Availability: Keep the hardware running, PostgreSQL will keep running. Our PostgreSQL server simply DOESN'T have problems. The only two times it's been shut down in YEARS were for planned hardware upgrades.
You're half right, half wrong. Moore's law DOES deal with transistor count. However, it says that it will double every 18 months, not every 6 months. (originally, it was 24 months, but later revised.)
In practice, however, the actual computational power has been doubling about every 18 months as well.
As evidence, look at where we were 10 year ago: The big, bad processer to have was a 33 MHz 486. Today's high-end processers have MORE than 64 times the computational power of the 486 of a decade ago - and there's no indication that we're not going to keep on track for another decade.
Right, but we keep making individual transistors smaller and smaller, letting them use less and less power. Of course, CPU manufacturers tend to simply add more transistors and/or increase the frequency to make up for the power savings.
Look at the newest, fastest Athlons - they produce less heat than considerably older versions. Why? Smaller manufacturing process. And that's going to keep on going...
Supposedly, this offers encryption with less computational demand. And, supposedly, it's not going to be in use for 5 to 10 years.
If that's the case, my quesion is this: Why bother? Moore's law says that in the 10 years that it will take to get this implemented, CPU's will be *64 times faster* than they are today.
Just think: "Wow! With this new encryption technology, encrypted 100 megabit networking only takes 0.05% of my processer instead of 0.1%!"
What if this could also be applied to DDR II (QDR) memory? THAT would give some REALLY impressive bandwidth.
The crappy side is that even if it can be applied, it's virtually guaranteed that the memory industry will take the wimp's way - first introduce DDR II, then wait a few years to introduce the dual-band DDR II. No sense in skipping a generation, that would just mean less revenue, right?
Any engineer will tell you that he can give you any tolerances you want, it's a matter of how much you want to spend. In this case, the issues are:
1. Bearings. You're going to spin that platter at 15,000 RPM 24 hrs. per day for years on end? At nearly 10 BILLION rotations per year, if you want reliability, those bearings had better be preeeeety good. And that means - more money.
2. Platter surface. Same as above. You want to spin that thing around thirty or fourty billion times with the heads nothing more than maybe a thousandth of an inch away? Better be awfully tough stuff, and it better be permanently bonded to the platter. Oh, wait - you're going to bump and kick your computer while it's on, aren't you? How good for the platter surface (or the heads) will those collisions be? Better coating, better heads: More money.
3. Electronics. Drives get HOT. You want your electronics to last a long time? They better be made for high-temp operation. That means.... yep, more money.
In the end, each company has had to ask itself this:
"Will we spend more money on quality drives, and hope that customer recognition pays off, or will we skimp a little here and there, and sell them by the boat-load?"
You can guess which one they've chosen. Why did they choose it? You guessed it... more money.
Solder wires to some straight pins, and push them through the headphone cable. That way, when you have to give the set back, they probably won't notice.
"State of the Art" is a good way to run your pocket book into the ground. Jumping on the newest, fanciest programming language doesn't usually make a business successful.
Here's yet another example: My company's (former) largest competitor invested *millions* into Sun hardware and development in Java. Why? "State of the Art". And guess what! With all of their "state of the art" infrastructure, their system was still slow as molasses.
What did we do? We spent less than a tenth of what they did to develop with Perl on x86 servers. Our site handles huge traffic loads pretty effectively, and we did it without running ourselves to the bankruptcy court.
But with my Sony, I can get around most of the FBI warnings, "mandatory previews", and other annoying features. Even though you may not be able to hit "next" all of the time, between trying the "title", "next", "menu", and as a last resort, the fast-forward buttons, most of them can be passed up.
Now if they just weren't there at all, now THAT would be nice.
However, as others have pointed out, up until recently Zimbabwe was a major agricultural exporter. I find it hard to believe that they lack sufficient milling capacity to grind the corn themselves -- unless things really have deteriorated that badly since Mugabe took power
There are several conditions:
Droughts causing low crop yields
Commercial farms which were siezed by war veterans have produced crops at a greatly reduced rate.
