so it shoots itself in the head
Lovely.
Yeah, Sun's storage sucks. It's sucked since the A3500 hit the market. The only thing they sell that isn't garbage is the D1000 (how hard was that to build? It was outsourced anyway...) and the HDS storage (9970/9980)...
+50, Lemonade through the nose.
One of the funniest responses I've ever read on Slashdot (and I've read lots)
ABSOLUTELY agreed.
I've had the same experience with the XServe support guys. They rock.
I had a disk mirroring issue, and the guys really knew this stuff cold..... And that was the 1st guy I got on the phone!
I don't even get support like that from Sun, with our Platinum contract and everything!
Only if you plug it in.
(in other words, YEAH. But you gotta supply it.)
Steve Jobs did that on March 24, 2001.
What, low uid's impress you or somethin'? ... makes me wish I signed up back when I started READING slushdot, way before 32whateverthehelliam :)
No, I don't need 66MHz. I need PCI-X.
33MHz really is slow these days.
Try running a backup server, or anything I/O intensive, across those things.... it's not much better than slowbus (SBus)!
I've actually run one of our 6800's out of I/O... we're being constrained because of the slowwwww PCI slots. Time for us to move it to either an HP rp8420 or an IBM p5 570...
Somehow, I doubt all that.
...
It's not on life support -- not even close. You don't port to a completely new architecture (Superdome/Integrity) for life support purposes.
As for customers -- VMS customers are *extremely* loyal.
We got a call from a major financial institution on 9/11 asking if we had any spare VAXes around. They needed to replace a few hundred(!!) VAX7000s that were lost in the Trade Center.
We had a few in our storage facility (We're on GS1280's right now... bless our developers who have managed to keep their code upgradeable!), but by the time we got back to them, they had found a warehouse in Texas that had a bunch of them
And of course, HP is making service revenue off each and every one of those VAXes -- as well as the AlphaServers they're migrating to as we speak.
VMS is going to be here for a long, long time. It's stable, it's robust, it's secure. Why not?
Yep, gotta love that DEC support. It's amazing to me how many DECies are still around, now working for HP. In fact, we had a Superdome install about a month ago, and our onsite Alpha support guy ran into someone during the install that he hasn't seen in 10 years -- it was like a reunion!
:-)
I'm a UNIX admin, and right now, I'm playing Sun and HP-UX. We've got the Alpha/VMS "Platinum" support model right now on HP-UX, and I have to tell you, it couldn't be better. That "NOTHING is too much trouble" thing is still true. You have to be a Big Fish, and you have to pay for it, but I'm still getting that "Whatever we can do to help means literally WHATEVER" support.
Fantastic stuff; I'm happy to see that HP's willing to adopt that support model (at least for us, anyway) on the Superdome/HPUX side. I'm sure it's causing some infighting between ingrained HPers and DECies, but as the customer, that's not my concern.
Oops, I meant to break Alpha/OpenVMS and Himalaya out as separate topics.... hence the (3) things that HP got. :-)
I'll tell you EXACTLY what HP got.
.... It's a shame too, I really liked Sun equipment, and *especially* Solaris. But 33mhz PCI buses on your high-end SF25k servers? Give me a break!
3 things:
1) The "legendary" DEC service & support models. Nothing -- and I mean NOTHING, not even IBM -- can compare. Nobody's support is like DEC's. Their support is SO good, it's absurd. I can really consider the dedicated support team I've got as an extension of my admin staff.
2) Two profitable businesses: Alpha/OpenVMS and NonStop (a/k/a Himalaya). As fashionable as it is to bash VMS, guess what, it's still around, and it's still VERY profitable.
VMS shops will continue to use VMS for a long, long time. In fact, as I recall, DEC/Compaq/HP is obligated to continue support through at least 2017. Cool stuff. (Isn't that when the lights go out on Broadway? Ba-dum-bum.)
NonStop is what runs, well, everything. Most SS7 networks are *highly* dependant on Nonstop. Yeah, sure, it's ridiculously expensive -- but it works. If you need 99.999%+ uptime, nothing else provides it --- not even the mainframe.
If you look at this merger through PC eyeglasses, yeah, it probably doesn't make much sense. But then if you look at it with the enterprise market in mind, it makes LOTS of sense.
Now, I'm not wild about the prospect of using the Itanium chips, but I have to say, the idea of running OpenVMS on the same systems with HP-UX, along with Linux, is definitely cool. Even nicer is that HP-UX (which is arcane in a lot of ways) will get some "real" features like TruClustering. Can't wait to see that!
Interesting times are ahead with HP.... I think they're a real powerhouse, and especially now that the integration of both companies is really rolling along, they're going to be a Big Force in the enterprise space.
I think it's going to come down to IBM and HP. Sun's just dropping the ball on SO many fronts lately (Bring back the Blueprints Engineers!!) that it's hard for me to count them as real players in the market right now
Yeah, right.
Nothing -- and I mean NOTHING -- even comes close to the Holy Trio of Enterprise Systems (HTES)...
VxVM, VxFS, VCS.
Volume Manager, File System, Cluster Server.
They're the gold standard. Can Linux *really* do what VxVM can do?
Don't think so. LVM isn't even close. And that's not a slap against Linux -- AIX's LVM and HP's LVM aren't close either. (Sun's Online Disksuite is just a joke)
VCS? What's better than VCS? Sun Cluster? Yeah, right. MC/SG? Try again. HACMP? No way. All 3 try to do too much. VCS has the right balance of simplicity with flexibility, that it just works.
Seriously, nobody's been able to compete with Veritas. Nothing's even close.