For every idot who's lost data from a RAID array there are 10,000 people who've lost data from drive failures.
Despite everyone's constant loud claims to the contrary, there is a class of data that is important enough that you not lose it to want it on a RAID array but not important enough that you not lose it to want to back it up. For me.. the line between these is gray but a basic rule of thumb is anything on a machine that I didn't create myself (i.e. that can be recreated fairly easily from external sources) does not need to be backed up. This class of data is probably roughly 99% of the data on most people's desktops. For this class of data, losing data is an annoying and time consuming headache but it's not catastrophic. Also, in a lot of cases like installed OS's and programs, backups are of limited use unless you can restore from scratch (which most people can't) In my setup I have a RAID 6 array backed up to a RAID 5 array that holds my unique data. Both my and my wife's primary desktops have mirrors as well though to keep the machines running reliably.
So, to those people who say RAID does nothing to prevent data loss.. you're obviously wrong. Stop talking.
The problem of rebuilds failing because there are multiple disks with undetected errors is a huge problem with RAID 1-5 arrays. The good thing is you can usually copy nearly all the data off an array that won't rebuild, and, if you have a way to force a disk back into the array that failed you can usually recover the data that you couldn't get from the other disks. This requires the ability to re-insert a disk to an array that was removed from it without rebuilding which is a damaging event, more damaging the longer the disk has been out.. but! it is possible. This is one of the reasons I like Linux's software RAID.
This is also why I think RAID 6 is the minimum level RAID you should use in large arrays now. The odds of running into a read failure on rebuild are increasing dramatically as disk size increases. It's also why RAID 1-5 arrays need to be much more careful about decoupling "failed" drives. They usually spit them out at the first write failure which is a terrible idea given that the array becomes completely unprotected as soon as that happens. RAID 5 controllers need to be smarter about how they manage rebuilds, and should scan all areas of remaining disks before de-coupling a failed drive and rebuilding any unreadable sectors on the other disks if possible. The good thing is RAID 6 completely eliminates the issue. Delaying de-coupling is still a good idea but the odds of a rebuild failing are negligible.
Having done RAID many times in different ways over the years I'd say that as long as the version of Windows 7 you plan on running supports it, I'd do it. But.. be warned! There may be issues.
For one, I ran into one copy of Comodo firewall that completely blew up the networking stack on XP if dynamic disks (required for Windows software RAID) were present. Also, any BSOD will be followed by a pointless RAID rebuild where it completely copies the contents of one of the drives to the other slowing disk IO for a long time. Also, the entire disks including partitions that aren't RAIDed will need to be configured as dynamic disks which can cause issues accessing them from DOS and Linux. Not usually an issue but it's worth pointing out. One of the things I liked to do was install an OS on each drive that was bootable and have the Stripe/Mirrored partitions accessible from both OSs. If you are going for this type of configuration there is something important in the order with which you create the partitions and convert the disks to dynamic disks. I think you have to convert each disk to dynamic from the OS that boots from that disk in order for it to remain bootable. If OS that boots from disk 1 converts disk 2 the OS on disk 2 will be rendered unbootable (if I remember correctly). If you are going for this type of config, install each OS and have it convert it's disk to dynamic, then create the mirror/stripe partition(s). Other than that, the ability to put the disks in any Windows machine and access the data makes software RAID the clear winner here.
In response to all the posts about how better to safeguard data by backing it up, he's not asking about that. He's asking about which way we'd recommend doing RAID. Your suggestions are off topic. Every time RAID is discussed the same arguments are made. Backups aren't RAID. RAID isn't a backup. Enough already.