Mugabe is impeding the remaining commercial farms, stopping plantation of corn on over 50% of the country's farmland.
Combine the three, and you're in for some serious starvation.
steve
Re:This is probably a stupid question
on
AMD's 64-Bit Chip
·
· Score: 2
Nice and simple. Two chips, two busses, and interactions between the two different systems add up to a bloated, buggy, and (most importantly) expensive system.
> because Robert Jordan began by blatantly > ripping off JRRT
Jeez, if you want to see someone REALLY rip off JRRT, look at Dennis L. McKiernan. Let's see, halflings go off to fight a war against the evil lord who has returned with his dark hoardes. They go through the ancient dwarvish mines where the dwarves mined too deeply, and uncovered an ancient dread. Foul water at the door, with a many-armed beast, which blocks the door after they enter. It doesn't get much more plagiaristic than that.
"Adjust your soldier's personality and watch how it changes his decisions"
There are some very healthy lessons to be learned from that:
- That your decisions don't have to be knee-jerk reactions. - That your decisions have long-reaching effects on your life. - That if you want a good life, making the right decisions early is the best way to make it happen.
It seems that there are fewer and fewer chances for children to learn those lessons as time goes on. The thought of a million kids, each taking responsibility for making the most of their own life.... it seems too good to be true.
Google has a great deal of servers behind load-balancers. If one of their machiens konks over dead, no big deal - you've only lost 1.500th of your capacity, and nobody knows the difference. If you're talking about a single file server, though, konk out a drive, and you're in deep trouble.
As if the tens of thousands of times this has been hashed out weren't enough already...
The question of IDE vs. SCSI is not (or should not) be about speed. Really. There are nice, fast drives in each camp. If speed is all that matters to you, go with IDE, it'll be a lot cheaper.
So are there any advantages to SCSI? Sure. But not for the majority of people. SCSI's beauties are:
- You can hook a LOT of drives to one controller - You can hook most any kind of device to the controller - You can hook devices up both inside and outside of the case - You can use much longer cables - When the controller is waiting on one command, it can issue other commands while it's waiting
SCSI was designed for systems where you would either have many, many devices connected to the controller, or where many different processes (or users) would be accessing the hardware simultaneously - and in either of those situations, it *does* perform better than IDE. However, the portion of systems that will actually enter into that area are very, very few. In general, "if you have to ask, you don't need it."
As for straight speed, if you're looking for all-out throughput, don't rely on a single drive, get a RAID array - be it IDE or SCSI. By getting a faster drive, you can increase your throughput by what - 10%? 20%? A two-drive array will nearly double your throughput, and with quality controllers, it's fairly linear up through three to five drives - again, depending on the quality of the controller.
Well, the $20 Pentium firewall isn't quite the same - while it can seperate one part of the LAN from another (or different networks), the advantage of the card is that it protects your machine from *everything* else, at least theoretically.
Having a principle firewall on the border of your network isn't challenged, but in a setting with many computers which can't be closely individually monitored (libraries, college campuses, etc.), these will at least help to prevent one person from attacking/abusing other machines on the same network.
First, you can't just stick a gigabit card in a machine and expect it to work at full capacity. The basic design of ethernet was not really designed for gigabit speeds, but we've managed to squeeze it out - barely.
With 10 mbit cards, having the card generate an interrupt with ever incoming frame wasn't too bad. And on 100-mbit, it's still managable - but at a full gigabit, it really, really starts to bog down the machine. Some cards get around that by using interrupt coalescing, where they buffer up a few frames before they trigger an interrupt. That has a drawback, though: It increases latency. The trade-off has to be at some point, and not choosing the RIGHT point can affect either throughput or latency.
Furthermore, to get the full benefit out of your card, you generally need to enable jumbo-frames on both the card and the switch - and of course, your switch has to support that feature.
