I've pretty much given up on everything else. SuSE is by far the best desktop distro I've run across, I'll be upgrading to 9.1 within the week (I buy one of their distros once a year or so).
However, everyone's so called "Server" installations just manages to piss me off. I don't want a GUI. I don't want a whole bunch of crap that all these commercial distros (including RedHat and SuSE) seem to believe I need.
Slackware is the only Linux distro I put on my servers anymore. I control exactly what's on them, and that's the way I like it. Except for certain servers running OpenBSD, all my workhorse servers have Slackware.
This was originally developed while he (Gary Riley) worked for NASA at the Johnson Space Cener. It was available in source form since before I started working with it in 1993.
As a proud patriot and republican voter I fully and wholly disagree with you.
>> And how bad is really some collateral damage?
If you really need an answer to this, there are solutions for your problems, I recommend investigating them. Start with therapy and medication.
>> Casualties makes the enemy frightned and less willing to fight.
This is true when you're killing enemy soldiers. When you start killing their wives and children, you tend to piss them off.
>> And the enemy should expect some collateral damage when they start a war, collaterall damage will in fact make them less willing to start a war next time.
Ok, this is just nonsense. With only one super power on the planet now (that would be us), no one at a national level is going to start a war against us. We started this war in Iraq, and the objective was to liberate the people, not obliterate them (oh yeah, and look for WMD, whatever).
Like it or not,it is likely that we'll be starting a couple more wars in the future. Ostensibly against nations who get caught harboring terrorists. And again, the goal will not be to hurt the people, but remove the current government.
What your theories are on who we choose to attack and who we negotiate with when the majority of nations over there have terrorist ties, don't interest me.
We are a technology development company focusing on the intellectual
property aspects of innovation in the information, sales and services
sectors of business.
Or more clearly stated: we are an intellectual property management company focusing on beating other companies to the USPTO with braindead obvious inventions related to the internet and more specifically e-commerce.
Did anyone notice the company's email address?
mercado37@yahoo.com
Gotta love them companies leading the world into the new age *sigh*
I don't think we have that much to worry about. The more crap they pull, the more people they piss off. $20 for a CD? Thanks! May I have another? Personally, I haven't expanded my music collection (CD, MP3, whatever) since Napster shutdown. It's too hard to find anything good these days, and I'm not wasting $20 on crap (or something I don't know).
Is it any wonder my interest in new music is dwindling to nothing? And I'm sure I'm not the only one.
I get an annual allocation of options from my employer. It's a large established company, trading publically. The number of options is fine, the percentage of the company is nil. Before the market downturn (that hit EVERYONE) my options were valuing up to a rather significant bonus, since last year however, they're not worth a whole lot.
As a large established company, they're not going anywhere, so I can hold them pretty much as long as I work there. If I take another job, I've gotta cash them quickly, or lose them.
Obviously, there's no way to use them as income. But if the stock goes up again (before I leave), I cash out and get a nice bonus. OTOH, if I leave before they're worth something, I value them up, and build it into a signing bonus or extra salary *grin*. Either way, it's supplemental income, at best.
Bret
Integrity is the key, once you can fake that....
(--BOFH)
Communication of information (songs, pictures, documents, credit card numbers, purchase orders, etc) from one entity (Person, Business, Cowboy Neal, whatever) directly to another over the internet is the INTERNET (surprise)!
For example, I take pictures of my family and make them available on the internet (lets call the software I use to do this a "web server"). Next, I provide my friends and relatives with a client (we'll call this software a "web browser") and instruct them on how to use it to see my family pictures.
Is this not P2P? If not, then lets extend the example:
My friends and relatives use their "web browser" program to copy my pictures and place those copies on their own "web server" programs and tell their friends and family how to access them.
Businesses develop software to do things like advertise their inventories. Other entities use software to view (and even forward to other entities) copies of this information to aid determining if a purchase is warranted. Just because at least one of the entities is a business doesn't make this NOT P2P.
