I live in a pretty questionable apartment complex in a rent-controlled (read: again questionable) part of a large urban area not too far from RedHat. I'd be happy to perform this service for the lowly donation of a linux-happy webcam. I even have DSL for the purpose.
BTW, I don't necessarily mean to associate RedHat with urban badness; in fact they are about 18 miles from me:).
Has any thought been put into doing a "reproducibilty challenge" against Gartner? That is,
Ask Gartner their methodology,
Document it so that it can be reproduced;
Make sure and get Gartner to say "yep that's how we did it".
Dan&Co reproduces the methodology and compares the numbers
Might be overly scientific. Maybe consider it an "Open Source" version of market research because of the peer review and verification. Some may think this approach has no place in the market research area.
But, I (for one) think it would be interesting to call Gartner's Bluff (if indeed that's what this is). I personally place them (and ZD, etc) into the Shill category.
...apart from the acting performances of Ewan MacGregor and Nicole Kidman (he's great,
she's not).
Ok, it's official. Katz is gay.
(attempt at humor here, folks).
the SGI 1450 Server as 16 procs
2733/16 = 170(TCP) per CPU
Um, the SGI's page for the 1450 states that it's only a 4-way machine.
That yields 683.25 TPC/CPU for Linux, versus 212.375 TPC/CPU for win2k on the Compaq hardware. In other words, Linux is more than three times faster according to my calculations.
One of us has our numbers wrong... I've got SGI's website to back mine up... Am I missing something here?
Regarding w2k datacenter, it only supports up to 32 CPUs. Anybody know of any shipping x86 hardware at that scale? Last I heard Unisys had the only box, to be resold by Compaq... Until Compaq backed out (don't mean to troll here, but conspiracies abound that MS pushed it that way in light of w2k dc's "shortcomings" - YMMV).
[1]... What the heck happens to all those outgoing chips?
eBay baby, eBay. Seriously, for a company such as Intel that works on the principals of planned obsolesence, there must be hundreds (nay, thousands) of slightly older chips *somewhere*.
[2]... recycling some of the 300's and 400's and making a Quad motherboard running say a PIII 400...
There's no such thing as the PIII 400. Also (as someone else mentioned) you can't get n-way SMP with PII's unless N=2. Finally, price out some quad-proc mobos... expensive... Note also that a 1.33ghz thunderbird machine will, for all intents and purposes, stomp a 4-way 400mhz machine for LOTS cheaper.
Thrashing isn't necessarily that specific. Thrashing might mean the head is dancing alot, doing unnecessary seeks... It could be multiple re-reads, as in the case of a pending hardware failure or poorly-tuned buffers... etc. The Jargon File seems to agree with both of us.
Actually the LA on this machine generally hangs around 3~5. It's the typical thing, a webserver running apache+mod_perl+mysql, lots and lots of hits/day. The load is mostly MySQL handling CGI-sourced queries.
I'm most impressed with the disk I/O I guess. Ever seen disks being worked, but the word "thrashing" never came to mind? The E250 is nice for that anyway, what with all the internal storage available.
I agree wholeheartedly that Solaris "ramps up" more nicely than Linux, though.
Dude, the C64 (Commodore) was truly advanced. It had an awesome SID chip that could do all these wonderful things, and the machine was only 1mhz (.999082 in the UK b/c 50hz!).
But I'm not here to talk about the C64.
There was a program for the 5.25 disk drive that played "A bicycle built for two" by vibrating the heads at varying frequencies. It was a machine language program downloaded into the drive's 2kb buffer, which redirected the boot (IPL?) routine- the code was downloaded to the drive, then the drive was reset.
Its last task after playing the music was to restore the original boot vector, and reset again!
This was undoubtedly inspired by (written by?!!?) the same guys who would cause old IBM drum disks to walk across the floor.
There was also a hack by which you could place a normal audio casette into the Commodore C2n Datasette drive, and read the varying audio pitches as rudimentary digital data- and output it via the SID chip to play audio tapes. Boy did the quality suck! But you could recognize the tune being played, barely.
It has fueled an entire sector of the internet, by giving people who don't know code a
chance to still have a semi-dynamic site. Not everyone wants to learn php/asp/perl/whatever to build a simple site.
There is nothing at all simple about building a dynamic site. Or to respond properly to your assertion, dynamic sites really don't fall into the category of "simple".
