The chicken scenario described removed any curiosity I had about looking into the library further. Why? Because it's very similar to the Java 101 bouncing ball thread demo (one thread per ball) which is used to show why 1 thread per ball doesn't scale to first time would be multi-threaded programmers.
This is definitely a good suggestion. I'd only keep data needing solid backups on the SAN. Systems should have their own boot disks and pagefile locations locally (lowers SAN utilization).
Then again, it truly sounds like he probably needs to review his SAN architecture. I'd probably have the DB on its own set of spindles, and having 2 domain controllers, with the primary being standalone and the secondary (and potentially tertiary if needed) doubling as print servers. Other than that, we'd need a lot more information to truly address how he should go about altering hsi configuration to handle the load(s).
In Java, this is allowed because of performance issues. You can make it almost 100% thread safe (note I said almost) by synchronizing every method, but there's still some gotchas in the JDK.
Multi-threaded programming is a skill that comes with a level of understanding, much like students of mathematics must reach a level of understanding to comprehend Algebra, Calculus, Differential Equations, and Partial Differential Equations (yeah, that last one's a bear, especially when you apply it to various physical models) respectively. It's why you can be a wiz at adding or subtracting, but utterly fail at algebra, or a wiz at algebra, but never be able to differentiate or integrate a function, and so on.
Writing code is easy. Any moron can do it, as the Java and PHP hordes have shown. Writing good code is harder, designing good OO code is even harder, and designing and writing good multi-threaded code is yet a step beyond that. It's why there are so few well-written multi-threaded apps, and most of those are in server land.
Even so, this is a "bad" implementation. There's absolutely no reason for there to be 1 thread per chicken. That's inherently not scalable. What you really want are an optimum number of threads for the number of cores in a pool that handle work units (chickens). This will scale much higher than the 1 thread per object model discussed in this topic.
Oh, and there's no such thing as "easy" multi-threading. Hell, the average programmer can't even grasp OO, so what makes them think they can grasp threading which has many many more aspects to it?
On X-Rays, they figured that one out pretty darn quickly. After all, there were view your body, hand and feet devices available to the public. They were already in use 1 year after their discovery.
We'll never know. Ohio's election officials apparently tailored the "random" samples for verification. No one AFAIK has accurately reported on the number of disenfranchised voters that were hounded or otherwise discouraged from voting by Republican croonies.
The funny thing? Until the shrub ran for office, I was a staunch Republican that would have said you're insane for even implying that I'd ever support a Democrat. Democrats be da deevil, ya no. (weak Waterboy reference)
It's not NASA that's the problem, but the current "brain" sitting behind the oval office. You know, the same one that got a minority of the popular vote in the last 2 elections?
Well, considering that UNIX98 also included Y2K compliance....
But, I cannot answer your questions. I can offer an anecdote that my personal latest sighting of one of those specific UNIX versions was in 2002. I'd assume (I know) that later versions came on later hardware, and thus are 64 bit.
As for BSD/Linux, the majority of those installations are probably still 32 bit, but I'd expect that to change within the next 10 years, well before 2038. (OMG - a 10 year hardware cycle for PC hardware?!?! If you're not running Windows, sure!;)
But if it's Solaris 7 or above, and especially if you're running on a Ultrasparc IIe or III or above, you're already running 64 bit UNIX. See this list
Isn't the purpose of technology to make things easier, not harder? This would be another major negative to Vista - it makes things harder, and doesn't help me in any way that I've heard.
I don't need transparent windows for the most part either. A second monitor is much more helpful. The exception is the way that Apple's Mac OSX does property windows. Those being somewhat transparent is helpful at times.
As for VMs, why would I bother? XP is stable (despite the common perception here) - why pay the penalty for a VM if I'm going to spend 70-80% of my time in Windows anyway? First XP is not stable. It is apparently stable enough for your usage, but it is not stable. A subtle but important difference. (Try sleeping XP sometime....) I agree that running a VM for 80% of your time seems senseless. There could be cases where it's a good idea though, or where you don't even notice it.
Interesting. I was wondering what happened. My WPA connection with my MBP fails to restart when coming out of sleep, sometimes for minutes. Once, after connecting to an open wireless network, it actually required a reboot! (The first in 3 months)
I've noticed that if I enter the same room as my router and bring the MBP out of sleep mode, it tends to connect within seconds, and that connection then holds even when leaving the house. I've also noticed a lot more potential networks to connect to of the format nWIREnnn which didn't show up before. These are fleeting, and I haven't checked whether they are the same consistent set of networks, or whether it's some cruft that's confusing the driver.
My router is a Netgear 802.11g router, btw.
As for connecting up to open networks, cycling your airport will sometimes let you connect without rebooting. I should also note that I haven't updated in about a week or two, so I haven't seen the latest update and don't know how that would affect me.