Ebay is P2P in the exact same model as Napster. I have an item (presumably one that I've created or which I've paid to obtain). I advertise over a central medium that I wish someone else to have this item. For a little closer tie to the analogy, lets say this item is copyable, so if one person receives a copy, it doesn't diminish anyone else's ability to receive a copy.
Another peer comes along, decides they want a copy for themselves. They then consider the cost of acquiring a copy (Napster = click here, Ebay = win the bidding and send the money). Once the aquisition requirements are met, then owner sends the item.
Just because with the Ebay model the owner is required to take an action to complete the transaction and the transaction isn't finalized over the internet doesn't mean it isn't P2P.
When you go from those two using Apache/PHP, you don't double the LOE. You have about 15 min of compile time for Apache and PHP.
You're kidding right?
First you have to talk to the network planning people to find out if you can be placed on an existing machine or they have to buy a new one. Regardless, they'll have to spin their wheels for a while. All these people will be charging hours to the project. This is why Sprint has a very nicely designed (in your words) network.
Next, no company generating multiple billions in revenue (that wants to keep generating it) puts anything into production without proving it in a test environment, run by a testing group. This group also needs to acquire/configure an environment as similar to production as possible, then deploy your software for testing. All these people will be charging hours to the project.
15 minutes my ass. As long as upper management is in agreement (to keep all the lower level managers from delaying things while they "decide" what to do), then the conversion can be LOEd at 1000 hours and be comfortable knowing there'll be plenty to spare.
No wonder sprint's doing so bad with ppl like you in charge.
Do us all a favor, and stick to things you have a clue about.
Boy this is Deja-Vu. I work at Sprint, and if you're talking about the provisioning software that People's Choice wrote before Sprint bought them, I reviewed that product before it got passed on. I reviewed the PHP code, the hardware specs and the "Sprint" anticipated load.
My recommendation was that the only necessary change be moving from MySQL to Oracle (future growth, it's better/cheaper to be on Oracle now than have to move later). If necessary, to accomodate heavier than expected load, move the Apache/PHP to an Ultra or small E series (say E250). I estimated less than 1000 hours LOE (about 200 development tops + mgmt, testing, etc overhead) to switch from MySQL to Oracle. I could have done the thing in 20, the PHP was reasonably clean. Switching from Linux/Intel to Sparc/Solaris would have been trivial and, at most, doubled the LOE.
Imagine my surprise when two months later I had heard that the consulting firm that did the work had reported 22,000 hours and 60% completed! After that, our department didn't have any input, so I lost track of what was going on.
I don't wonder why my stock options suck right now *grumble*
The upside to all of this is that we Linux (Unix) users have alternatives to BIND (and Sendmail, and Apache, and WuFTPD and...) and don't have to wait for patches when a problem occurs.
However, many of the posts thus far have advocated everyone dumping BIND for the poster's favorite flavor of DNS. They are correct, to a point, and I agree with those posters, to a point. However, all of us dumping BIND is not the answer. Just some of us should. Others should remain with BIND.
Competing apps should continue to compete with each other for users, but should not merge, or even share code (where possible).
When worms like this show up, having diverse products will often minimize the damage caused. As opposed to our favorite proprietary vendor. When something affects M$, it affects every M$ user, thus earning the title "M$ Virus", "M$ Worm", etc. When something strikes BIND and less than half of the Unix world uses BIND, it's rather difficult to label it a "Linux" or "Unix" problem.
I had a consulting firm try to recruit me a while back. Their non-compete agreement (which was the deal breaker) claimed that their "unique consulting process" was how they differentiated themselves from their competitors. Therefore, I couldn't go to work for any competitor of theirs for fear that I would divulge the secrets of their process and unduly aid them.
There are a couple of kickers here: first, they posted the meat of their process on their web site, because their "unique process" is also the main selling point when marketing their services.
Secondly, their process was developed by experienced software consultants, who didn't get their experience at said company (because it's a fairly new company), but at previous companies. One has to wonder what NCAs were violated in the development of "the process".
