My wife (an American citizen) moved here to Ottawa a few years ago and just got her Permanent Resident status a couple weeks ago. I can't say I'm upset that the possibility of going down there instead didn't even cross my mind.
Every day I send her news articles about the rights of US Citizens evaporating like cheap perfume. She's scared to travel to her home country anymore, and wouldn't, if it weren't for her remaining family and friends state-side.
That said, the article -- while scary as hell -- just seems like small fries now. I wasn't aware American citizens really had any rights left after W got through with them.
Amen to Stephen Baxter. Funny, I am currently reading Titan at the moment, having exhausted everything else he's written since Manifold: Space.
I'm currently in the middle of an entire acronym-laden chapter about the battered and careworn hulk of Columbia (sigh) deorbiting for re-entry. The story takes place in 2004 -- go figure.
Stephen Baxter is probably peeing his pants right now. I'm pretty chuffed, myself.
Oh sure, that's how it spread -- but you just know that some twat somewhere opened the original Pandora's box on this one in an email attachment titled "Cindy Crawford Strip Tease.scr" and that was that.
... to stop executing screensavers, executables, etc. when they appear in their mailbox? I guess it doesn't matter since it only takes the weakest link to compromise an entire corporate network.
Linux might be a bit of a hard line. Going back to PINE, on the other hand, suddenly seems considerably more appealing.
Here's to hoping there isn't any military hardware being launched into space publically or privately -- although I guess it's a couple decades too late to be fussy about it now.
Had no idea about the post but it makes sense -- thanks for the insight. Whose interests will subsidize the continued development of private space flight now? If the X Prize is won, will the other contendors abandon their craft, or wait to get approached by Boeing?
... When I read the news as well, about the upcoming test flight.
For one who's done a lot of reading of both science fiction as well as cosmology, the stars seem so far out of reach for my short lifetime.
Getting into space isn't exactly reaching the stars, but it's the first step on the journey. I hope the mission goes smoothly and its success is a sign of things to come.
Since what you've quoted compared to what you've said are two different terms that might be being confused, let me clarify one point beforehand -- When I say I "upgraded" from 4x60GB to 4x200GB, it was done all at once and was a completely new volume which was built and formatted from scratch and the data restored from a backup, not an upgrade in the sense that I gradually added more storage to an already extant array without breaking it. The 60 GB drives were completely retired from service; I coined it an upgrade because the controller (and everything else -- RAID level, server, file system, role, purpose, etc.) remained the same as before.
Adding storage to a RAID 5 array without breaking it is, however, possible -- provided you have sufficient unused channels on your controller for the additional drive(s) as well as software (or possibly hardware, or a combination of both) support to do so. Having a drive of similar or identical geometry to those in your existing array is also highly desirable (or possibly necessary).
I've never delved into the technical details of how the procedure actually happens -- nor would I care to, as I have more important things to do than understand how it works so long as it works -- but let me take a guess.
A RAID 5's data is staggered across the containers in fixed-size blocks (often 16KB, IIRC, although this is usually adjustable). I imagine that when you add an extra disk to a volume, the software performs an operation similar to defragmenting a drive -- except that the blocks are re-organised (and necessary parity re-generated) across all of the disks to use the new drive (instead of making the files contiguous).
I have no idea if any other RAID levels besides level 5 support the addition of disks to increase the capacity of an existing volume, so someone else would have to answer for those. Also, anyone in the know may feel free to shoot my theory full of holes with documented fact.
As for partitioning a big drive down to a smaller size for inclusion in an array -- I'm not sure that's actually possible. In the case of a hardware RAID, a RAID table is written to every drive (before even the partition table, MBR, or anything else, I believe) containing information about the array -- like the ID number of the drive, the type and dimensions of the array, the condition of the array, etc. This would preclude the possibility of partitioning an individual drive manually. Software RAIDs may operate differently however -- someone in an earlier post did mention that you could still use the slack space in a larger disk with a soft RAID but I wouldn't know how to do it (or if that's even true).
