I'd like to be enlightened as to the benefits of running a Microsoft cluster in a scientific environment. Really. You've given me nothing new to compare my experiences running Microsoft-based clusters versus my experience running Linux clusters.
I suppose if you REALLY ran windows clusters for a living you'd know which tools. That would be Cluster Administrator. Which is higher-level than the "mind-boggling" array of files in/etc. But it's a whole lot less effort than getting an MCSE.
Please, provide some facts. Otherwise your response is, well, zealous and overwrought.
Wrong. I get paid a decent salary running Windows clusters.
but you're writing them off as idiots because they used the MS package. Wrong. This article is an advertisement disguised as news.
Nevermind they might be saving money Might is a pretty big maybe.... I *know* a Linux-based cluster costs less. Especially as we get into 2008 pricing.
in the long run In the long run we are all dead. Please define the time frame in which chosing Microsoft saves money. More facts please.
by paying less people to administrate it Will they actually get the job done? It's impossible to know either way unless they cure cancer with the thing. Facts please.
MS tools get the job done. Which tools are these? Gui's? I imagine the average win32 admin WOULD be at a loss in a Linux cluster./etc/ is just mind-boggling!
However, I would be surprised if this cost them much at all besides time. They are probably a large enough customer that they get many MS products and services for free. Except it isn't "free." Someone way outside your pay grade signed a contract and might have paid Microsoft. (or not if the customer is a good PR win)
In addition, the publicity for MS makes it worth it to MS to offer tons of incentives. This story is an advertisement disguised as news.
I work at an EDU org and MS pricing is a lot less than retail... a lot less. And a Linux-based cluster is even less. I don't see any motivation to maximize the educational institutions resources in your response. None!
Now more than ever, I'm concerned about the basic capabilities of American research institutions maximize their resources. Sigh...
1. The url provided has no prices whatsoever. Imagine that!
2. What are the license constraints in this academic pricing? You know, razors are darn cheap compared to the price of razor blades. That first line of coke is pretty cheap too.
Seriously though. I don't know if I should be concerned or not. Part of being young is working with the mistaken belief they can become millionaires working for World Peace. (or whatever their heart's desire) Part of it also is they don't comprehend the complexity of the underlying delivery systems.
Now, if the Bank of Mom and Dad does not sustain their magical thinking, then they'll get in line pretty fast once they have to choose between washing their clothes or eating.
and I have a very hard time believing most of the claims of fact in this story.
"When we deployed Windows on our cluster, which has more than 1,000 nodes, we went from bare metal to running the Linpack benchmark programs in just four hours,"
Hmmm. And what installer was this? Is it available commercially? How much is the license for the version with this mythical four-hour installer?
"The performance of Windows HPC Server 2008 has yielded efficiencies that are among the highest we've seen for this class of machine," Pennington said.
What "class" would that be? I imagine it would explicitly exclude Free clusters.
One should question whether the efficacy of any institution/research project using their grant money wisely given the amount of money required to fulfill Microsoft's licensing requirements.
Furthermore, If research projects are actually considering wasting their grant dollars on Microsoft licenses, then the outlook for American R&D is grim.
If MS would simplify it... Show me the business case for a company the size of MS to get involved in this. There isn't one. RSA was as big as they got and they weren't strong enough to stay out on their own.
Nevermind the fact that it is simple!!! Compared to all of the time and effort it took to learn how to use a computer, it's ridiculously simple. People just don't want to pay for it or even feel the need to get it.
and it takes some legislation that some people don't like to raise awareness?
I've used PGP encrypted attachments for years. Works great.
SMTP over TLS is a good start. TLS is supposed to replace ssl, but who knows when that will happen. If you want to get mad-tricky, there's stunnel.
VOIP over TLS is another good start. It's not widely implemented, but widely available.
Chat can also be handled over TLS. Either through a VOIP softphone which is widely available or possibly XMPP.
If I offered TLS services for chat, VOIP and email, for a $5/month, exactly how many takers would there be? Not enough to be worth my trouble and the crypto-overhead when you get into lots of users. Which is why it isn't widely available.
As for Reader j1976, getting free advice on./ is not a good start. Let me give you a tip. Your politically most powerful users in your organization will probably kill the project as soon as they understand it will change the way they work. Nevermind that it would be a minor change. That's not the point. The point is they don't want to be on the receiving end of this change. Which, will **irreparably** harm your career prospects.