It occurred to me that since the NCA didn't make much sense, I could only conclude that the company had little long-term economic incentive for me to stay, so it compensated by putting up legal barriers to keep me from leaving. Thinking about it further, it seems pretty clear that for small companies, the key threat is not an employee giving "valuable intellectual property" to a competitor, but the fact that an employee with "valuable intellectual skills" walked out the door. The damage done to a small company just by a key employee leaving can be huge.
At this consulting company, I looked around at the developer side and saw a very small core. It later became obvious that they couldn't afford to hire too many good geeks and had to use every trick in the book (besides paying them more) to keep them on. Specifically it became obvious three months after they tried to get me when they ran out of money and went under.
While I'm all for taking spammers out and having them shot, it just hit me that I've (we've) been a bit hypocritical about the topic.
Consider: I (we) hold that it should not be illegal to break into computers, it should simply be impossible (and the only way to make it impossible is to allow people to try, knowing that at least some will report their results to the community).
I (we) believe it should not be illegal to break encryption. These activities should also be encouraged to aid in evolving the technology.
In both cases, most people are taking the easy way out (legislation) while we're voicing our opposition and insisting that in the long term, any legislation would be detrimental to the industry. Besides, we're the ones breaking into computers and breaking encryption. Well maybe not you personally, but there's certainly a very loud bunch of us who are quick to voice their indignation when another geek gets in trouble for practicing those activities.
Now, however, the shoe's on the other foot. We are the recipients of the all dreaded spam! We complain loudly and bitterly. We cheer righteously when laws are passed and even louder when those laws actually manage to punish the hated spammer.
I'm just not so sure anymore. I'm now thinking that a wiser course of action would be to follow our own advice. The same sage advice we offer when we are the perpetrators and others are the victims: "Don't create new laws, create better technology".
Let me begin by stating that M$'s products are complete trash. I own exactly one machine running Windoze, exclusively for running games I can't get on Linux and opening the odd Office document that Star Office has trouble with. I also have serious issues with many of their past and current business practices. In short, the nice things I have to say about them are few and very far between.
That said, I have to agree with Decado (in principle). It is common practice for companies to acquire patents for defensive purposes, even silly ones. Then trade licenses when approached by a legal predator. Granted, acquiring these costs a fair sum of money and resources, but are fairly cheap when compared with legal costs and potential licensing fees.
In this age of patent insanity, companies that don't acquire a defensive cache are leaving themselves open to outrageous license fees for trivial things, or face horrendous legal costs on top of losing the case and paying the "full price" fees. I think we can find enough examples of that already happening.
Until the patent system gets better organized, acquiring these things is a corporate necessity. Our efforts are better spent attacking predators like Rambus and the patent system they're abusing than silly patents sitting in somebody's basement.
So, let's heartily continue bashing M$, but only for the things they actually are doing and the crap they produce.
Actually, I'm writing an application right now in Java. We're delivering content via JSP, with a distributed Java backend (using RMI). We're only going to have a couple hundred operators hitting this app so performance isn't a huge issue.
However, our developement/testing/production environments are NT+Sun/Sun/HP. Java's portability saves us more headaches than everything else combined causes.
Considering the poster probably has a bit more control over his environments and he's talking a two orders of magnitude larger user base, I wouldn't bother with Java at all.
Stick with the HTTP protocol so you don't have to develop a client and go with C or C++ for the server.
Design is a major issue when talking performance, but there's more to it than that. The poster mentioned using MySQL on the backend. That means there's quite a bit of work to do before we even start mentioning design.
Someone with a fair understanding of data analysis needs to go through and figure out what the data storage needs are. Now pick your database: MySQL is reasonably fast for small databases on small machines. But it reaches its breaking point relatively quickly. My experience indicates that PostGreSQL is the next step up the ladder. With a user base in the 5 figure range, I would run Postgres on it's own machine and watch it closely. If it seems to have problems keeping up (and you're not on too small a machine) you'll have to start looking at a big database (e.g. Oracle).