I am however fairly certain that just about any type of RAID array is limited to the minimum available container size -- if you have one 2 GB and three 200 GB discs and made a RAID 0, 1, 10, or 5 (no idea about any of the rest) -- the maximum size you could make an array by using ALL of the containers would be 2GB per container (creating a maximum usable array size of 8 GB by opting for RAID 0) which is obviously a pitiful waste, whether or not you were able to use the slack 198 GB in the remaining three drives. You'd be better off excluding the 2 GB drive entirely and making a RAID 5 out of the three 200 GB drives, where you would have approximately 400GB usable space with redundancy should one of the drives fail.
I thank you for your corrections and clarifications -- I'm one of the many individuals who is happy to benefit from the advantages of a RAID array without knowing the pendantic details of how it -- and every other level of RAID -- operate.
That said, a few points worth making:
First, I was speaking solely for the benefit of the home user who, like myself, will probably not have any exposure to (or use for) any RAID levels besides 0, 1, the so-called "marketing term" 10, and 5 -- I'm not sure why I bothered mentioning 3 at all but there you are.
Your statement that RAID 0+1 or 10 is an recent invention to make a product appear more feature-rich is essentially true -- because in most cases, the hardware implementation (which is almost invariably only two channels on a motherboard or controller card) will provide little performance benefit over a RAID 1 on the same hardware. That said, it is immensely more important to mention RAID 10 than it was to mention RAID 2, considering the context of the question -- The likelyhood of a home consumer running into the former versus the latter while shopping for a solution to their problem is rather obvious.
Your mention of the "typical person who buys a cheap, 4 port IDE RAID card wants the most capacity he can get out of 4 drives so RAID 0 or 5 is inevitable," only makes apparent that the body of your remarks aren't merely tangential but diametrically opposite to the point: The poster isn't asking for help setting up a mission-critical computer for a Fortune 500's financials. It's a home user who is tired of losing data to hard drive failures. Besides the stipulation that the solution provide the perceived benefit of redundancy, it is assured that cost and accessibility are of the highest concern. You yourself state the typical cheap person is typically not performance sensitive and typically won't appreciate the performance benefit between RAID levels -- so why bother mentioning it at all, besides to typically toot your own horn?:)
Second, regarding '"data of data loss" by 4x' [sic], I admit my mathematics are bad at best, but at least err on the side of precaution -- Yours miss the point of the risk while misquoting me. The more drives you have, the greater the risk that one of them will fail prematurely. In the event of a disk failure (or just about any other type of fatal failure/user error) on a RAID 0, you are always guaranteed to lose 100% of the data regardless of which drive fails first (which could be in a matter of weeks, if someone sells you a returned drive -- which is exactly what happened to me recently, fortunate for me it was on a RAID 5), meaning you come out with 0% of your data at the end of the tunnel. In the case of two JBOD drives, regardless of which drive fails first you are still only out 50% of the data -- or less, because you can mitigate the risk by cross-duplication of critical data between the two devices (or even outright duplicate it in its entirety while you have the room). This is of course greatly simplified for the sake of argument, but it is sufficient to say that regardless of the risk, RAID 0 is not the solution for any individual seeking protection/redundancy of data since it only increments -- and not decrements -- the risk of losing data, the former being the reason for my dismissive mathematics, and the whole raison d'être of the article.
The third and final point that neither of us made (nor did I see mentioned in my skim through other posts) is without question the most important: don't rely on your RAID array to protect your data. Period. Any (non-zero) RAID array offers redundancy but is still a far cry from a backup; think of it as having borrowed time (should you lose a drive to hardware failure) to make the backups you should have been making all along, since -- as we have already covered -- neither performance nor high availability are major considerations in your decisi
Sorry, no. Most hardware RAID controllers support hot-swap and hot-rebuild capability -- especially RAID 5, which is most commonly used for high-availability business operations. They will rebuild the missing drive from parity during idle cycles with little-to-no noticeable loss to the user.
Mixing and matching makes and models is generally not a good idea -- particularly because the disk geometry is the crucial part here -- the geometry is going to determine how well the drives perform together; even if the drive is much faster but the geometries aren't similar, you're in trouble. It's a good idea to buy really popular drives for your RAID array that are likely going to be around for a while.