To make matters even worse, you can't always pump out (or receive) a full gigabit in any other than testing situations. Say you're receiving a large incoming file via FTP, NFS, or the protocol of your choice. Can your machine *really* write data to the disk at over 100 megabytes per second? And if it can, can it really handle both receiving a gigabit from the card, processing it, and writing the gigabit out to the disk? Unless you've got a very large amount of money in the machine, it probably won't.
steve
This is *nothing*
on
Google Juice
·
· Score: 0, Redundant
What that guy did is literally *nothing* compared to what the company I work for has been doing for the last 3 years....
But not entirely. Sure, they can't just pull advances out of thier butts - but at the same time, how long have nickel-metal-hydride and lithium-ion batteries been around for? A looong time. But can you buy some? A very few stores have NiMH rechargeables, but unless you go searching at specialty shops, you're not going to find any regular-sized lithium batteries. If the battery industry was really concerned with getting more power to us, they'd shift production to lithium cells, and then the cost would drop...
When our company started, we started in an affordable location, and stayed there. We built on an inexpensive, yet powerful Linux-based infrastructure. We pinched every penny. We never recieved venture capital, but we've been around 3 years, and we're turning profits.
Our competitors started out in expensive cities. They spent millions on their infrastructure, to impress the investors - and the investors flocked. They burned through several rounds of multi-million-dollar funding, but never really produced anything. Now, they're all out of business or have all 'downsized' and moved to cheaper locations.
How in the world to people with so little business-sense (or even just plain COMMON financial sense) get in charge of these companies? It never ceases to amaze me that so many people have been (and still are) so STUPID.
We're only talking about $5, huh? Yeah. $5 to this site. $5 to another. $10 to another, $3 to another, ad infinitum.
I'm not going to pay every site that I go to. I don't want the total cost, the hastle of paying a hundred people, and I don't like the idea.
So you realize that I'm not being a hypocrite, Besides the web site that I work for as a day job, I run a couple of sites on the side that make money. Even the ones with quite a bit of good, quality content don't charge the end-users any amount of money, and certainly not for subscriptions. That's just not how I want the net to work.
Slashdot Mirror
Availability: Keep the hardware running, PostgreSQL will keep running. Our PostgreSQL server simply DOESN'T have problems. The only two times it's been shut down in YEARS were for planned hardware upgrades.
steve
You're half right, half wrong. Moore's law DOES deal with transistor count. However, it says that it will double every 18 months, not every 6 months. (originally, it was 24 months, but later revised.)
In practice, however, the actual computational power has been doubling about every 18 months as well.
As evidence, look at where we were 10 year ago: The big, bad processer to have was a 33 MHz 486. Today's high-end processers have MORE than 64 times the computational power of the 486 of a decade ago - and there's no indication that we're not going to keep on track for another decade.
steve
Right, but we keep making individual transistors smaller and smaller, letting them use less and less power. Of course, CPU manufacturers tend to simply add more transistors and/or increase the frequency to make up for the power savings.
Look at the newest, fastest Athlons - they produce less heat than considerably older versions. Why? Smaller manufacturing process. And that's going to keep on going...
steve
Supposedly, this offers encryption with less computational demand. And, supposedly, it's not going to be in use for 5 to 10 years.
If that's the case, my quesion is this: Why bother? Moore's law says that in the 10 years that it will take to get this implemented, CPU's will be *64 times faster* than they are today.
Just think: "Wow! With this new encryption technology, encrypted 100 megabit networking only takes 0.05% of my processer instead of 0.1%!"
steve
What if this could also be applied to DDR II (QDR) memory? THAT would give some REALLY impressive bandwidth.
The crappy side is that even if it can be applied, it's virtually guaranteed that the memory industry will take the wimp's way - first introduce DDR II, then wait a few years to introduce the dual-band DDR II. No sense in skipping a generation, that would just mean less revenue, right?
steve
It's the old adage:
"good, fast, cheap - pick any two."
Any engineer will tell you that he can give you any tolerances you want, it's a matter of how much you want to spend. In this case, the issues are:
1. Bearings. You're going to spin that platter at 15,000 RPM 24 hrs. per day for years on end? At nearly 10 BILLION rotations per year, if you want reliability, those bearings had better be preeeeety good. And that means - more money.