I've done software RAID for a long time and I'm a firm believer that it's the right way to do RAID in a lot more cases than it's typically used. Hardware RAID has many proponents and is obviously a profitable industry so there is a lot of money being spent based on it's perceived advantages so I'm in the minority here but that doesn't mean I'm wrong. Expensive RAID array's advantages are getting harder to justify now that Serial ATA has brought dedicated per-drive bandwidth, fast cold (if not hot) swapabililty, and now that multi-core CPUs can handle the load of the software RAID work gracefully. The niche where software RAID makes sense is getting bigger every year. Many people argue that software RAID is too slow, that you need the hardware controller to offload the calculations. I'd like to throw out the fact that I use an 8 drive Linux based software RAID 6 array for my primary storage. Bonnie++ clocks it at 215.2 MB/s on block writes and 263.5 MB/s on block reads. CPU utilization is 45/31% on those respectively but with a quad core machine, using half of one core to do my IO processing when it's writing at that speed is perfectly acceptable. Remember, this is RAID 6 with 2 differently calculated parity chunks, not goofy simple RAID 5 and it's CPU is just a AMD Phenom 9600 Quad with 8 gigs of RAM, not something exotic. I run many virtual machines on top of this array simultaneously and they are nice and snappy. The configuration works quite well and I'd recommend it for a inexpensive, high speed virtual machine server configuration.
For my Windows desktop, I'm currently using my motherboard's Intel Matrix Raid capabilities and have configured half of my drive as a mirror and half as a stripe. The setup has worked seamlessly and I'd recommend it as a reasonable alternative to software RAID. It's slightly cleaner from Windows's point of view and the Intel Matrix controllers are fairly common and, from what I understand, I could put the disks into any machine with Intel matrix raid and access the data in a pinch. I honestly probably would have done software RAID had Vista supported mirroring.
I'm going to have to agree with TFA on most points, including the bad security model. I'm a huge Linux fan and I do think eventually we'll all be running it but to get to that point we need to fix a lot.
The user based security model doesn't match well with the way desktop computers are used nowadays.
They missed something in the article and that is 3D. 3D in Linux is broken and while the fix is being worked on, it's not done and we won't know if it really fixes the problems until its done.
This guy stole $200,000 worth of stuff before he was caught. What about the guy who stole $100,000 worth of stuff? The guy who stole a few cameras? The guy who just snagged one laptop for his kid? If it takes stealing $200,000 worth of equipment before they will catch you then why not steal?
The outrage shouldn't be over how much bad this guy did, but how broken the system is if he could steal that much before being caught.
Furthermore, I think anybody who needs an external authority to tell them that they shouldn't kill, steal, or harm others, is morally flawed. Government is an external authority that tells people that they shouldn't kill, steal, or harm others... But do you *need* that authority to tell you it's wrong? Would you steal from your neighbor if it was legal? Or kill them? Or beat them up? If you were.. you have a shitty moral compass and I don't want you as my president. We've already got one like that.
Not discussing Ron Paul here is deeply sad. I expect the Slashdot crowd to be able to see past the glossing over of issues that the corporate media pushes and be able to see the importance of the battle of philosophies of government that the Paul campaign represents. The other 3 candidates are pretty generic politicians. They'll say whatever they think is most likely to get them elected. There are generic platitudes and dealing with as few issues in the most generic way possible and your typical political BS but no real substance there. In the meantime there are debates raging over philosophies of government, its just not being reported on so you don't hear about it much. For some reason actually covering politics, not politicians and political trivia is distasteful in American media.
In general Ron Paul's most controversial policy seems to be the one opposing the war and unfortunately he's not very charismatic about explaining why he holds his position. He mostly points to the constitution to defend his position expecting people to understand the reasons behind why the constitution says what it does, but that goes over most people's heads these days.
There are huge competing ideologies in our government right now and a Ron Paul candidacy shines an important light on a few debates we need to be having. They are uncomfortable topics for a lot of powerful people and so getting the debate started is difficult but its important enough that we Slashdotters should be behind it 100%.