Also, the other hardware you're running on has some performance implications. Do you have a large amount of physical memory? The more information you can keep in memory, the faster your system. How are your disk file systems layed out (NFS? RAID?). The when you do have to go to the file system, these having resolved these questions will affect performance.
Now we can talk about languages and delivery mechanisms.
You mentioned keeping an eye towards portability. Unfortunately, there are trade offs there as well. If you want speed, portability is your enemy. Java and Perl are great languages (I use them and recommend them often), but they are relatively poor performers. You can pretty much eliminate any interpreted language (e.g. Tcl) and web script (e.g. PHP, ASP, ColdFusion).
The heavy lifters are still C and C++. But even if you write your CGI in C, you're still incurring the CGI penalty (which is very expensive).
If you insist on using Apache, then start by writing an Apache module in C or C++. Even faster than that is to skip Apache altogether and write the entire server yourself. You want this to be web delivered still, which is fine as the HTTP protocol isn't too difficult to implement.
Once you've figured all this out... NOW you can start your design.
As a resident of KC, I can promise KC isn't boring, and it's not even a matter of knowing where to look.
Mind you, when compared to Chicago we might look a little (ok, a lot) quiet... that and we only have one baseball team here:)
As for advertising, I didn't see jack! I work at the Sprint Campus RIGHT NEXT DOOR to where it was held and the first notice I saw that it was happening was the prior Thursday. Did I see this notice in any of the local media? Nope, Linux Weekly News (www.lwn.net). I'm surprised anyone in KC knew the event was happening.
Slashdot Mirror
I've pretty much given up on everything else. SuSE is by far the best desktop distro I've run across, I'll be upgrading to 9.1 within the week (I buy one of their distros once a year or so).
However, everyone's so called "Server" installations just manages to piss me off. I don't want a GUI. I don't want a whole bunch of crap that all these commercial distros (including RedHat and SuSE) seem to believe I need.
Slackware is the only Linux distro I put on my servers anymore. I control exactly what's on them, and that's the way I like it. Except for certain servers running OpenBSD, all my workhorse servers have Slackware.
This was originally developed while he (Gary Riley) worked for NASA at the Johnson Space Cener. It was available in source form since before I started working with it in 1993.
What's the shipping on that?
As a proud patriot and republican voter I fully and wholly disagree with you.
>> And how bad is really some collateral damage?
If you really need an answer to this, there are solutions for your problems, I recommend investigating them. Start with therapy and medication.
>> Casualties makes the enemy frightned and less willing to fight.
This is true when you're killing enemy soldiers. When you start killing their wives and children, you tend to piss them off.
>> And the enemy should expect some collateral damage when they start a war, collaterall damage will in fact make them less willing to start a war next time.
Ok, this is just nonsense. With only one super power on the planet now (that would be us), no one at a national level is going to start a war against us. We started this war in Iraq, and the objective was to liberate the people, not obliterate them (oh yeah, and look for WMD, whatever).
Like it or not,it is likely that we'll be starting a couple more wars in the future. Ostensibly against nations who get caught harboring terrorists. And again, the goal will not be to hurt the people, but remove the current government.
What your theories are on who we choose to attack and who we negotiate with when the majority of nations over there have terrorist ties, don't interest me.
Now that I've had a bit more experience, my line more frequently goes like this:
Fast, Cheap, Good, you pick ONE then I'll pick one. And, by the way, if you leave 'Good' on the table, I'll be picking that one.
I would think that tiny dicks (or at least the self esteem levels associated with the perception) would render sterilization redundant.
Well, let's see here. File sharing of copyrighted material is becoming a felony. Let's follow the trend, shall we?
... and 4 CDs. Get's charged with a felony. What are the best defenses for felony charges?
Some file sharer get's caught with 200GB of music
Addiction/Insanity!
Lawyer: Your honor, my client is addicted to music. His income is insufficient to purchase the music legally so he trades online.