My question is about advertising. Linux in general (and RedHat in particular) is in the minds, discussions, and hands of a lot of people right now, especially techies. In fact, you might say that the current popular growth of Linux relies on word-of-mouth among techies and developers.
Many of us feel like it's ready for the enterprise, and have even done some enterprise deployments (see also: IBM). However, it can be a hard sale, because many of the corporate types ("suits" in the present vernacular) just don't have the offline exposure or introduction to Linux. I stress offline here.
I guess what I'm getting at is, how do you feel about conventional marketing/advertising, most notably television or mainstream magazine advertising? I truly believe the Linux enterprise setting would benefit greatly from a television commercial or advert in Forbes or WSJ. Linux has a strong foothold in the techie community, but I'm finding that most managers, CEO's, etc don't have it much on their radar because of a lack of this secondary (primary?) exposure, which seems like a cheap way to generate massive non-techie awareness if done properly. Think of SAS, SAP, Intel, Sun, etc. and the impact of their television commercials (the Sun starship commercials are particularly entertaining, and have been mentioned to me several times in business discussions). Novell too, except that their commercials are a little goofy.
You probably shouldn't be running bind (or anything else). Linux's security problems are almost always created by people leaving stuff
up/on/open when they don't need to.
These "people" are you and me, the admins. This problem is clearly the admin's fault.
Insert standard "wish-the-distros-would-wise-up-and-ship-closed-by -default-installations" thought here...
There is very little truth in your statement these days. On most recent distros you have to choose explicitly to be a server. If you don't, you have to explicitly choose to install and enable BIND. Truth be known, I doubt there are very many KDE workstations out there running named.
No, the blame lies in lazy (or nonexistant?) sysadmins. Let's face it; why is your server running BIND if it doesn't need to (you chose it from the install...)? If the machine is a nameserver, then when the advisory came out in January, did you patch up right away? If not, WHY NOT?. The vendors got updated RPMs and whatnot out fairly quickly.
For the non-existant admin problem, things like the Redhat network will help tremendously.
Not trying to flame here, but your ranting sounds like the parents who blame high-school shootings on video games and movies, when they should be pointing in the mirror. To all the slack admins out there: Enough of this sh*t. Suck it up and do your damn jobs.
FWIW, installs are getting very savvy these days, taking up the slack for the poor job a lot of admins out there are doing; check out RH's latest beta (wolverine?) install - it does ipchains config during the install.
I can see one reason for Sun doing this, and that's the fact that gray market hardware is killing their new hardware sales.
But now, why buy a $400 Ultra 1 (with only 143mhz CPU, etc) when you can spend about twice that and get the latest and greatest entry level workstation? (ignoring the value of SCSI and SBUS)
Bitchin' Hemos,
Thanks for all the details. When the chip part numbers are mentioned in the front-page summary, by Gods this is truly News for Nerds!
Proof that sometimes, the Slash editors/moderators really do read us nerds like open books, think like us, etc.
If I were you, I'd get Dell (or the reseller/vendor, whoever) to drop off a couple of identical machines, configure one with RH6.2 and one with NT, and let you guys test them. Do not be afraid to tweak the Apache server based on your experience and knowledge. That will show management where their skillsets lie.
Run one for a week or two, move the content, and run the other for a couple of weeks. Then bring up the MS licensing costs, the Apache statistics on web presence, etc, along with an "Oh by the way I'm a Unix guy" and see what they say.
If Dell refuses to supply the test machines, make sure and bring that up to your management- Explain that basically the vendor is unwilling to justify their claim. Then you might be able to pick another vendor such as IBM or Compaq. I understand that Compaq has a fairly liberal test/loan program for such things.
Slashdot Mirror
I live in a pretty questionable apartment complex in a rent-controlled (read: again questionable) part of a large urban area not too far from RedHat. I'd be happy to perform this service for the lowly donation of a linux-happy webcam. I even have DSL for the purpose.
:) .
BTW, I don't necessarily mean to associate RedHat with urban badness; in fact they are about 18 miles from me
Thanks,
DragonWyatt
Now we just need Buck Rogers to return with that gorgeous sidekick he had ...
Ah, yes, Erin Gray.
I would drink her bath water.
- The training is paid for
- Other expenses (travel, food, lodging) are paid for
- We get normal pay for the time.
A pretty sweet deal, me thinks. It's nice also that it's done by # days, rather than cost, etc.The guy wants a WAN router. Therefore he NEEDS a PCI slot for the WAN card.
