Think you're safe when you've got your computer room all tidy and clean?
Forget it. We had a problem with the airconditioning at a medium size company in Delft which had it's heat-outlets on the top of the roof.
These outlets were not so well protected to cold as was shown two years ago when, after a freezingly cold weekend we came into the server room and it was really boiling hot. Problem was the huge ventilators on the roof were stuck frozen.
It was the only time we had all windows and doors open in the middle of winter. But this cold could well have started a fire.
If you want maximum performance and your OS of choice (as a games developer) is a worthless performer but still the OS all your potential costumers run you choose for no security to get the maximum!
If you want maximum performance on a linux machine you can choose to run your program in a stripped down version of your OS with only the features enabled that you want. This way you can keep your system secure and play games.
If windows is stripped up in modular design you could run a 'secure' version for games and still use the newest bloated windows version.
The point here is that no matter what law gets passed we'll always find a way to get around it.
Do you really believe the record labels and film producing companies can get enough momentum to create an un-crackable copyrights protection system? I don't.
Maybe the big companies can pay expensive lawsuits, but I'm sure not going to buy a harddisk or other media that check's the content of whatever I want to put on it.
Picking out Microsoft software for my company.
on
Clever New Windows Worm
·
· Score: 2, Interesting
In 1997 (I think it was could have been 1998 though) the company I work for Delft Hydraulics used Z-mail as the windows platform e-mail client (they used popmail, a text based e-mail client on dos).
I was presented the task of picking out a browser and an e-mail client for the windows95 platform we were preparing to roll out (about 400 computers used by the people that design dykes and harbours for places all over the world).
I knew some software but to be fair I started looking around for all kinds of e-mail packages and browsers. Z-mail was not really an option because it was unstable and required a lot of ram. After playing around with some five or six different e-mail packages the choices became evident.
The advantage of having a browser e-mail combination ruled out all of the separate e-mail programs, not that I found a lot of great ones. (Pegasus, Z-mail, pine, IMC and Eudora where all missing some functionality I whished for our company.)
So the choice was between Microsoft's Internet Explorer in combination with Outlook Express (I never considered Outlook an option since we use sendmail for mail exchange from the early beginnings of the internet in the 80's) or Netscape Communicator (including Navigator, Mail, Calendar and some more stuff).
I summed up the advantages and disadvantages for all products and stated that the software of my choise was the Netscape package.
But, my superiors ruled out Netscape. They did not want to pay $50,- per computer for 'just a browser and an e-mail package' when they could get Internet Explorer and Outlook Express for 'free'. Back then I was in no position to tell them the $50,- was really worth not using all software of one vendor. Today I could, but not back then. So am I to blaim for getting Outlook Express into the company?
1 month after we started to roll out windows95 everywhere the Netscape Communicator package was suddenly available at no cost. But by then Netscape had lost and Microsoft had put it's monopoly foot deep into our company.
We are still using windows95 with Microsoft Office and Internet Explorer and Outlook Express to this very day. All email virus and worm checking is performd by our e-mail server and a strong firewall in combination with PC viruschecking software should keep browser virus out.
Sure, 'the client' or 'the boss' want to have something they think works. But the important thing is: you know better than they do!
So, you kindly advise them to use the 'right thing'. In the end, it's cheaper to use a un*x platform for most of the jobs, because (you know the drill) un*x is more stable, better to understand, gives you more insight into what software does what. And with Linux you can even fix the software your software depends on because the source is available!
"The customer is always right, but wants to be advised by you!" If the costumer whants to use Microsoft software, sure, give it to him. But it might be trivial to give them an alternative that works better because the system below is better.
Want an example? Try giving the costumer that wants a web-based application an IIS solution based on ASP and as a comparison show them what a computer with less processingpower/memory/diskspace/etc. and apache/php can do! With less code!
Freedom is a state of mind, not something you can buy.
To your opinion, is a High Availability cluster a paralell system?
By HA cluster I mean a set of computers bundled together to build up redundancy to protect the application(s) running on it against hardware failure.
I think this protection gets more important the bigger you build your paralell systems. What would the value of a paralell system be if one of it's million parts brings it down completly?
To take this drift even further, I think that for large dimension paralell systems there is no way avoiding the NUMA approach. Performance is just one of the problems you'll face when building a paralell system, even though it may be the most important reason for starting to build them.
In the end it all comes down to scalability and the cost of expanding.
So the best I can make out of your comment (and the website for that matter) is that it is a piece of software that takes the fuss out of loadbalancing over a number of linux machines?
Then what about the performance of this stuff?
And is the design of this software interesting enough to support it?