My biggest issue with the government as it exists now is our foreign policies and how they keep pushing "Making the world safe for democracy" while, in practice, they do the exact opposite. Any action by us, designed to influence the government of a foreign country is an action that subverts and does damage to democracy. This goes for toppling, subsidizing, backing militarily, or any other type of support for any government, good or bad. Any time we do these things we do them either because they are in the best interest of Americans or because Americans think its the right thing to do. The problem is that this inevitably taints whatever outcome, good or bad, with the fact that it was, at least partially, a reflection of the interests of Americans not the interests of the citizens of that country. As democracy is only really democracy if the government is one that solely represents and acts upon the will of it's citizenry. Any government which is serving the interests of the US in any way, isn't really a democracy. All our meddling in other countries, no matter how well intentioned, undermines peace and security there. This is true, not just in theory, but it's visible in practice in almost every country the CIA has tried to advance the "American" agenda. The disasters they have wrought have killed millions and subverted democracy all over the world and very likely even here in the US. Our government shouldn't, and arguably doesn't have the authority to do these things and yet they keep happening.
Just look at the cold war for example: We were so worried about Russia spreading communism to the world that we were willing to commit atrocities to keep it from happening. Look a few years later and not only are they not trying to push communism on other countries, they've abandoned it themselves. Now.. instead of committing atrocities to save a bad thing from happening.. looking back all we really did was commit atrocities. In fact, contrary to the goal of making the world safe for us, these middlings have made the world a very unsafe place for us. This is the thing the war-mongers don't want people to see. They want you to see everything in terms of either we keep fighting until we win or we give up and they win. A battle of ideas isn't like that though. Often, the harder you fight the worse you make everything.
The 3 major Republican candidates are trying to sell me protection from nebulous bad guys who are out to get me. I'm not buying it. They are ending up using that power
In the time it took to develop Perl 6, other programming languages have been conceived, implemented, used and abandoned.
Fortran has sucked since the 1950s. C has sucked since the 1970s. Java has sucked for 13 years. How long should "doing it right" take?
Theres a lot of revolutionary features in parrot. Its unpopular so its fun to bash it but I expect the.net people are laughing publicly and privately frantically trying to catch up so as to not lose market share permanently. If parrot does nothing else but push widespread adoption of the technologies they pioneered, it'll be huge. If the other guys are asleep at the switch.. I expect it will gain lots of fans.
The same way people who bash Linux just don't get it, the people who bash parrot just don't get it. Peel back the covers and you'll be impressed.
C might be "high level" by some definition of the word but where I come from C is a low level language one step up from assembly. Calling C a high level language with Java, C# and Ruby around is like calling a 4 story building tall when theres a 400 story skyscraper next to it.
Parrot might be able to give you what you are looking for if it saves compiled bytecode, or something like that, to disk and then mmap() it back into memory to actually execute it.
You are assuming that parrot isn't creating a libfoo.so.0 file. Parrot is designed to be able to compile programs the old-fashioned way (once, not on the fly, saved to a file) as well as on the fly.
.net programs require the.net framework to be installed to be used. Its my understanding that one of parrot's goals is to be able to create standalone OS binaries that don't need a giant environment be installed to run. This may seem like a small issue, temporary and easy to fix, but its not.
Say, for example,.net 4.2 includes this great new feature that will make your little tool that you want to run on the 4000 machines you administrate that much better but they all have 3.5 installed. Unfortunately.. 4.2 has a bug that breaks an important application everyone uses. Even more unfortunately, you don't figure that out until you install it on those 4000 machines so you can run your utility. You're fired.
The issue I brought up about shared memory for shared libraries is another.net limitation which many people dismiss as unimportant but deeply limits the utility of the languages in creating code in wide-spread use on a machine. A library that is used by every application on the system (networking library, registry type library, etc) that is created in a way that its memory is unsharable is extremely inefficient in comparison to one that is sharable. With java (and I believe.net) a fundamental design principal is that each process's memory is completely isolated from all other processes (for the components written in.net and java). This leads to deep inefficiencies in memory use but more importantly in cache use and probably accounts for some of the "molasses" feeling of those environments.
Yes.. but.. (from what I understand) unlike.net and java, you will be able to compile binary versions of your applications for distribution that don't rely on a giant, bloated, pre-installed runtime *and* they aren't compiled to bytecode at program launch. Also, the register based parrot engine underneath is supposed to be much faster than the purely stack based.net and java VMs.