2. Platter surface. Same as above. You want to spin that thing around thirty or fourty billion times with the heads nothing more than maybe a thousandth of an inch away? Better be awfully tough stuff, and it better be permanently bonded to the platter. Oh, wait - you're going to bump and kick your computer while it's on, aren't you? How good for the platter surface (or the heads) will those collisions be? Better coating, better heads: More money.
3. Electronics. Drives get HOT. You want your electronics to last a long time? They better be made for high-temp operation. That means.... yep, more money.
In the end, each company has had to ask itself this:
"Will we spend more money on quality drives, and hope that customer recognition pays off, or will we skimp a little here and there, and sell them by the boat-load?"
You can guess which one they've chosen. Why did they choose it? You guessed it... more money.
Solder wires to some straight pins, and push them through the headphone cable. That way, when you have to give the set back, they probably won't notice.
steve
Then why don't they use the low-power CPU's designed just for embedded circuits?
"State of the Art" is a good way to run your pocket book into the ground. Jumping on the newest, fanciest programming language doesn't usually make a business successful.
Here's yet another example: My company's (former) largest competitor invested *millions* into Sun hardware and development in Java. Why? "State of the Art". And guess what! With all of their "state of the art" infrastructure, their system was still slow as molasses.
What did we do? We spent less than a tenth of what they did to develop with Perl on x86 servers. Our site handles huge traffic loads pretty effectively, and we did it without running ourselves to the bankruptcy court.
steve
steve
But with my Sony, I can get around most of the FBI warnings, "mandatory previews", and other annoying features. Even though you may not be able to hit "next" all of the time, between trying the "title", "next", "menu", and as a last resort, the fast-forward buttons, most of them can be passed up.
Now if they just weren't there at all, now THAT would be nice.
steve
Here's a schematic for a 'touch switch', the type used on the 'touch lamps':
http://www.paia.com/~paia/touchsw.htm
Use it to connect a 9-volt battery and a piezo buzzer.
steve
There are several conditions:
Combine the three, and you're in for some serious starvation.
steve
Nice and simple. Two chips, two busses, and interactions between the two different systems add up to a bloated, buggy, and (most importantly) expensive system.
steve
> pity about the casting of Elrond tho
I couldn't help but expect "Welcome to Rivendell... Mister Baggins."
steve
> because Robert Jordan began by blatantly
> ripping off JRRT
Jeez, if you want to see someone REALLY rip off JRRT, look at Dennis L. McKiernan. Let's see, halflings go off to fight a war against the evil lord who has returned with his dark hoardes. They go through the ancient dwarvish mines where the dwarves mined too deeply, and uncovered an ancient dread. Foul water at the door, with a many-armed beast, which blocks the door after they enter. It doesn't get much more plagiaristic than that.
steve
I read on the blurb for Soldiers:
"Adjust your soldier's personality and watch how it changes his decisions"
There are some very healthy lessons to be learned from that:
- That your decisions don't have to be knee-jerk reactions.
- That your decisions have long-reaching effects on your life.
- That if you want a good life, making the right decisions early is the best way to make it happen.
It seems that there are fewer and fewer chances for children to learn those lessons as time goes on. The thought of a million kids, each taking responsibility for making the most of their own life.... it seems too good to be true.
steve
Google has a great deal of servers behind load-balancers. If one of their machiens konks over dead, no big deal - you've only lost 1.500th of your capacity, and nobody knows the difference. If you're talking about a single file server, though, konk out a drive, and you're in deep trouble.
steve
As if the tens of thousands of times this has been hashed out weren't enough already...
The question of IDE vs. SCSI is not (or should not) be about speed. Really. There are nice, fast drives in each camp. If speed is all that matters to you, go with IDE, it'll be a lot cheaper.