From the information on the site I'm not clear what the intention of this software is. Is it:
a. a way to create a huge compute node (as in Beowulf cluster)
or is it:
b. a way to create a scalable and robust service node (as in a google cluster)?
I think it could be quite interesting to use this software, but to get a large userbase (and thus robust and well debugged software) it seems to me that there should be a well defined goal.
I'm not trying to post a troll, but I'm trying to find out if:
- this software would make hardware failures have less inpact on running programs like database servers or large simulation jobs?
- this software has a greater impact on performance than MPI alike solutions
Sure they are working on it, but even on/. the only news about Office suites is about M$'s Office 10 and.NET nonsense.
I don't want Solaris to be shipped with my desktop, I want Solaris to be shipped with an Open Source desktop. And one that is a bit more easy to configure (for the enterprize) than CDE wich I think is a load of cr*p.
Sure, I want SUN to build some kind of system that:
Runs X11 applications
Is a lot more friendly to network bandwith
Is opensource
Is build to be ported (ie can be compiled to run on linux/windows/mac/hp/sgi without much trouble)
Runs 'whatever the new protocol name may be' apps that use modern hardware to the best
I think SUN is one of the companies out there that has the momentum to do so. And maby while they are at it they might implement a new type of NFS too, one that has better file-locking, beter security, beter quota support, etc.
I would be glad if SUN took just a little more time to build NEW things on the software side, I think they can do it, and I think they should.
You are right, it's more of the same 'corporate BS'. And I wonder if we should even discuss these sort of matters, but then again, they are bashing eachother with with 'technical details' and that could be interesting.
In the end the 'corporate BS manager' decides what is being programmed by the 'corporate BS programmers' and what is not.
So Microsoft doesn't get it huh?
Ok, fair enough for me.
But what about SUN?
Please blame me for not reading enough and not asking around enough, but wasn't SUN going to:
1. Break staroffice as a package apart and building it into an opensource office package made up of CORBA objects so it could be used in different GNOME applications?
2. When will solaris be shipped with GNOME as the default desktop?
3. When will SUN ship it's systems with a protocol a bit less bloated than X? (Ica?)
And as I heared someone say a few weeks ago, who cares what the glue is, as long as it works. IE, it doesn't matter if you use XML, EJB, Perl, Java, C, or any other cross platform tool to build your software just as long as the software does what you expect it to.
Who cares, Patents are just a plain pain in the ass. In most of the cases they are couterintuitive and protect the rights of the companies that have 'big money'.
But on the other hand it's quite fun to see that M$ is losing at a game they prove to be quite good at time after time.
For as far as the name X-box goes, who cares what it is called, as long as what's inside works.
But opening a way to buy music while not letting the 'monopolist' record companies be the only ones to distribute it may make very well sense. Music that's not as good as other music may have a lower price. Plus, using the internet as a distribution medium is much cheaper than CD's that need packaging and handling.
MD5 may not be the real means to track users or the real source of the files, but it may surely improve the quality of the files that you get from others.
Using MD5 you can know for sure if your file is complete and was not corrupted in traffic.
Sure, clustering aint just Beowulf, and even then Beowulf is not the only High Performance solution.
But even in High Performance solutions availability and scalability are things that are not to be forgotten. (Unless you don't mind your High Performance cluster to crash every week or so due to harddisk failures and overheating Pentium chips.)
To come back to the submission question, to my opinion a Distribution for a Beowulf cluster should have:
- a means to automate installation completely. If you miss this it would take you a lot of time to install a new machine each time one of the machines in your (500 node) cluster crashes.
- an easy way to update your nodes (for the same reason)
- a clear and understandable filesystem layout, that also protects your nodes from the clustered processes (you wouldn't want your/var to be filled up any time a job runs out of its limits)
- hardware support for the devices of your choice (which include hardware raid mirrored disks, gigabit ethernet cards, fast io devices)
- a good 'out of the box' security policy
Now I don't know what distro has these features, but I might have to know somewhere soon in the future as I try to find alternatives to the really expensive O2000 10proc r10000 HPC cluster and the evenly expensive SUN E3500 / A3500 HA cluster we use at my job.
Besides that I think the point noted above are also true for a distribution that just has to provide a platform for a serious bussiness server.
For as far as I'm concerned there are some key functions I mis in scalability for the linux solution (or they just exist and I haven't looked good enough):
- a filesystem that is journaled, life growable, has exellent performance on >1 TerraByte sizes and can be attached to two or more machines enabling failover (like veritas vxfs) (could coda help?)
- an architecture that has the capability of real number crunching (like the O2000/O3000) while maintaining reliability and low prices (maybe alpha's are a solution here, or I just have to cool down and settle for 1000Mhz Pentium IV machines)
Slashdot Mirror
Think you're safe when you've got your computer room all tidy and clean?