On the lighter side, the 85 percent rule states that 85 percent of the human species is genetically stupid, and nothing any of us do will give these people a clue. There is an additional 10 percent that have been corrupted by exposure to the 85 percent, so they too become stupid. This leaves us with only 5 percent of the species even worth knowing. I think you're numbers are a little optimistic. Less than 1% is worth knowing, and most of those are being corrupted by Management or Business knowledge, leaving just the slathering late-night drooling cream of the crop you see here on./
You're assuming that their statement "reformatted the backup hard drive" meant a single physical hard drive. It could just as easily been a RAID array, or a SAN. I would certainly not bet on what they meant by "hard drive" in that story, nor that they even got the actual technical facts right.
The story states he deleted the mains, then reformatted the backup, and the tapes were corrupted. Sounds like a top-notch tech there if it were merely single hard drives. Getting a backup RAID array or SAN online is a little different though, and might lend itself to massive corruption if done incorrectly. Hint, reformatting a single disk is trivial to undo, so much so that even a reformat done by an ignorant twit should result in a recoverable drive with most of the information available. A RAID array or SAN however can be painful, although I doubt this was on a SAN.
I'm willing to bet you're underestimating that data by at least an order of magnitude. Remember, these are government documents they're scanning. There's no such thing as a 1 page document to get money from the government.:)
But I agree with your statement that DVDs are pointless. a couple of 100+GB tapes are a much better way to go, and much easier to deal with considering that this is mostly static data.
That is cool - but think how fast that gets handed to the server that's serving you. Think about how much faster it'd be in RAID 10 (or was that 0 + 1 for Mirror + Striping)
The problem with ethernet connected drives is that the lag is higher than that of internal drives, which is the reason that internal drives still exist.
There are definitely things to like about this, though, as a $0.50 ethernet chip these days should be cheap enough to add transparently to almost any drive. I like the fact that ethernet cables are almost as small as SATA/SAS cables.
I'd agree, "prior art" is almost irrelevant on this one. Considering they've been in use for about 50 years I'd say this is a prime example of how the PTO is completely, hopelessly broken. What will they patent next? The worm screw pump?
But, I read the patent, and it is even worse than that. It's basically describing the basics of pointer storage as applied to linked lists.
Slashdot Mirror
The chicken scenario described removed any curiosity I had about looking into the library further. Why? Because it's very similar to the Java 101 bouncing ball thread demo (one thread per ball) which is used to show why 1 thread per ball doesn't scale to first time would be multi-threaded programmers.
This is definitely a good suggestion. I'd only keep data needing solid backups on the SAN. Systems should have their own boot disks and pagefile locations locally (lowers SAN utilization).
Then again, it truly sounds like he probably needs to review his SAN architecture. I'd probably have the DB on its own set of spindles, and having 2 domain controllers, with the primary being standalone and the secondary (and potentially tertiary if needed) doubling as print servers. Other than that, we'd need a lot more information to truly address how he should go about altering hsi configuration to handle the load(s).
In Java, this is allowed because of performance issues. You can make it almost 100% thread safe (note I said almost) by synchronizing every method, but there's still some gotchas in the JDK.
Multi-threaded programming is a skill that comes with a level of understanding, much like students of mathematics must reach a level of understanding to comprehend Algebra, Calculus, Differential Equations, and Partial Differential Equations (yeah, that last one's a bear, especially when you apply it to various physical models) respectively. It's why you can be a wiz at adding or subtracting, but utterly fail at algebra, or a wiz at algebra, but never be able to differentiate or integrate a function, and so on.
Writing code is easy. Any moron can do it, as the Java and PHP hordes have shown. Writing good code is harder, designing good OO code is even harder, and designing and writing good multi-threaded code is yet a step beyond that. It's why there are so few well-written multi-threaded apps, and most of those are in server land.
Even so, this is a "bad" implementation. There's absolutely no reason for there to be 1 thread per chicken. That's inherently not scalable. What you really want are an optimum number of threads for the number of cores in a pool that handle work units (chickens). This will scale much higher than the 1 thread per object model discussed in this topic.
Oh, and there's no such thing as "easy" multi-threading. Hell, the average programmer can't even grasp OO, so what makes them think they can grasp threading which has many many more aspects to it?
On X-Rays, they figured that one out pretty darn quickly. After all, there were view your body, hand and feet devices available to the public. They were already in use 1 year after their discovery.
We'll never know. Ohio's election officials apparently tailored the "random" samples for verification. No one AFAIK has accurately reported on the number of disenfranchised voters that were hounded or otherwise discouraged from voting by Republican croonies.
The funny thing? Until the shrub ran for office, I was a staunch Republican that would have said you're insane for even implying that I'd ever support a Democrat. Democrats be da deevil, ya no. (weak Waterboy reference)
Great analogy, whether it's appropriate or not!