Binaries that launch and run at the speed of C with the ability to write in a high level language? Sounds like the future of programming to me.
The big question I have is will multiple programs be able to share memory to store a commonly used library like C can. One of the things I think causes big inefficiencies in modern languages is the way they don't share memory well. If I have a C library that has, say, 5 megs of pure code in it and I'm running 10 copies of a program that uses it, all 10 can use the same 5 meg copy with only parts that are written to localized to the application. With high level languages, I need to use 50 megs of memory for each. Can Parrot share like C does and have all copies of the program reference the one copy stored in memory? Java and.net can't use this since their libraries aren't implemented on a level where that is possible. Parrot probably could but I'm not sure if it does. Does anyone know the answer to that?
Think about it for a second... If a highway has a carrying capacity of 400 cars per minute and you try to push 500 cars per minute through it how many cars per minute will end up sitting and waiting somewhere?
People seem deeply perplexed by this problem of "who is responsible" for them having to sit in traffic and for some reason always expect it to be some drivers fault. Unless the traffic jam you are in lasts about 10 seconds, or there is a real accident, then odds are the trigger event for the slowdown is irrelevant. The slowdown was inevitable.
Your road isn't big enough. It's operating at near or over saturation. Any road that is being pushed over its saturation point will buckle and traffic will stop, plain and simple. The trigger for the slowdown can be a person hitting the breaks, a momentary breakdown, even someone deciding they need to play it safe and leave a little more room between them and the next car can start the wave that ends up in a traffic jam. But the question isn't how the wave starts.. that question is meaningless. The question is why doesn't it dissipate like all the other waves and that answer is saturation. If the road wasn't saturated, the wave that got started when the guy hit his breaks too hard would have lasted a few cars and then dissipated as people worked around it and space between cars was compressed. Traffic would have absorbed the shock and kept moving.
Instead of tinkering with exotic breaking systems why not simply build more and bigger roads to accommodate the traffic. Better yet.. make tax incentives for companies who have employees work from home so less people are actually on the road.
Since Wi-Fi is generally 11-54 mbit they're only talking 1-5 gbit. The mentioned use is for video so it sounds like they are trying to connect displays to devices that generate output, i.e. replacing a monitor cable. For comparison DVI is 3.7 gbit, DVI-D 7.4. Most likely they are talking about the 1-2 gbit range since if it was in the 5gbit range they'd probably have said so instead of 100x wifi. That data rate would only be useful for low-resolution displays like HDTVs, not for general purpose computer monitor use. The devices would likely need to be close to each other due to the high frequencies. It sounds like they may be targeting removing the cable requirements home theater systems or something similar.
Personally.. I like cables for hooking up video. Wireless is buggy, snoopable, power hungry, and hard to set up (with 4 transmitters and 4 receivers, how to you configure what displays where?) Cables, while bulky and sometimes annoying have an incredibly easy UI. Plug one end here, the other end there, the things are connected. Want to change it? plug the wire in somewhere else.
Linux needs to be able to dislodge Microsoft Office in order to become a viable option for large-scale replacement of Windows in the corporate environment. Open Office can handle most of the document creation work and Firefox/Thunderbird do a great job at handling web browsing and email.
What is needed now is something that can integrate with an Exchange server's calendaring and also integrate with a robust open-source calendering server system to replace Outlook's calendaring functionality with an open, standards based system. We need to embrace and extend Outlook.
I've been saying it for 7 years now... 2010 is the year. This patent thing may put it off a few years but I doubt it. Every year Linux gets better and better, closer and closer.
Of the 235 probably 150 will be trivial to write around or just be thrown out as obvious.. most of the rest will take significant time and effort to work around but that work will eventually be done. I suspect somewhere around 10 will cause features to be removed from Linux distributions but they'll be features we can live without. During the process the patent system will hopefully be reigned in.
Any way this ends up it will be an interesting and ultimately healthy fight for FOSS.