As far as question 2 -- It depends. Hardware RAID vendors usually offer this option. Adaptec does for its ATA/SATA models (but the maximum number of drives you can connect is four, I believe) and Promise only offers this for SOME of theirs -- check the specs closely. Just about any SCSI RAID card will allow you to add more drives later. As far as software RAIDs go -- I have no idea. I'll leave that to the softRAID geeks to answer.
Really depends on the type of RAID you'd like to implement.
RAID 0 stripes the data across 2 or more drives and therefore offers no redundancy (in fact, in a two-disk stripe you mutiply danger of data loss x4 compared to two individual drives -- because you not only double the possibility of failure with two disks as opposed to one, but stand to lose all of the data on both drives should one fail). In any event, no point in discussing it further since redundancy is the point.
RAID 1 offers redundancy by exactly duplicating the contents of a drive onto another drive, and needs exactly two drives. This is considered the most "fail-safe" method of RAID array although offers no performance benefits whatsoever.
RAID 10 (or 1+0 or 0+1) is a combination of RAID 0 and 1 and is nearly always done with four drives, although technically it can be done with six or eight (if your controller supports them). It offers both performance benefit and redundancy, although the cost of the "wasted" drive space is quite high.
RAID 3 involves using 3 or more drives, one of which contains parity information to rebuild the lost drive should any of the other drives fail. This is one of the least popular RAID formats and has more or less been totally replaced by RAID 5.
RAID 5 involves using 3 or more drives and writes parity information across all drives in the array, allowing one drive to fail with little to no performance loss. The failed drive can be replaced and the RAID rebuilt. Depending on your hardware/software, this can often be done hot without having to power down the system at all. It is one of the most commonly implemented RAID solutions because of the good mix between drive use (the price goes down the more drives you have in the array yet you can have as little as three), redundancy, and high availability.
There are others out there like RAID 50 but nothing worth mentioning, especially for a home user.
The only question left to you is whether the RAID will be run by hardware or software (software might be a good choice if you are already running Linux on the server, but you'll have to ask someone else about it because I don't know a thing about it). Personally I chose the hardware route years ago and bought an Adaptec 2400A, which is a four-channel hardware ATA-RAID card capable of RAID0, 1, 10, and 5 -- guess which I use. I use all four channels, each with a 200GB SATA hard drive. I've lived through a couple drive failures, a full drive upgrade (when I first bought the card it was 4x60GB drives) and even once where two drives RAID tables got zapped (I'll NEVER put my drives in removable cages again) and never lost a byte of data -- so the CAD$500 or so for the investment on the card was worth it.
600GB of storage means not having to worry about all those unlicenced-in-North-America-anime torrents running out of space any time soon.
... But I sure do like me some Ghost in the Shell: Stand Alone Complex.
I figured out that if I bought the Japanese domestic DVDs of all 26 episodes -- with no subs, of course -- I'd be paying something in the neighbourhood of CAD$1100 after purchase, shipping, and duty for one season of a television series I wouldn't understand.
So, y'know.
I agree. I think it's wonderful that Gnome and KDE are becoming "feature-rich" enough to compare/compete with Microsoft's offerings -- along with the size and advanced hardware requirements that comes along with it. Sure, they may turn out to be less 'tight' than a comparably-featured installation of Windows XP, but these things can improve over time (if there's a need or a desire to do so), and should anyway be chalked up to the price of the freedom to choose -- after all, you could be being forced to boot KDE every time you turn on your PC (does that remind you of anything?).
If KDE/Gnome's level of integration, overall appearance, etc. is what's required to make software companies -- followed by users -- decide they can now realistically support/run Linux, I'm all for it. There will still always be lightweight desktop alternatives for people who don't want or need the level of refinement that make Gnome and KDE 'big-boned'.
I generalise the situation by this simple exprsesion: The refinement/weight of the desktop is inversely proportional to the breadth of applications it performs -- to the point that for very specific applications (such as an htpc) you're going to run X without any desktop manager at all -- or just give X a miss entirely and go with a framebuffer.