Claiming that the IT people who are snooping at stuff they know they shouldn't is purely a lack of planning on behalf of management is just trying to act like these people involved don't bear any individual responsibility. I can't agree with that assertion.
Well said. I totally agree that there is an individual responsibility in there that *should* kick in. That doesn't make a story people want to read.
At some hypothetical point, information becomes very sensitive. That information needs to be accessible by a select few. I was trying to make that point by painting the issue as black/white.
GPL or not, based on Sun's behavior around their Indiana/OpenSolaris (That can't be called OpenSolaris because Sun said no)will pretty much kill whatever advantages going GPL/community driven may bring.
Articles like this one just perpetuate numerous cultural and organizational phenomena of taking risks then blaming someone else for losing the bet. Management's role in creating the situation is totally ignored by most of the comments to my initial reply.
Because, some people aren't supposed to be seeing certain things
Running with that assumption for a moment, most of the replies totally ignore the *fact* that Management is unwilling to pay OR EVEN CONSIDER using a system that would guard those "certain things."
-PGP encrypt attachments? No way. -Password on a zipped archive? Probably not. -A system-wide approach via PKI? Not on your life.
Management has *intentionally* set themselves up for failure and they blame the IT worker? This is the classic case of sh!t rolling downhill.
Maybe I'm missing the point but I don't see where there is an issue.
In nearly all IT environments, either you trust your IT staff, or you have some killer PKI. Reality suggests management in the typical company wouldn't pay for or be bothered to use, so we're back to IT having super-snooping powers.
Step 1: Make two things that are not comparable in any way sound the same. Apple-vs-Microsoft Microsoft-vs-Novell are not comparable situations. Period.
Step 2: Use false logic to support your argument.
Unfortunately, the harm Microsoft (via Novell) represents to the commercial OSS community is obvious to anyone who can maintain a few minutes of objectivity while examining the facts.
Moderators clearly fell for your abuse of logic. Congratulations.
As someone that runs a VOIP server, I can speak from limited experience.
1. Unlike email, The offender needs a block of voip numbers to do any meaningful spitting. Those blocks aren't as costless as sending spam. Let's argue for a minute they don't need blocks. The VOIP server should not be allowed to process more than ~2 calls out per number. That's a configuration issue. On proprietary voip server software, I don't know if that's possible, but on openser it is.
2. This _should_ be the responsibility of the VOIP host, except we know that most current providers won't do it for free. It can, and should be automated. ex. *69 reports the call as spam. Even if the call is coming from a peering host, the source can be halted swiftly.
3. DB queries on call volume should identify the offender within 30 minutes anyway.
The article is an advertisement disguised as news.
At best, they would argue they are a payment processor. A Bank does many other types of cash and lending operations.
FYI: You can start your own payment processing service like any business. Where you are going to get screwed, is when you need to connect to banks for electronic funds transfers. ACH rates are low, but getting into the network is not cheap or easy.
Payment processing is one of the monopolies absolutely no one cares about despite the broad harm to consumers. F*i*r*s*t D,a,t,a and V^i^s^a.
Why is this happening now instead of years ago? The harm Intel has created is egregious and has been obvious for a long time.
Did someone at the White House get up on the wrong side of bed one morning? Maybe the White House didn't like what the Executives were doing with their political action funds?
There's a very good reason they chose those ads. Do you think the non-profits have the resources to litigate this? What would they litigate exactly?
It's a big win for BT, and probably Comcast here in the U.S because there are so many legal issues that none of the harmed companies can afford to litigate it. It would be a career's-worth of work for both sides, with the ISP getting the vast majority of their wishes met either through litigation or purchasing legislation.
I had another thought. What if the plan is to aggregate advertisers? This would destroy the sites that makes any money based on advertising, or have them go to BT for their ad revenue.
The -eventual- outcome would be every ISP that can afford to do it will create something vaguely like television only with some extra free info out there where they can't sell adverts.
In the time that it takes for the case to make it's way through court, they could make plenty of progress toward this end without consequences.
This is how it works people. Smaller companies hit on a good idea all of the time. Every once in a while, the idea appeals to a very large group of consumers. Big companies just wait. Sometimes for quite a while.