Quit suggesting all these embedded solutions that don't offer WAN functionality (or the expandability needed to provide it).
- Ask Gartner their methodology,
- Document it so that it can be reproduced;
- Make sure and get Gartner to say "yep that's how we did it".
- Dan&Co reproduces the methodology and compares the numbers
Might be overly scientific. Maybe consider it an "Open Source" version of market research because of the peer review and verification. Some may think this approach has no place in the market research area.But, I (for one) think it would be interesting to call Gartner's Bluff (if indeed that's what this is). I personally place them (and ZD, etc) into the Shill category.
Thoughts anyone?
...apart from the acting performances of Ewan MacGregor and Nicole Kidman (he's great, she's not).
Ok, it's official. Katz is gay.
(attempt at humor here, folks).
Mea Culpa, apparently this is not a single machine...
the SGI 1450 Server as 16 procs
2733/16 = 170(TCP) per CPU
Um, the SGI's page for the 1450 states that it's only a 4-way machine.
That yields 683.25 TPC/CPU for Linux, versus 212.375 TPC/CPU for win2k on the Compaq hardware. In other words, Linux is more than three times faster according to my calculations.
One of us has our numbers wrong... I've got SGI's website to back mine up... Am I missing something here?
Regarding w2k datacenter, it only supports up to 32 CPUs. Anybody know of any shipping x86 hardware at that scale? Last I heard Unisys had the only box, to be resold by Compaq... Until Compaq backed out (don't mean to troll here, but conspiracies abound that MS pushed it that way in light of w2k dc's "shortcomings" - YMMV).
[1] ... What the heck happens to all those outgoing chips?
... recycling some of the 300's and 400's and making a Quad motherboard running say a PIII 400...
eBay baby, eBay. Seriously, for a company such as Intel that works on the principals of planned obsolesence, there must be hundreds (nay, thousands) of slightly older chips *somewhere*.
[2]
There's no such thing as the PIII 400. Also (as someone else mentioned) you can't get n-way SMP with PII's unless N=2. Finally, price out some quad-proc mobos... expensive... Note also that a 1.33ghz thunderbird machine will, for all intents and purposes, stomp a 4-way 400mhz machine for LOTS cheaper.
Note:
Not for sale or use in California.
Thrashing isn't necessarily that specific.
Thrashing might mean the head is dancing alot, doing unnecessary seeks... It could be multiple re-reads, as in the case of a pending hardware failure or poorly-tuned buffers... etc.
The Jargon File seems to agree with both of us.
Actually the LA on this machine generally hangs around 3~5. It's the typical thing, a webserver running apache+mod_perl+mysql, lots and lots of hits/day. The load is mostly MySQL handling CGI-sourced queries. I'm most impressed with the disk I/O I guess. Ever seen disks being worked, but the word "thrashing" never came to mind? The E250 is nice for that anyway, what with all the internal storage available.
I agree wholeheartedly that Solaris "ramps up" more nicely than Linux, though.
Put Linux on an E450 and compare it with Solaris on an E450 and you would get a much better comparison.
I admin an E250 running Linux, and in my experience, it blows Solaris (2.6 anyway) away on the same hardware.
Dude, the C64 (Commodore) was truly advanced. It had an awesome SID chip that could do all these wonderful things, and the machine was only 1mhz (.999082 in the UK b/c 50hz!).
But I'm not here to talk about the C64.
There was a program for the 5.25 disk drive that played "A bicycle built for two" by vibrating the heads at varying frequencies. It was a machine language program downloaded into the drive's 2kb buffer, which redirected the boot (IPL?) routine- the code was downloaded to the drive, then the drive was reset.
Its last task after playing the music was to restore the original boot vector, and reset again!
This was undoubtedly inspired by (written by?!!?) the same guys who would cause old IBM drum disks to walk across the floor.
There was also a hack by which you could place a normal audio casette into the Commodore C2n Datasette drive, and read the varying audio pitches as rudimentary digital data- and output it via the SID chip to play audio tapes. Boy did the quality suck! But you could recognize the tune being played, barely.
It has fueled an entire sector of the internet, by giving people who don't know code a chance to still have a semi-dynamic site. Not everyone wants to learn php/asp/perl/whatever to build a simple site.
There is nothing at all simple about building a dynamic site. Or to respond properly to your assertion, dynamic sites really don't fall into the category of "simple".