Schools are government organizations and if you are a child, the schools rules are forced upon you. A government organization removing people's free speech rights is a horrible idea and quite correctly prohibited by the constitution.
Administrators like to think of the students like stores think of customers... here's how things are, take it or leave it. The problem is, there is no "leave it". Truancy laws say you have to go somewhere and essentially no children have the financial means to chose to go somewhere else.
If you take away a kids right to free speech during school hours you haven't really taken it away, you've just pushed it to non-school hours. This seems reasonable. If you take it away at home, however, you've removed it completely.
Repression of free speech is unconstitutional because it is a stupid idea. Inevitably it leads to the opposite of what it intends. Bad thoughts and ideas fester in private and die when the light of many eyes are shown upon them.
Just because something affects a school doesn't mean school administrators have control over it. They have tons of tools at their disposal, ones that will actually work. Try those. Get creative.
The metric system is better for people doing calculations.
The imperial system is better for people who cook, fix, build and actually do things.
I'm tired of scientists whining about how everybody doesn't make their math easier, especially when computers do 99% of their math for them anyway.
The English units are intuitive. 1 cup, 1 pint, 1 teaspoon are designed around quantities that are easy to visualize. 350 milliliters is designed for easy math but it's hard to visualize 350 1 milliliter things.
Also, in the English measurement system you find systems divisible by 3 and 4 which are very handy for building things. Metric is divisible by 2, 5 and 10 which is useful for doing math.
The metric system was designed before computers to make certain math in science easier. That goal is outdated. Metric units are inferior now.
The question as to what takes a road a saturation and pushes it over the edge is meaningless. It can be anything. Someone trying to go too fast, someone hitting the breaks a bit too hard, someone changing lanes.. any little blip can start things off. As many people have noted, once the road is at saturation, eventually something will push it over the limit.
The real question here is "How do we stop traffic jams?"
The answer is simple: don't saturate the road. Trying to avoid traffic jams on saturated roads is like trying to dodge raindrops.
A road below saturation has room enough between cars to handle little impulses. The impact of someone slowing traffic will be absorbed and dampened by the extra space instead of amplified. Roads that are built with excess capacity do this naturally so one way to achieve this is to simply make roads bigger. Most roads are built so far below the load that would naturally be put on them though that it would require a massive (2-3x) increase in their capacity to make this work.
A similar solution is to simply have more highways. Right now they only account for about 2% of the roads in the US and handle about 25% of the traffic. Taking land to build roads is always controversial but every second you sit in a traffic jam should bring you one second closer to understanding why it has to happen.
Other methods would include things like blocking access to the highway when it is near capacity. There have been failed attempts at this in California but they weren't designed to keep the roads under the saturation point so they were doomed to fail. A correct solution would have to outright deny access to people if the road is near saturation until it dropped below. This solution would probably be more annoying than the problem though.
I had a similar problem with drive caching swapping out all my running programs. I have a linux box where I back up one drive to another each night at 3am. The machine has 2 gigs of RAM but when I came in the next day it had swapped out all my programs to use it for disk cache and never swapped them back in. Every time I clicked anything in a program theres was nasty delay as it loaded that functionality off of disk. The UI was horribly sluggish for a while and I kept finding new things that were swapped out as the day went on. It was very frustrating.
I ended up adding a new cron job at 6am that turns off swap space and then turns it back on. Everything is back in real RAM by the time I get to the machine.
I keep seeing these statistics thrown around, but they just don't ring true.
There's a reason!
The thing they are looking at is "assets" calculated by how much you have - how much you owe. Under this model a very wealthy couple who together make $300,000 per year, own a million dollar home, drive $70,000 cars, have a nanny, a maid, take expensive vacations, buy anything they could possibly want and are by most measures very wealthy could be considered having nothing. The fictional couple above, if they are young or not saving properly for retirement, may have debt exceeding their assets in which case they would be considered as having no net worth at all. To this study they would be "poor" which they are most definitely not.