All big companies, Microsoft included, have one guy running around corporate going "This UMPC thing is going to be big! We need to target it." This guy is completely ignored because there's no market data and Management pretty much ignores him because he's saying stuff like this all of the time.
Meanwhile, Asus figured out how to deliver the goods on the cheap. Microsoft's Asus rep ignored Asus's info about UMPC's because Microsoft's rep is used to waiting for corporate to deliver the pinata filled with money.
When Asus gets things rolling, Management panics because their high-priced market research has just come back with a new report saying cheap UMPC's are growing into a huge market. Some ass-kisser in Marketing is then tasked with stomping on the Linux Distro by preparing a pinata filled with money to deliver to Microsoft's Asus rep.
There's more waiting. More market research. More waiting. Presentations. Approvals. Meetings. More waiting.
Microsoft corporate delivers pinata to Asus rep. Microsoft's OS is then available as a SKU worldwide ~1-3 years after Asus's product launch.
1. Contact production company for info on who to send spec scripts. 2. Write and submit spec scripts.
This is where it gets sketchy. If they are reputable, they'll figure out some way to throw you a bone. You might not get paid, but you might get a credit. If they are not reputable, they'll steal your ideas outright. It's a coin toss.
Slashdot Mirror
I'd like to be enlightened as to the benefits of running a Microsoft cluster in a scientific environment. Really. You've given me nothing new to compare my experiences running Microsoft-based clusters versus my experience running Linux clusters.
I suppose if you REALLY ran windows clusters for a living you'd know which tools. /etc. But it's a whole lot less effort than getting an MCSE.
That would be Cluster Administrator. Which is higher-level than the "mind-boggling" array of files in
Please, provide some facts. Otherwise your response is, well, zealous and overwrought.
So basically you have no facts,
Wrong. I get paid a decent salary running Windows clusters.
but you're writing them off as idiots because they used the MS package.
Wrong. This article is an advertisement disguised as news.
Nevermind they might be saving money
Might is a pretty big maybe.... I *know* a Linux-based cluster costs less. Especially as we get into 2008 pricing.
in the long run
In the long run we are all dead. Please define the time frame in which chosing Microsoft saves money. More facts please.
by paying less people to administrate it
Will they actually get the job done? It's impossible to know either way unless they cure cancer with the thing. Facts please.
MS tools get the job done. /etc/ is just mind-boggling!
Which tools are these? Gui's? I imagine the average win32 admin WOULD be at a loss in a Linux cluster.
I needed a laugh. Thanks.
However, I would be surprised if this cost them much at all besides time. They are probably a large enough customer that they get many MS products and services for free.
Except it isn't "free." Someone way outside your pay grade signed a contract and might have paid Microsoft. (or not if the customer is a good PR win)
In addition, the publicity for MS makes it worth it to MS to offer tons of incentives.
This story is an advertisement disguised as news.
I work at an EDU org and MS pricing is a lot less than retail ... a lot less.
And a Linux-based cluster is even less. I don't see any motivation to maximize the educational institutions resources in your response. None!
Now more than ever, I'm concerned about the basic capabilities of American research institutions maximize their resources. Sigh...
Hmmm... Reasonable eh?
1. The url provided has no prices whatsoever. Imagine that!
2. What are the license constraints in this academic pricing? You know, razors are darn cheap compared to the price of razor blades. That first line of coke is pretty cheap too.
I know the price of my preferred clusters, $0. Usage constraints? None. http://debianclusters.cs.uni.edu/index.php/Main_Page
I loose my mind!
Seriously though. I don't know if I should be concerned or not. Part of being young is working with the mistaken belief they can become millionaires working for World Peace. (or whatever their heart's desire) Part of it also is they don't comprehend the complexity of the underlying delivery systems.
Now, if the Bank of Mom and Dad does not sustain their magical thinking, then they'll get in line pretty fast once they have to choose between washing their clothes or eating.
and I have a very hard time believing most of the claims of fact in this story.
"When we deployed Windows on our cluster, which has more than 1,000 nodes, we went from bare metal to running the Linpack benchmark programs in just four hours,"
Hmmm. And what installer was this? Is it available commercially? How much is the license for the version with this mythical four-hour installer?