Check out this guy's work... He's hacked in H.323 support for ethereal, and provides source and all.
Previously the only product that could do true H.323 decodes was the Shomiti (which is a damn fine piece of equipment, by the way...)
Hi Bob and greetings from home (RTP, NC).
My question is about advertising. Linux in general (and RedHat in particular) is in the minds, discussions, and hands of a lot of people right now, especially techies. In fact, you might say that the current popular growth of Linux relies on word-of-mouth among techies and developers.
Many of us feel like it's ready for the enterprise, and have even done some enterprise deployments (see also: IBM). However, it can be a hard sale, because many of the corporate types ("suits" in the present vernacular) just don't have the offline exposure or introduction to Linux. I stress offline here.
I guess what I'm getting at is, how do you feel about conventional marketing/advertising, most notably television or mainstream magazine advertising? I truly believe the Linux enterprise setting would benefit greatly from a television commercial or advert in Forbes or WSJ. Linux has a strong foothold in the techie community, but I'm finding that most managers, CEO's, etc don't have it much on their radar because of a lack of this secondary (primary?) exposure, which seems like a cheap way to generate massive non-techie awareness if done properly. Think of SAS, SAP, Intel, Sun, etc. and the impact of their television commercials (the Sun starship commercials are particularly entertaining, and have been mentioned to me several times in business discussions). Novell too, except that their commercials are a little goofy.
Your thoughts?
You probably shouldn't be running bind (or anything else). Linux's security problems are almost always created by people leaving stuff up/on/open when they don't need to.
y -default-installations" thought here...
These "people" are you and me, the admins. This problem is clearly the admin's fault.
Insert standard "wish-the-distros-would-wise-up-and-ship-closed-b
There is very little truth in your statement these days. On most recent distros you have to choose explicitly to be a server. If you don't, you have to explicitly choose to install and enable BIND. Truth be known, I doubt there are very many KDE workstations out there running named.
No, the blame lies in lazy (or nonexistant?) sysadmins. Let's face it; why is your server running BIND if it doesn't need to (you chose it from the install...)? If the machine is a nameserver, then when the advisory came out in January, did you patch up right away? If not, WHY NOT?. The vendors got updated RPMs and whatnot out fairly quickly.
For the non-existant admin problem, things like the Redhat network will help tremendously.
Not trying to flame here, but your ranting sounds like the parents who blame high-school shootings on video games and movies, when they should be pointing in the mirror. To all the slack admins out there: Enough of this sh*t. Suck it up and do your damn jobs.
FWIW, installs are getting very savvy these days, taking up the slack for the poor job a lot of admins out there are doing; check out RH's latest beta (wolverine?) install - it does ipchains config during the install.
I can see one reason for Sun doing this, and that's the fact that gray market hardware is killing their new hardware sales.
But now, why buy a $400 Ultra 1 (with only 143mhz CPU, etc) when you can spend about twice that and get the latest and greatest entry level workstation? (ignoring the value of SCSI and SBUS)
http://rute.raleigh.nc.us/dist (Not an official mirror though, just three T1's :)
Bitchin' Hemos,
Thanks for all the details. When the chip part numbers are mentioned in the front-page summary, by Gods this is truly News for Nerds!
Proof that sometimes, the Slash editors/moderators really do read us nerds like open books, think like us, etc.
Thanks, again!
Have your managers read this headline, and see how it turns out...
Not really related to performance... Well, maybe the world's economic performance *cough* *cough*.
If I were you, I'd get Dell (or the reseller/vendor, whoever) to drop off a couple of identical machines, configure one with RH6.2 and one with NT, and let you guys test them. Do not be afraid to tweak the Apache server based on your experience and knowledge. That will show management where their skillsets lie.
Run one for a week or two, move the content, and run the other for a couple of weeks. Then bring up the MS licensing costs, the Apache statistics on web presence, etc, along with an "Oh by the way I'm a Unix guy" and see what they say.
If Dell refuses to supply the test machines, make sure and bring that up to your management- Explain that basically the vendor is unwilling to justify their claim. Then you might be able to pick another vendor such as IBM or Compaq. I understand that Compaq has a fairly liberal test/loan program for such things.
Imagine a Beowulf cluster of those things!
(Score: -1, Flamebait)
and was originally intended to get some Funny mods.
But, it seems as though either
Guess my attempt at being a karma whore backfired eh?