Conversely, the bloated desktop junkies who want 3D screen savers, animated everything and nth-degree alpha-channel bitmap skins on everything on a machine where they check their email, play solitaire, and fill out TPS reports all day long aren't going to care if their desktop is bloated since they weren't using 98% of their 2.5GHz workstations for anything useful, anyway.
... But how will we ever make a sequel to Grim Fandango? By today's standards the engine is unremarkable. A remake could have more detailed graphics with the scenes rendered realtime; the characters were designed for very low-polycount rendering already.
What really set Grim Fandango apart was the writing, and the audio. The music and the voice acting were second to none. Without them the game loses its character.
In any event, the remakes likely won't get much further without having to start lifting audio, too -- I'm fairly certain LucasArts started doing that not long after Maniac Mansion 2...
Slashdot Mirror
Every day I send her news articles about the rights of US Citizens evaporating like cheap perfume. She's scared to travel to her home country anymore, and wouldn't, if it weren't for her remaining family and friends state-side.
That said, the article -- while scary as hell -- just seems like small fries now. I wasn't aware American citizens really had any rights left after W got through with them.
I'm currently in the middle of an entire acronym-laden chapter about the battered and careworn hulk of Columbia (sigh) deorbiting for re-entry. The story takes place in 2004 -- go figure.
Stephen Baxter is probably peeing his pants right now. I'm pretty chuffed, myself.
Oh sure, that's how it spread -- but you just know that some twat somewhere opened the original Pandora's box on this one in an email attachment titled "Cindy Crawford Strip Tease.scr" and that was that.
Linux might be a bit of a hard line. Going back to PINE, on the other hand, suddenly seems considerably more appealing.
Had no idea about the post but it makes sense -- thanks for the insight. Whose interests will subsidize the continued development of private space flight now? If the X Prize is won, will the other contendors abandon their craft, or wait to get approached by Boeing?
For one who's done a lot of reading of both science fiction as well as cosmology, the stars seem so far out of reach for my short lifetime.
Getting into space isn't exactly reaching the stars, but it's the first step on the journey. I hope the mission goes smoothly and its success is a sign of things to come.
Adding storage to a RAID 5 array without breaking it is, however, possible -- provided you have sufficient unused channels on your controller for the additional drive(s) as well as software (or possibly hardware, or a combination of both) support to do so. Having a drive of similar or identical geometry to those in your existing array is also highly desirable (or possibly necessary).
I've never delved into the technical details of how the procedure actually happens -- nor would I care to, as I have more important things to do than understand how it works so long as it works -- but let me take a guess.
A RAID 5's data is staggered across the containers in fixed-size blocks (often 16KB, IIRC, although this is usually adjustable). I imagine that when you add an extra disk to a volume, the software performs an operation similar to defragmenting a drive -- except that the blocks are re-organised (and necessary parity re-generated) across all of the disks to use the new drive (instead of making the files contiguous).
I have no idea if any other RAID levels besides level 5 support the addition of disks to increase the capacity of an existing volume, so someone else would have to answer for those. Also, anyone in the know may feel free to shoot my theory full of holes with documented fact.
As for partitioning a big drive down to a smaller size for inclusion in an array -- I'm not sure that's actually possible. In the case of a hardware RAID, a RAID table is written to every drive (before even the partition table, MBR, or anything else, I believe) containing information about the array -- like the ID number of the drive, the type and dimensions of the array, the condition of the array, etc. This would preclude the possibility of partitioning an individual drive manually. Software RAIDs may operate differently however -- someone in an earlier post did mention that you could still use the slack space in a larger disk with a soft RAID but I wouldn't know how to do it (or if that's even true).