Judge: Six months in rehab, two years probation. *bang*
RIAA Lawyer: *stunned bunny look*
We are a technology development company focusing on the intellectual property aspects of innovation in the information, sales and services sectors of business.
Or more clearly stated: we are an intellectual property management company focusing on beating other companies to the USPTO with braindead obvious inventions related to the internet and more specifically e-commerce.
Did anyone notice the company's email address?
mercado37@yahoo.com
Gotta love them companies leading the world into the new age *sigh*
I don't think we have that much to worry about. The more crap they pull, the more people they piss off. $20 for a CD? Thanks! May I have another? Personally, I haven't expanded my music collection (CD, MP3, whatever) since Napster shutdown. It's too hard to find anything good these days, and I'm not wasting $20 on crap (or something I don't know).
Is it any wonder my interest in new music is dwindling to nothing? And I'm sure I'm not the only one.
I get an annual allocation of options from my employer. It's a large established company, trading publically. The number of options is fine, the percentage of the company is nil. Before the market downturn (that hit EVERYONE) my options were valuing up to a rather significant bonus, since last year however, they're not worth a whole lot.
....
As a large established company, they're not going anywhere, so I can hold them pretty much as long as I work there. If I take another job, I've gotta cash them quickly, or lose them.
Obviously, there's no way to use them as income. But if the stock goes up again (before I leave), I cash out and get a nice bonus. OTOH, if I leave before they're worth something, I value them up, and build it into a signing bonus or extra salary *grin*. Either way, it's supplemental income, at best.
Bret
Integrity is the key, once you can fake that
(--BOFH)
Communication of information (songs, pictures, documents, credit card numbers, purchase orders, etc) from one entity (Person, Business, Cowboy Neal, whatever) directly to another over the internet is the INTERNET (surprise)!
For example, I take pictures of my family and make them available on the internet (lets call the software I use to do this a "web server"). Next, I provide my friends and relatives with a client (we'll call this software a "web browser") and instruct them on how to use it to see my family pictures.
Is this not P2P? If not, then lets extend the example:
My friends and relatives use their "web browser" program to copy my pictures and place those copies on their own "web server" programs and tell their friends and family how to access them.
Businesses develop software to do things like advertise their inventories. Other entities use software to view (and even forward to other entities) copies of this information to aid determining if a purchase is warranted. Just because at least one of the entities is a business doesn't make this NOT P2P.
Ebay is P2P in the exact same model as Napster. I have an item (presumably one that I've created or which I've paid to obtain). I advertise over a central medium that I wish someone else to have this item. For a little closer tie to the analogy, lets say this item is copyable, so if one person receives a copy, it doesn't diminish anyone else's ability to receive a copy.
Another peer comes along, decides they want a copy for themselves. They then consider the cost of acquiring a copy (Napster = click here, Ebay = win the bidding and send the money). Once the aquisition requirements are met, then owner sends the item.
Just because with the Ebay model the owner is required to take an action to complete the transaction and the transaction isn't finalized over the internet doesn't mean it isn't P2P.
Fayd (running with scissors)
You're kidding right?
First you have to talk to the network planning people to find out if you can be placed on an existing machine or they have to buy a new one. Regardless, they'll have to spin their wheels for a while. All these people will be charging hours to the project. This is why Sprint has a very nicely designed (in your words) network.
Next, no company generating multiple billions in revenue (that wants to keep generating it) puts anything into production without proving it in a test environment, run by a testing group. This group also needs to acquire/configure an environment as similar to production as possible, then deploy your software for testing. All these people will be charging hours to the project.
15 minutes my ass. As long as upper management is in agreement (to keep all the lower level managers from delaying things while they "decide" what to do), then the conversion can be LOEd at 1000 hours and be comfortable knowing there'll be plenty to spare.
No wonder sprint's doing so bad with ppl like you in charge.
Do us all a favor, and stick to things you have a clue about.