Forget it. We had a problem with the airconditioning at a medium size company in Delft which had it's heat-outlets on the top of the roof.
These outlets were not so well protected to cold as was shown two years ago when, after a freezingly cold weekend we came into the server room and it was really boiling hot. Problem was the huge ventilators on the roof were stuck frozen.
It was the only time we had all windows and doors open in the middle of winter. But this cold could well have started a fire.
If you want maximum performance and your OS of choice (as a games developer) is a worthless performer but still the OS all your potential costumers run you choose for no security to get the maximum!
If you want maximum performance on a linux machine you can choose to run your program in a stripped down version of your OS with only the features enabled that you want. This way you can keep your system secure and play games.
If windows is stripped up in modular design you could run a 'secure' version for games and still use the newest bloated windows version.
The point here is that no matter what law gets passed we'll always find a way to get around it.
Do you really believe the record labels and film producing companies can get enough momentum to create an un-crackable copyrights protection system? I don't.
Maybe the big companies can pay expensive lawsuits, but I'm sure not going to buy a harddisk or other media that check's the content of whatever I want to put on it.
In 1997 (I think it was could have been 1998 though) the company I work for Delft Hydraulics used Z-mail as the windows platform e-mail client (they used popmail, a text based e-mail client on dos).
I was presented the task of picking out a browser and an e-mail client for the windows95 platform we were preparing to roll out (about 400 computers used by the people that design dykes and harbours for places all over the world).
I knew some software but to be fair I started looking around for all kinds of e-mail packages and browsers. Z-mail was not really an option because it was unstable and required a lot of ram. After playing around with some five or six different e-mail packages the choices became evident.
The advantage of having a browser e-mail combination ruled out all of the separate e-mail programs, not that I found a lot of great ones. (Pegasus, Z-mail, pine, IMC and Eudora where all missing some functionality I whished for our company.)
So the choice was between Microsoft's Internet Explorer in combination with Outlook Express (I never considered Outlook an option since we use sendmail for mail exchange from the early beginnings of the internet in the 80's) or Netscape Communicator (including Navigator, Mail, Calendar and some more stuff).
I summed up the advantages and disadvantages for all products and stated that the software of my choise was the Netscape package.
But, my superiors ruled out Netscape. They did not want to pay $50,- per computer for 'just a browser and an e-mail package' when they could get Internet Explorer and Outlook Express for 'free'. Back then I was in no position to tell them the $50,- was really worth not using all software of one vendor. Today I could, but not back then. So am I to blaim for getting Outlook Express into the company?
1 month after we started to roll out windows95 everywhere the Netscape Communicator package was suddenly available at no cost. But by then Netscape had lost and Microsoft had put it's monopoly foot deep into our company.
We are still using windows95 with Microsoft Office and Internet Explorer and Outlook Express to this very day. All email virus and worm checking is performd by our e-mail server and a strong firewall in combination with PC viruschecking software should keep browser virus out.
I also own a company, and I work for one.
Sure, 'the client' or 'the boss' want to have something they think works. But the important thing is: you know better than they do!
So, you kindly advise them to use the 'right thing'. In the end, it's cheaper to use a un*x platform for most of the jobs, because (you know the drill) un*x is more stable, better to understand, gives you more insight into what software does what. And with Linux you can even fix the software your software depends on because the source is available!
"The customer is always right, but wants to be advised by you!" If the costumer whants to use Microsoft software, sure, give it to him. But it might be trivial to give them an alternative that works better because the system below is better.
Want an example? Try giving the costumer that wants a web-based application an IIS solution based on ASP and as a comparison show them what a computer with less processingpower/memory/diskspace/etc. and apache/php can do! With less code!
Freedom is a state of mind, not something you can buy.
To your opinion, is a High Availability cluster a paralell system?
By HA cluster I mean a set of computers bundled together to build up redundancy to protect the application(s) running on it against hardware failure.
I think this protection gets more important the bigger you build your paralell systems. What would the value of a paralell system be if one of it's million parts brings it down completly?
To take this drift even further, I think that for large dimension paralell systems there is no way avoiding the NUMA approach. Performance is just one of the problems you'll face when building a paralell system, even though it may be the most important reason for starting to build them.
In the end it all comes down to scalability and the cost of expanding.
So the best I can make out of your comment (and the website for that matter) is that it is a piece of software that takes the fuss out of loadbalancing over a number of linux machines?
Then what about the performance of this stuff?
And is the design of this software interesting enough to support it?
Is anyone really using it?
From the information on the site I'm not clear what the intention of this software is. Is it:
;)
a. a way to create a huge compute node (as in Beowulf cluster)
or is it:
b. a way to create a scalable and robust service node (as in a google cluster)?