"The performance of Windows HPC Server 2008 has yielded efficiencies that are among the highest we've seen for this class of machine," Pennington said.
What "class" would that be? I imagine it would explicitly exclude Free clusters.
One should question whether the efficacy of any institution/research project using their grant money wisely given the amount of money required to fulfill Microsoft's licensing requirements.
Furthermore, If research projects are actually considering wasting their grant dollars on Microsoft licenses, then the outlook for American R&D is grim.
In less time than it took to read your post.
Corporate agriculture makes Microsoft's anti-competitive behavior look saintly by comparison to their own shenanigans.
Some of the subtler, but most important battles won by would be the conspicuous absence of recommended daily sugar intake being listed by the FDA.
http://www.accidentalhedonist.com/index.php/2006/01/24/tariffs_and_subsidies_the_literal_cost_o
http://thehill.com/business--lobby/lobby-league-25-agriculture-2004-12-08.html
I could, actually, really really, provide "encrypted" email, VOIP and chat over TLS. Easily since I already do it for myself and some family members.
If I charged $5/month for 5 addresses I'm pretty sure I wouldn't get enough takers to make it worth my time.
I'd like to hear otherwise.
It's too complex for most.
No. They don't have an urgent need. They'd do it if there was an urgent reason for it.
sending my and all of their friends their public key
That's what a key server is for. http://packages.debian.org/etch/onak
If MS would simplify it...
Show me the business case for a company the size of MS to get involved in this. There isn't one. RSA was as big as they got and they weren't strong enough to stay out on their own.
Nevermind the fact that it is simple!!! Compared to all of the time and effort it took to learn how to use a computer, it's ridiculously simple. People just don't want to pay for it or even feel the need to get it.
and it takes some legislation that some people don't like to raise awareness?
./ is not a good start. Let me give you a tip. Your politically most powerful users in your organization will probably kill the project as soon as they understand it will change the way they work. Nevermind that it would be a minor change. That's not the point. The point is they don't want to be on the receiving end of this change. Which, will **irreparably** harm your career prospects.
I've used PGP encrypted attachments for years. Works great.
SMTP over TLS is a good start. TLS is supposed to replace ssl, but who knows when that will happen. If you want to get mad-tricky, there's stunnel.
VOIP over TLS is another good start. It's not widely implemented, but widely available.
Chat can also be handled over TLS. Either through a VOIP softphone which is widely available or possibly XMPP.
If I offered TLS services for chat, VOIP and email, for a $5/month, exactly how many takers would there be? Not enough to be worth my trouble and the crypto-overhead when you get into lots of users. Which is why it isn't widely available.
As for Reader j1976, getting free advice on
Claiming that the IT people who are snooping at stuff they know they shouldn't is purely a lack of planning on behalf of management is just trying to act like these people involved don't bear any individual responsibility. I can't agree with that assertion.
Well said. I totally agree that there is an individual responsibility in there that *should* kick in. That doesn't make a story people want to read.
At some hypothetical point, information becomes very sensitive. That information needs to be accessible by a select few. I was trying to make that point by painting the issue as black/white.
Well said sir.
GPL or not, based on Sun's behavior around their Indiana/OpenSolaris (That can't be called OpenSolaris because Sun said no)will pretty much kill whatever advantages going GPL/community driven may bring.
Articles like this one just perpetuate numerous cultural and organizational phenomena of taking risks then blaming someone else for losing the bet. Management's role in creating the situation is totally ignored by most of the comments to my initial reply.
Because, some people aren't supposed to be seeing certain things
Running with that assumption for a moment, most of the replies totally ignore the *fact* that Management is unwilling to pay OR EVEN CONSIDER using a system that would guard those "certain things."
-PGP encrypt attachments? No way.
-Password on a zipped archive? Probably not.
-A system-wide approach via PKI? Not on your life.
Management has *intentionally* set themselves up for failure and they blame the IT worker? This is the classic case of sh!t rolling downhill.
Maybe I'm missing the point but I don't see where there is an issue.
In nearly all IT environments, either you trust your IT staff, or you have some killer PKI. Reality suggests management in the typical company wouldn't pay for or be bothered to use, so we're back to IT having super-snooping powers.
Step 1: Make two things that are not comparable in any way sound the same. Apple-vs-Microsoft Microsoft-vs-Novell are not comparable situations. Period.