Boy this is Deja-Vu. I work at Sprint, and if you're talking about the provisioning software that People's Choice wrote before Sprint bought them, I reviewed that product before it got passed on. I reviewed the PHP code, the hardware specs and the "Sprint" anticipated load.
My recommendation was that the only necessary change be moving from MySQL to Oracle (future growth, it's better/cheaper to be on Oracle now than have to move later). If necessary, to accomodate heavier than expected load, move the Apache/PHP to an Ultra or small E series (say E250). I estimated less than 1000 hours LOE (about 200 development tops + mgmt, testing, etc overhead) to switch from MySQL to Oracle. I could have done the thing in 20, the PHP was reasonably clean. Switching from Linux/Intel to Sparc/Solaris would have been trivial and, at most, doubled the LOE.
Imagine my surprise when two months later I had heard that the consulting firm that did the work had reported 22,000 hours and 60% completed! After that, our department didn't have any input, so I lost track of what was going on.
I don't wonder why my stock options suck right now *grumble*
The upside to all of this is that we Linux (Unix) users have alternatives to BIND (and Sendmail, and Apache, and WuFTPD and ...) and don't have to wait for patches when a problem occurs.
However, many of the posts thus far have advocated everyone dumping BIND for the poster's favorite flavor of DNS. They are correct, to a point, and I agree with those posters, to a point. However, all of us dumping BIND is not the answer. Just some of us should. Others should remain with BIND.
Competing apps should continue to compete with each other for users, but should not merge, or even share code (where possible).
When worms like this show up, having diverse products will often minimize the damage caused. As opposed to our favorite proprietary vendor. When something affects M$, it affects every M$ user, thus earning the title "M$ Virus", "M$ Worm", etc. When something strikes BIND and less than half of the Unix world uses BIND, it's rather difficult to label it a "Linux" or "Unix" problem.
I had a consulting firm try to recruit me a while back. Their non-compete agreement (which was the deal breaker) claimed that their "unique consulting process" was how they differentiated themselves from their competitors. Therefore, I couldn't go to work for any competitor of theirs for fear that I would divulge the secrets of their process and unduly aid them.
There are a couple of kickers here: first, they posted the meat of their process on their web site, because their "unique process" is also the main selling point when marketing their services.
Secondly, their process was developed by experienced software consultants, who didn't get their experience at said company (because it's a fairly new company), but at previous companies. One has to wonder what NCAs were violated in the development of "the process".
It occurred to me that since the NCA didn't make much sense, I could only conclude that the company had little long-term economic incentive for me to stay, so it compensated by putting up legal barriers to keep me from leaving. Thinking about it further, it seems pretty clear that for small companies, the key threat is not an employee giving "valuable intellectual property" to a competitor, but the fact that an employee with "valuable intellectual skills" walked out the door. The damage done to a small company just by a key employee leaving can be huge.
At this consulting company, I looked around at the developer side and saw a very small core. It later became obvious that they couldn't afford to hire too many good geeks and had to use every trick in the book (besides paying them more) to keep them on. Specifically it became obvious three months after they tried to get me when they ran out of money and went under.
While I'm all for taking spammers out and having them shot, it just hit me that I've (we've) been a bit hypocritical about the topic.
Consider: I (we) hold that it should not be illegal to break into computers, it should simply be impossible (and the only way to make it impossible is to allow people to try, knowing that at least some will report their results to the community).
I (we) believe it should not be illegal to break encryption. These activities should also be encouraged to aid in evolving the technology.
In both cases, most people are taking the easy way out (legislation) while we're voicing our opposition and insisting that in the long term, any legislation would be detrimental to the industry. Besides, we're the ones breaking into computers and breaking encryption. Well maybe not you personally, but there's certainly a very loud bunch of us who are quick to voice their indignation when another geek gets in trouble for practicing those activities.
Now, however, the shoe's on the other foot. We are the recipients of the all dreaded spam! We complain loudly and bitterly. We cheer righteously when laws are passed and even louder when those laws actually manage to punish the hated spammer.