I am however fairly certain that just about any type of RAID array is limited to the minimum available container size -- if you have one 2 GB and three 200 GB discs and made a RAID 0, 1, 10, or 5 (no idea about any of the rest) -- the maximum size you could make an array by using ALL of the containers would be 2GB per container (creating a maximum usable array size of 8 GB by opting for RAID 0) which is obviously a pitiful waste, whether or not you were able to use the slack 198 GB in the remaining three drives. You'd be better off excluding the 2 GB drive entirely and making a RAID 5 out of the three 200 GB drives, where you would have approximately 400GB usable space with redundancy should one of the drives fail.
That said, a few points worth making:
First, I was speaking solely for the benefit of the home user who, like myself, will probably not have any exposure to (or use for) any RAID levels besides 0, 1, the so-called "marketing term" 10, and 5 -- I'm not sure why I bothered mentioning 3 at all but there you are.
Your statement that RAID 0+1 or 10 is an recent invention to make a product appear more feature-rich is essentially true -- because in most cases, the hardware implementation (which is almost invariably only two channels on a motherboard or controller card) will provide little performance benefit over a RAID 1 on the same hardware. That said, it is immensely more important to mention RAID 10 than it was to mention RAID 2, considering the context of the question -- The likelyhood of a home consumer running into the former versus the latter while shopping for a solution to their problem is rather obvious.
Your mention of the "typical person who buys a cheap, 4 port IDE RAID card wants the most capacity he can get out of 4 drives so RAID 0 or 5 is inevitable," only makes apparent that the body of your remarks aren't merely tangential but diametrically opposite to the point: The poster isn't asking for help setting up a mission-critical computer for a Fortune 500's financials. It's a home user who is tired of losing data to hard drive failures. Besides the stipulation that the solution provide the perceived benefit of redundancy, it is assured that cost and accessibility are of the highest concern. You yourself state the typical cheap person is typically not performance sensitive and typically won't appreciate the performance benefit between RAID levels -- so why bother mentioning it at all, besides to typically toot your own horn? :)
Second, regarding '"data of data loss" by 4x' [sic], I admit my mathematics are bad at best, but at least err on the side of precaution -- Yours miss the point of the risk while misquoting me. The more drives you have, the greater the risk that one of them will fail prematurely. In the event of a disk failure (or just about any other type of fatal failure/user error) on a RAID 0, you are always guaranteed to lose 100% of the data regardless of which drive fails first (which could be in a matter of weeks, if someone sells you a returned drive -- which is exactly what happened to me recently, fortunate for me it was on a RAID 5), meaning you come out with 0% of your data at the end of the tunnel. In the case of two JBOD drives, regardless of which drive fails first you are still only out 50% of the data -- or less, because you can mitigate the risk by cross-duplication of critical data between the two devices (or even outright duplicate it in its entirety while you have the room). This is of course greatly simplified for the sake of argument, but it is sufficient to say that regardless of the risk, RAID 0 is not the solution for any individual seeking protection/redundancy of data since it only increments -- and not decrements -- the risk of losing data, the former being the reason for my dismissive mathematics, and the whole raison d'être of the article.
The third and final point that neither of us made (nor did I see mentioned in my skim through other posts) is without question the most important: don't rely on your RAID array to protect your data. Period. Any (non-zero) RAID array offers redundancy but is still a far cry from a backup; think of it as having borrowed time (should you lose a drive to hardware failure) to make the backups you should have been making all along, since -- as we have already covered -- neither performance nor high availability are major considerations in your decisi
Sorry, no. Most hardware RAID controllers support hot-swap and hot-rebuild capability -- especially RAID 5, which is most commonly used for high-availability business operations. They will rebuild the missing drive from parity during idle cycles with little-to-no noticeable loss to the user.
As far as question 2 -- It depends. Hardware RAID vendors usually offer this option. Adaptec does for its ATA/SATA models (but the maximum number of drives you can connect is four, I believe) and Promise only offers this for SOME of theirs -- check the specs closely. Just about any SCSI RAID card will allow you to add more drives later. As far as software RAIDs go -- I have no idea. I'll leave that to the softRAID geeks to answer.
Really depends on the type of RAID you'd like to implement.