It's not NASA that's the problem, but the current "brain" sitting behind the oval office. You know, the same one that got a minority of the popular vote in the last 2 elections?
Tang
Well, considering that UNIX98 also included Y2K compliance....
;)
But, I cannot answer your questions. I can offer an anecdote that my personal latest sighting of one of those specific UNIX versions was in 2002. I'd assume (I know) that later versions came on later hardware, and thus are 64 bit.
As for BSD/Linux, the majority of those installations are probably still 32 bit, but I'd expect that to change within the next 10 years, well before 2038. (OMG - a 10 year hardware cycle for PC hardware?!?! If you're not running Windows, sure!
But if it's Solaris 7 or above, and especially if you're running on a Ultrasparc IIe or III or above, you're already running 64 bit UNIX. See this list
Isn't the purpose of technology to make things easier, not harder? This would be another major negative to Vista - it makes things harder, and doesn't help me in any way that I've heard.
I don't need transparent windows for the most part either. A second monitor is much more helpful. The exception is the way that Apple's Mac OSX does property windows. Those being somewhat transparent is helpful at times.
That explains a lot, and may be why wireless is getting less usable. Maybe it's time to go look at a different band?
64 bit unices available:
HP UX 11.x
AIX 4.3
Solaris 7+ (note that UltraSparc processers IIe and III on up require 64 bit kernels)
SGI Irix since at least 6.5
Compaq Tru64
All of these are pre-2000.
BTW, a funny thing I found while looking this up - SCO as of 2004 had no UNIX98 compliant OSes. Maybe that explains their lawsuit based business plan?
I agree that running a VM for 80% of your time seems senseless. There could be cases where it's a good idea though, or where you don't even notice it.
Don't worry - that asteroid will hit us in 2036, thus solving the great 32-bit unix timestamp.
(does anyone run 32-bit unix anymore? Windows, yes, but Unix?)
Interesting. I was wondering what happened. My WPA connection with my MBP fails to restart when coming out of sleep, sometimes for minutes. Once, after connecting to an open wireless network, it actually required a reboot! (The first in 3 months)
I've noticed that if I enter the same room as my router and bring the MBP out of sleep mode, it tends to connect within seconds, and that connection then holds even when leaving the house. I've also noticed a lot more potential networks to connect to of the format nWIREnnn which didn't show up before. These are fleeting, and I haven't checked whether they are the same consistent set of networks, or whether it's some cruft that's confusing the driver.
My router is a Netgear 802.11g router, btw.
As for connecting up to open networks, cycling your airport will sometimes let you connect without rebooting. I should also note that I haven't updated in about a week or two, so I haven't seen the latest update and don't know how that would affect me.
Tripedalism
You're assuming that their statement "reformatted the backup hard drive" meant a single physical hard drive. It could just as easily been a RAID array, or a SAN. I would certainly not bet on what they meant by "hard drive" in that story, nor that they even got the actual technical facts right.
The story states he deleted the mains, then reformatted the backup, and the tapes were corrupted. Sounds like a top-notch tech there if it were merely single hard drives. Getting a backup RAID array or SAN online is a little different though, and might lend itself to massive corruption if done incorrectly. Hint, reformatting a single disk is trivial to undo, so much so that even a reformat done by an ignorant twit should result in a recoverable drive with most of the information available. A RAID array or SAN however can be painful, although I doubt this was on a SAN.
I'm willing to bet you're underestimating that data by at least an order of magnitude. Remember, these are government documents they're scanning. There's no such thing as a 1 page document to get money from the government. :)
But I agree with your statement that DVDs are pointless. a couple of 100+GB tapes are a much better way to go, and much easier to deal with considering that this is mostly static data.
Scenario #3 is what they want. You do recall who owns 33% of XM, right? (ClearChannel)
That is cool - but think how fast that gets handed to the server that's serving you. Think about how much faster it'd be in RAID 10 (or was that 0 + 1 for Mirror + Striping)
The problem with ethernet connected drives is that the lag is higher than that of internal drives, which is the reason that internal drives still exist.
There are definitely things to like about this, though, as a $0.50 ethernet chip these days should be cheap enough to add transparently to almost any drive. I like the fact that ethernet cables are almost as small as SATA/SAS cables.
I'd agree, "prior art" is almost irrelevant on this one. Considering they've been in use for about 50 years I'd say this is a prime example of how the PTO is completely, hopelessly broken. What will they patent next? The worm screw pump?
But, I read the patent, and it is even worse than that. It's basically describing the basics of pointer storage as applied to linked lists.
.. And never take that final fatal promotion! I'd just like to get within 2 or 3 promotions of that final promotion!!!