I think it could be quite interesting to use this software, but to get a large userbase (and thus robust and well debugged software) it seems to me that there should be a well defined goal.
I'm not trying to post a troll, but I'm trying to find out if:
- this software would make hardware failures have less inpact on running programs like database servers or large simulation jobs?
- this software has a greater impact on performance than MPI alike solutions
- this software has security impact of any sort?
Just my 2 eurocents
I don't want Solaris to be shipped with my desktop, I want Solaris to be shipped with an Open Source desktop. And one that is a bit more easy to configure (for the enterprize) than CDE wich I think is a load of cr*p.
Sure, I want SUN to build some kind of system that:
- Runs X11 applications
- Is a lot more friendly to network bandwith
- Is opensource
- Is build to be ported (ie can be compiled to run on linux/windows/mac/hp/sgi without much trouble)
- Runs 'whatever the new protocol name may be' apps that use modern hardware to the best
I think SUN is one of the companies out there that has the momentum to do so. And maby while they are at it they might implement a new type of NFS too, one that has better file-locking, beter security, beter quota support, etc.I would be glad if SUN took just a little more time to build NEW things on the software side, I think they can do it, and I think they should.
Is anyone using this? Why is there not more Hype to OpenOffice?
You are right, it's more of the same 'corporate BS'. And I wonder if we should even discuss these sort of matters, but then again, they are bashing eachother with with 'technical details' and that could be interesting.
In the end the 'corporate BS manager' decides what is being programmed by the 'corporate BS programmers' and what is not.
So Microsoft doesn't get it huh?
Ok, fair enough for me.
But what about SUN?
Please blame me for not reading enough and not asking around enough, but wasn't SUN going to:
1. Break staroffice as a package apart and building it into an opensource office package made up of CORBA objects so it could be used in different GNOME applications?
2. When will solaris be shipped with GNOME as the default desktop?
3. When will SUN ship it's systems with a protocol a bit less bloated than X? (Ica?)
And as I heared someone say a few weeks ago, who cares what the glue is, as long as it works. IE, it doesn't matter if you use XML, EJB, Perl, Java, C, or any other cross platform tool to build your software just as long as the software does what you expect it to.
Who cares, Patents are just a plain pain in the ass. In most of the cases they are couterintuitive and protect the rights of the companies that have 'big money'.
But on the other hand it's quite fun to see that M$ is losing at a game they prove to be quite good at time after time.
For as far as the name X-box goes, who cares what it is called, as long as what's inside works.
You have a point. Stealing things is not right.
But opening a way to buy music while not letting the 'monopolist' record companies be the only ones to distribute it may make very well sense. Music that's not as good as other music may have a lower price. Plus, using the internet as a distribution medium is much cheaper than CD's that need packaging and handling.
MD5 may not be the real means to track users or the real source of the files, but it may surely improve the quality of the files that you get from others.
Using MD5 you can know for sure if your file is complete and was not corrupted in traffic.
Sure, clustering aint just Beowulf, and even then Beowulf is not the only High Performance solution.
/var to be filled up any time a job runs out of its limits)
But even in High Performance solutions availability and scalability are things that are not to be forgotten. (Unless you don't mind your High Performance cluster to crash every week or so due to harddisk failures and overheating Pentium chips.)
To come back to the submission question, to my opinion a Distribution for a Beowulf cluster should have:
- a means to automate installation completely. If you miss this it would take you a lot of time to install a new machine each time one of the machines in your (500 node) cluster crashes.
- an easy way to update your nodes (for the same reason)
- a clear and understandable filesystem layout, that also protects your nodes from the clustered processes (you wouldn't want your
- hardware support for the devices of your choice (which include hardware raid mirrored disks, gigabit ethernet cards, fast io devices)
- a good 'out of the box' security policy
Now I don't know what distro has these features, but I might have to know somewhere soon in the future as I try to find alternatives to the really expensive O2000 10proc r10000 HPC cluster and the evenly expensive SUN E3500 / A3500 HA cluster we use at my job.
Besides that I think the point noted above are also true for a distribution that just has to provide a platform for a serious bussiness server.
For as far as I'm concerned there are some key functions I mis in scalability for the linux solution (or they just exist and I haven't looked good enough):
- a filesystem that is journaled, life growable, has exellent performance on >1 TerraByte sizes and can be attached to two or more machines enabling failover (like veritas vxfs) (could coda help?)
- an architecture that has the capability of real number crunching (like the O2000/O3000) while maintaining reliability and low prices (maybe alpha's are a solution here, or I just have to cool down and settle for 1000Mhz Pentium IV machines)