The problem with using net assets as a measure of wealth is people use these numbers in discussions of societal fairness and equity. These numbers are completely meaningless in that context. The only thing its really appropriate for doing is discussing how well people are saving for retirement because that's the primary reason to have money pooling in the way that would show up here. People use these numbers to point out economic inequity and imply things about standard of living which are completely inappropriate.
Most of WoW is fun and not bottable. People who think it's bottable generally don't understand that all the interesting difficult game content happens after you hit the level cap.
Blizzard has historically done a great job at catching the bad guys without catching innocent people. They probably caught some people who didn't deserve it but they were generally fair. This latest round wasn't like that at all. It would appear that they are currently targeting wow-glider and anti-afk macros so aggressively, they are using unreliable tactics prone to false positives.
I've been paying attention to this latest wave of bannings because my guild's main tank got hit and he has no idea why. I've talked to him at length and tried to figure out possible triggers. Nothing we've come up with yet would explain things. The most likely candidates include logging in from friends machines but he's talked to them and nobody seems to have installed any of the types of cheats that I know about.
Right now Blizzard's justice system is operating in a black box. They hide behind email auto-responders that claim to have "reinvestigated" their "extensive in-game logs" and technicians who handle appeals but don't seem to be given any details beyond "cheat detected". We don't know what their evidence is. We have to trust that when they ban it's legitimate.
Legally, Blizzard hides behind WoW's terms of service that says basically that they can yank your account anytime they feel like it with or without a valid reason. This is only part of the agreement between Blizzard and us though. Before investing hundreds and thousands of hours into characters that they can rip away at any time we expect them to be fair about doing so. We expect them to have a good reason and reasonable evidence and until now, they have upheld that.
In this latest round of bannings Blizzard banned unfairly. They banned innocent people based on flimsy faulty evidence and called them cheaters and lairs when they complained. The evidence that this was happening was there before this announcement but this is irrefutable proof. Unbanning the people who got banned for using Linux is a good start but what about all the other people who were innocent and got banned? Are we to believe that these are the only unfair bannings? The only mistake made happened to be in a tight active community with commercial representatives? Doubtful. I want to see Blizzard chow down on a big piece of humble pie this Thanksgiving and start actually looking at the evidence against people and un-ban the innocent victims of their latest dragnet banning scheme.
Blizzard usually does a great job at balancing protecting it's customers and protecting the gaming experience. This latest round of bannings stands in stark contrast to that. Someone, somewhere obviously got overzealous and decided to ban based on flimsy faulty evidence. I won't have any trust in the fairness of Blizzard's system until I see more people get un-banned.
Technology has made it increasingly easy to avoid interacting with other people. When was the last time you sat on your porch and chatted with a neighbor?
Like it or not technology is connecting people more, not less. We connect in different ways but those ways are easier, safer, more convenient and often more fun. How much porch sitting do you really think has been lost? Does it really compare with 10s of millions of people hanging out, having fun, helping each other and making friends in MMORPGs? How about the young couple with a baby at home who can't so much as catch a movie without hours of prep followed by worrying about what the person watching their kid is doing? For me, the kid goes to sleep at night and I start raiding.
Spending 4 hours with friends playing a game, joking around, playing music, making fun of each other, etc. It could be softball or WoW. More and more it's online games though since they are dramatically more accessible. I agree, there is something about real face to face friends lacking in online friends but there are upsides too.
Gamers may be divorced from the reality that OReilly cares about, but OReilly is also divorced from the reality that gamers are tuned into. Its intellectually lazy to whine about the destruction of what we had and ignore the value of what is being created in it's place.
I like the idea of electronic voting. In the end it will be more secure than paper when it is done right but as it is now, security and tamper resistance haven't even been made a priority. If these things had the kind of security that you find in an X-Box it would be an enormous improvement.
They argue essentially "We cant tell you how the system works because if you knew that the system would be compromised." The problem is that some people *do* know how it works and other people can and have figured it out. Therefor The system is already compromised.
A system done right will be able to be watched very carefully. It will have an audit trail, it will find problems and when inspected carefully, catch people attempting to commit voter fraud. Any system that doesn't allow you to double check it is not tamper evident and shouldn't be used. Any system that comes out perfectly and finds no issues is probably not tamper evident.