RAID 0 stripes the data across 2 or more drives and therefore offers no redundancy (in fact, in a two-disk stripe you mutiply danger of data loss x4 compared to two individual drives -- because you not only double the possibility of failure with two disks as opposed to one, but stand to lose all of the data on both drives should one fail). In any event, no point in discussing it further since redundancy is the point.
RAID 1 offers redundancy by exactly duplicating the contents of a drive onto another drive, and needs exactly two drives. This is considered the most "fail-safe" method of RAID array although offers no performance benefits whatsoever.
RAID 10 (or 1+0 or 0+1) is a combination of RAID 0 and 1 and is nearly always done with four drives, although technically it can be done with six or eight (if your controller supports them). It offers both performance benefit and redundancy, although the cost of the "wasted" drive space is quite high.
RAID 3 involves using 3 or more drives, one of which contains parity information to rebuild the lost drive should any of the other drives fail. This is one of the least popular RAID formats and has more or less been totally replaced by RAID 5.
RAID 5 involves using 3 or more drives and writes parity information across all drives in the array, allowing one drive to fail with little to no performance loss. The failed drive can be replaced and the RAID rebuilt. Depending on your hardware/software, this can often be done hot without having to power down the system at all. It is one of the most commonly implemented RAID solutions because of the good mix between drive use (the price goes down the more drives you have in the array yet you can have as little as three), redundancy, and high availability.
There are others out there like RAID 50 but nothing worth mentioning, especially for a home user.
The only question left to you is whether the RAID will be run by hardware or software (software might be a good choice if you are already running Linux on the server, but you'll have to ask someone else about it because I don't know a thing about it). Personally I chose the hardware route years ago and bought an Adaptec 2400A, which is a four-channel hardware ATA-RAID card capable of RAID0, 1, 10, and 5 -- guess which I use. I use all four channels, each with a 200GB SATA hard drive. I've lived through a couple drive failures, a full drive upgrade (when I first bought the card it was 4x60GB drives) and even once where two drives RAID tables got zapped (I'll NEVER put my drives in removable cages again) and never lost a byte of data -- so the CAD$500 or so for the investment on the card was worth it.
600GB of storage means not having to worry about all those unlicenced-in-North-America-anime torrents running out of space any time soon.
... But I sure do like me some Ghost in the Shell: Stand Alone Complex. I figured out that if I bought the Japanese domestic DVDs of all 26 episodes -- with no subs, of course -- I'd be paying something in the neighbourhood of CAD$1100 after purchase, shipping, and duty for one season of a television series I wouldn't understand. So, y'know.
I agree. I think it's wonderful that Gnome and KDE are becoming "feature-rich" enough to compare/compete with Microsoft's offerings -- along with the size and advanced hardware requirements that comes along with it. Sure, they may turn out to be less 'tight' than a comparably-featured installation of Windows XP, but these things can improve over time (if there's a need or a desire to do so), and should anyway be chalked up to the price of the freedom to choose -- after all, you could be being forced to boot KDE every time you turn on your PC (does that remind you of anything?).
If KDE/Gnome's level of integration, overall appearance, etc. is what's required to make software companies -- followed by users -- decide they can now realistically support/run Linux, I'm all for it. There will still always be lightweight desktop alternatives for people who don't want or need the level of refinement that make Gnome and KDE 'big-boned'.
I generalise the situation by this simple exprsesion: The refinement/weight of the desktop is inversely proportional to the breadth of applications it performs -- to the point that for very specific applications (such as an htpc) you're going to run X without any desktop manager at all -- or just give X a miss entirely and go with a framebuffer.
Conversely, the bloated desktop junkies who want 3D screen savers, animated everything and nth-degree alpha-channel bitmap skins on everything on a machine where they check their email, play solitaire, and fill out TPS reports all day long aren't going to care if their desktop is bloated since they weren't using 98% of their 2.5GHz workstations for anything useful, anyway.
What really set Grim Fandango apart was the writing, and the audio. The music and the voice acting were second to none. Without them the game loses its character.
In any event, the remakes likely won't get much further without having to start lifting audio, too -- I'm fairly certain LucasArts started doing that not long after Maniac Mansion 2...