I'm just not so sure anymore. I'm now thinking that a wiser course of action would be to follow our own advice. The same sage advice we offer when we are the perpetrators and others are the victims: "Don't create new laws, create better technology".
Let me begin by stating that M$'s products are complete trash. I own exactly one machine running Windoze, exclusively for running games I can't get on Linux and opening the odd Office document that Star Office has trouble with. I also have serious issues with many of their past and current business practices. In short, the nice things I have to say about them are few and very far between.
That said, I have to agree with Decado (in principle). It is common practice for companies to acquire patents for defensive purposes, even silly ones. Then trade licenses when approached by a legal predator. Granted, acquiring these costs a fair sum of money and resources, but are fairly cheap when compared with legal costs and potential licensing fees.
In this age of patent insanity, companies that don't acquire a defensive cache are leaving themselves open to outrageous license fees for trivial things, or face horrendous legal costs on top of losing the case and paying the "full price" fees. I think we can find enough examples of that already happening.
Until the patent system gets better organized, acquiring these things is a corporate necessity. Our efforts are better spent attacking predators like Rambus and the patent system they're abusing than silly patents sitting in somebody's basement.
So, let's heartily continue bashing M$, but only for the things they actually are doing and the crap they produce.
Actually, I'm writing an application right now in Java. We're delivering content via JSP, with a distributed Java backend (using RMI). We're only going to have a couple hundred operators hitting this app so performance isn't a huge issue.
However, our developement/testing/production environments are NT+Sun/Sun/HP. Java's portability saves us more headaches than everything else combined causes.
Considering the poster probably has a bit more control over his environments and he's talking a two orders of magnitude larger user base, I wouldn't bother with Java at all.
Stick with the HTTP protocol so you don't have to develop a client and go with C or C++ for the server.
Design is a major issue when talking performance, but there's more to it than that. The poster mentioned using MySQL on the backend. That means there's quite a bit of work to do before we even start mentioning design.
... NOW you can start your design.
Someone with a fair understanding of data analysis needs to go through and figure out what the data storage needs are. Now pick your database: MySQL is reasonably fast for small databases on small machines. But it reaches its breaking point relatively quickly. My experience indicates that PostGreSQL is the next step up the ladder. With a user base in the 5 figure range, I would run Postgres on it's own machine and watch it closely. If it seems to have problems keeping up (and you're not on too small a machine) you'll have to start looking at a big database (e.g. Oracle).
Also, the other hardware you're running on has some performance implications. Do you have a large amount of physical memory? The more information you can keep in memory, the faster your system. How are your disk file systems layed out (NFS? RAID?). The when you do have to go to the file system, these having resolved these questions will affect performance.
Now we can talk about languages and delivery mechanisms.
You mentioned keeping an eye towards portability. Unfortunately, there are trade offs there as well. If you want speed, portability is your enemy. Java and Perl are great languages (I use them and recommend them often), but they are relatively poor performers. You can pretty much eliminate any interpreted language (e.g. Tcl) and web script (e.g. PHP, ASP, ColdFusion).
The heavy lifters are still C and C++. But even if you write your CGI in C, you're still incurring the CGI penalty (which is very expensive).
If you insist on using Apache, then start by writing an Apache module in C or C++. Even faster than that is to skip Apache altogether and write the entire server yourself. You want this to be web delivered still, which is fine as the HTTP protocol isn't too difficult to implement.
Once you've figured all this out
As a resident of KC, I can promise KC isn't boring, and it's not even a matter of knowing where to look.
... that and we only have one baseball team here :)
.sig is a good .sig)
Mind you, when compared to Chicago we might look a little (ok, a lot) quiet
As for advertising, I didn't see jack! I work at the Sprint Campus RIGHT NEXT DOOR to where it was held and the first notice I saw that it was happening was the prior Thursday. Did I see this notice in any of the local media? Nope, Linux Weekly News (www.lwn.net). I'm surprised anyone in KC knew the event was happening.
Bret
(No