Slashdot Mirror
For every idot who's lost data from a RAID array there are 10,000 people who've lost data from drive failures.
Despite everyone's constant loud claims to the contrary, there is a class of data that is important enough that you not lose it to want it on a RAID array but not important enough that you not lose it to want to back it up. For me.. the line between these is gray but a basic rule of thumb is anything on a machine that I didn't create myself (i.e. that can be recreated fairly easily from external sources) does not need to be backed up. This class of data is probably roughly 99% of the data on most people's desktops. For this class of data, losing data is an annoying and time consuming headache but it's not catastrophic. Also, in a lot of cases like installed OS's and programs, backups are of limited use unless you can restore from scratch (which most people can't) In my setup I have a RAID 6 array backed up to a RAID 5 array that holds my unique data. Both my and my wife's primary desktops have mirrors as well though to keep the machines running reliably.
So, to those people who say RAID does nothing to prevent data loss.. you're obviously wrong. Stop talking.
The problem of rebuilds failing because there are multiple disks with undetected errors is a huge problem with RAID 1-5 arrays. The good thing is you can usually copy nearly all the data off an array that won't rebuild, and, if you have a way to force a disk back into the array that failed you can usually recover the data that you couldn't get from the other disks. This requires the ability to re-insert a disk to an array that was removed from it without rebuilding which is a damaging event, more damaging the longer the disk has been out.. but! it is possible. This is one of the reasons I like Linux's software RAID.
This is also why I think RAID 6 is the minimum level RAID you should use in large arrays now. The odds of running into a read failure on rebuild are increasing dramatically as disk size increases. It's also why RAID 1-5 arrays need to be much more careful about decoupling "failed" drives. They usually spit them out at the first write failure which is a terrible idea given that the array becomes completely unprotected as soon as that happens. RAID 5 controllers need to be smarter about how they manage rebuilds, and should scan all areas of remaining disks before de-coupling a failed drive and rebuilding any unreadable sectors on the other disks if possible. The good thing is RAID 6 completely eliminates the issue. Delaying de-coupling is still a good idea but the odds of a rebuild failing are negligible.
Having done RAID many times in different ways over the years I'd say that as long as the version of Windows 7 you plan on running supports it, I'd do it. But.. be warned! There may be issues.
For one, I ran into one copy of Comodo firewall that completely blew up the networking stack on XP if dynamic disks (required for Windows software RAID) were present. Also, any BSOD will be followed by a pointless RAID rebuild where it completely copies the contents of one of the drives to the other slowing disk IO for a long time. Also, the entire disks including partitions that aren't RAIDed will need to be configured as dynamic disks which can cause issues accessing them from DOS and Linux. Not usually an issue but it's worth pointing out. One of the things I liked to do was install an OS on each drive that was bootable and have the Stripe/Mirrored partitions accessible from both OSs. If you are going for this type of configuration there is something important in the order with which you create the partitions and convert the disks to dynamic disks. I think you have to convert each disk to dynamic from the OS that boots from that disk in order for it to remain bootable. If OS that boots from disk 1 converts disk 2 the OS on disk 2 will be rendered unbootable (if I remember correctly). If you are going for this type of config, install each OS and have it convert it's disk to dynamic, then create the mirror/stripe partition(s). Other than that, the ability to put the disks in any Windows machine and access the data makes software RAID the clear winner here.
In response to all the posts about how better to safeguard data by backing it up, he's not asking about that. He's asking about which way we'd recommend doing RAID. Your suggestions are off topic. Every time RAID is discussed the same arguments are made. Backups aren't RAID. RAID isn't a backup. Enough already.