Step 2: Use false logic to support your argument.
Unfortunately, the harm Microsoft (via Novell) represents to the commercial OSS community is obvious to anyone who can maintain a few minutes of objectivity while examining the facts.
Moderators clearly fell for your abuse of logic. Congratulations.
They are called smart card chips. You can get them packaged as surface mount too. It's got all crypto goodness one needs.
Except management can't comprehend decent crypto, so we'll have a few more decades of encryption keys stored on disk.
As someone that runs a VOIP server, I can speak from limited experience.
1. Unlike email, The offender needs a block of voip numbers to do any meaningful spitting. Those blocks aren't as costless as sending spam. Let's argue for a minute they don't need blocks. The VOIP server should not be allowed to process more than ~2 calls out per number. That's a configuration issue. On proprietary voip server software, I don't know if that's possible, but on openser it is.
2. This _should_ be the responsibility of the VOIP host, except we know that most current providers won't do it for free. It can, and should be automated. ex. *69 reports the call as spam. Even if the call is coming from a peering host, the source can be halted swiftly.
3. DB queries on call volume should identify the offender within 30 minutes anyway.
The article is an advertisement disguised as news.
At best, they would argue they are a payment processor. A Bank does many other types of cash and lending operations.
FYI: You can start your own payment processing service like any business. Where you are going to get screwed, is when you need to connect to banks for electronic funds transfers. ACH rates are low, but getting into the network is not cheap or easy.
Payment processing is one of the monopolies absolutely no one cares about despite the broad harm to consumers. F*i*r*s*t D,a,t,a and V^i^s^a.
Why is this happening now instead of years ago? The harm Intel has created is egregious and has been obvious for a long time.
Did someone at the White House get up on the wrong side of bed one morning? Maybe the White House didn't like what the Executives were doing with their political action funds?
Why now?
There's a very good reason they chose those ads. Do you think the non-profits have the resources to litigate this? What would they litigate exactly?
It's a big win for BT, and probably Comcast here in the U.S because there are so many legal issues that none of the harmed companies can afford to litigate it. It would be a career's-worth of work for both sides, with the ISP getting the vast majority of their wishes met either through litigation or purchasing legislation.
I had another thought. What if the plan is to aggregate advertisers? This would destroy the sites that makes any money based on advertising, or have them go to BT for their ad revenue.
The -eventual- outcome would be every ISP that can afford to do it will create something vaguely like television only with some extra free info out there where they can't sell adverts.
In the time that it takes for the case to make it's way through court, they could make plenty of progress toward this end without consequences.
Only complacent management at Microsoft.
Here's the loong tale of how this stuff happens.
This is how it works people. Smaller companies hit on a good idea all of the time. Every once in a while, the idea appeals to a very large group of consumers. Big companies just wait. Sometimes for quite a while.
All big companies, Microsoft included, have one guy running around corporate going "This UMPC thing is going to be big! We need to target it." This guy is completely ignored because there's no market data and Management pretty much ignores him because he's saying stuff like this all of the time.
Meanwhile, Asus figured out how to deliver the goods on the cheap. Microsoft's Asus rep ignored Asus's info about UMPC's because Microsoft's rep is used to waiting for corporate to deliver the pinata filled with money.
When Asus gets things rolling, Management panics because their high-priced market research has just come back with a new report saying cheap UMPC's are growing into a huge market. Some ass-kisser in Marketing is then tasked with stomping on the Linux Distro by preparing a pinata filled with money to deliver to Microsoft's Asus rep.
There's more waiting. More market research. More waiting. Presentations. Approvals. Meetings. More waiting.
Microsoft corporate delivers pinata to Asus rep. Microsoft's OS is then available as a SKU worldwide ~1-3 years after Asus's product launch.
Can anyone tell me if this will hit Konqueror in the new KDE?
I've been running KDE 4.1 through debian experimental. Buggy right now, but no show-stoppers. KDE and the new Konqueror are surprisingly fast.
1. Contact production company for info on who to send spec scripts.
2. Write and submit spec scripts.
This is where it gets sketchy. If they are reputable, they'll figure out some way to throw you a bone. You might not get paid, but you might get a credit. If they are not reputable, they'll steal your ideas outright. It's a coin toss.