So are there any advantages to SCSI? Sure. But not for the majority of people. SCSI's beauties are:
- You can hook a LOT of drives to one controller
- You can hook most any kind of device to the controller
- You can hook devices up both inside and outside of the case
- You can use much longer cables
- When the controller is waiting on one command, it can issue other commands while it's waiting
SCSI was designed for systems where you would either have many, many devices connected to the controller, or where many different processes (or users) would be accessing the hardware simultaneously - and in either of those situations, it *does* perform better than IDE. However, the portion of systems that will actually enter into that area are very, very few. In general, "if you have to ask, you don't need it."
As for straight speed, if you're looking for all-out throughput, don't rely on a single drive, get a RAID array - be it IDE or SCSI. By getting a faster drive, you can increase your throughput by what - 10%? 20%? A two-drive array will nearly double your throughput, and with quality controllers, it's fairly linear up through three to five drives - again, depending on the quality of the controller.
steve
Well, the $20 Pentium firewall isn't quite the same - while it can seperate one part of the LAN from another (or different networks), the advantage of the card is that it protects your machine from *everything* else, at least theoretically.
Having a principle firewall on the border of your network isn't challenged, but in a setting with many computers which can't be closely individually monitored (libraries, college campuses, etc.), these will at least help to prevent one person from attacking/abusing other machines on the same network.
steve
First, you can't just stick a gigabit card in a machine and expect it to work at full capacity. The basic design of ethernet was not really designed for gigabit speeds, but we've managed to squeeze it out - barely.
With 10 mbit cards, having the card generate an interrupt with ever incoming frame wasn't too bad. And on 100-mbit, it's still managable - but at a full gigabit, it really, really starts to bog down the machine. Some cards get around that by using interrupt coalescing, where they buffer up a few frames before they trigger an interrupt. That has a drawback, though: It increases latency. The trade-off has to be at some point, and not choosing the RIGHT point can affect either throughput or latency.
Furthermore, to get the full benefit out of your card, you generally need to enable jumbo-frames on both the card and the switch - and of course, your switch has to support that feature.
To make matters even worse, you can't always pump out (or receive) a full gigabit in any other than testing situations. Say you're receiving a large incoming file via FTP, NFS, or the protocol of your choice. Can your machine *really* write data to the disk at over 100 megabytes per second? And if it can, can it really handle both receiving a gigabit from the card, processing it, and writing the gigabit out to the disk? Unless you've got a very large amount of money in the machine, it probably won't.
steve
What that guy did is literally *nothing* compared to what the company I work for has been doing for the last 3 years....
steve
But not entirely. Sure, they can't just pull advances out of thier butts - but at the same time, how long have nickel-metal-hydride and lithium-ion batteries been around for? A looong time. But can you buy some? A very few stores have NiMH rechargeables, but unless you go searching at specialty shops, you're not going to find any regular-sized lithium batteries. If the battery industry was really concerned with getting more power to us, they'd shift production to lithium cells, and then the cost would drop...
steve
When our company started, we started in an affordable location, and stayed there. We built on an inexpensive, yet powerful Linux-based infrastructure. We pinched every penny. We never recieved venture capital, but we've been around 3 years, and we're turning profits.
Our competitors started out in expensive cities. They spent millions on their infrastructure, to impress the investors - and the investors flocked. They burned through several rounds of multi-million-dollar funding, but never really produced anything. Now, they're all out of business or have all 'downsized' and moved to cheaper locations.
How in the world to people with so little business-sense (or even just plain COMMON financial sense) get in charge of these companies? It never ceases to amaze me that so many people have been (and still are) so STUPID.
steve
We're only talking about $5, huh? Yeah. $5 to this site. $5 to another. $10 to another, $3 to another, ad infinitum.
I'm not going to pay every site that I go to. I don't want the total cost, the hastle of paying a hundred people, and I don't like the idea.
So you realize that I'm not being a hypocrite, Besides the web site that I work for as a day job, I run a couple of sites on the side that make money. Even the ones with quite a bit of good, quality content don't charge the end-users any amount of money, and certainly not for subscriptions. That's just not how I want the net to work.
steve