I've done software RAID for a long time and I'm a firm believer that it's the right way to do RAID in a lot more cases than it's typically used. Hardware RAID has many proponents and is obviously a profitable industry so there is a lot of money being spent based on it's perceived advantages so I'm in the minority here but that doesn't mean I'm wrong. Expensive RAID array's advantages are getting harder to justify now that Serial ATA has brought dedicated per-drive bandwidth, fast cold (if not hot) swapabililty, and now that multi-core CPUs can handle the load of the software RAID work gracefully. The niche where software RAID makes sense is getting bigger every year. Many people argue that software RAID is too slow, that you need the hardware controller to offload the calculations. I'd like to throw out the fact that I use an 8 drive Linux based software RAID 6 array for my primary storage. Bonnie++ clocks it at 215.2 MB/s on block writes and 263.5 MB/s on block reads. CPU utilization is 45/31% on those respectively but with a quad core machine, using half of one core to do my IO processing when it's writing at that speed is perfectly acceptable. Remember, this is RAID 6 with 2 differently calculated parity chunks, not goofy simple RAID 5 and it's CPU is just a AMD Phenom 9600 Quad with 8 gigs of RAM, not something exotic. I run many virtual machines on top of this array simultaneously and they are nice and snappy. The configuration works quite well and I'd recommend it for a inexpensive, high speed virtual machine server configuration.
For my Windows desktop, I'm currently using my motherboard's Intel Matrix Raid capabilities and have configured half of my drive as a mirror and half as a stripe. The setup has worked seamlessly and I'd recommend it as a reasonable alternative to software RAID. It's slightly cleaner from Windows's point of view and the Intel Matrix controllers are fairly common and, from what I understand, I could put the disks into any machine with Intel matrix raid and access the data in a pinch. I honestly probably would have done software RAID had Vista supported mirroring.
I'm going to have to agree with TFA on most points, including the bad security model. I'm a huge Linux fan and I do think eventually we'll all be running it but to get to that point we need to fix a lot.
The user based security model doesn't match well with the way desktop computers are used nowadays.
They missed something in the article and that is 3D. 3D in Linux is broken and while the fix is being worked on, it's not done and we won't know if it really fixes the problems until its done.
This guy stole $200,000 worth of stuff before he was caught. What about the guy who stole $100,000 worth of stuff? The guy who stole a few cameras? The guy who just snagged one laptop for his kid? If it takes stealing $200,000 worth of equipment before they will catch you then why not steal?
The outrage shouldn't be over how much bad this guy did, but how broken the system is if he could steal that much before being caught.
Not discussing Ron Paul here is deeply sad. I expect the Slashdot crowd to be able to see past the glossing over of issues that the corporate media pushes and be able to see the importance of the battle of philosophies of government that the Paul campaign represents. The other 3 candidates are pretty generic politicians. They'll say whatever they think is most likely to get them elected. There are generic platitudes and dealing with as few issues in the most generic way possible and your typical political BS but no real substance there. In the meantime there are debates raging over philosophies of government, its just not being reported on so you don't hear about it much. For some reason actually covering politics, not politicians and political trivia is distasteful in American media.
In general Ron Paul's most controversial policy seems to be the one opposing the war and unfortunately he's not very charismatic about explaining why he holds his position. He mostly points to the constitution to defend his position expecting people to understand the reasons behind why the constitution says what it does, but that goes over most people's heads these days.
There are huge competing ideologies in our government right now and a Ron Paul candidacy shines an important light on a few debates we need to be having. They are uncomfortable topics for a lot of powerful people and so getting the debate started is difficult but its important enough that we Slashdotters should be behind it 100%.
My biggest issue with the government as it exists now is our foreign policies and how they keep pushing "Making the world safe for democracy" while, in practice, they do the exact opposite. Any action by us, designed to influence the government of a foreign country is an action that subverts and does damage to democracy. This goes for toppling, subsidizing, backing militarily, or any other type of support for any government, good or bad. Any time we do these things we do them either because they are in the best interest of Americans or because Americans think its the right thing to do. The problem is that this inevitably taints whatever outcome, good or bad, with the fact that it was, at least partially, a reflection of the interests of Americans not the interests of the citizens of that country. As democracy is only really democracy if the government is one that solely represents and acts upon the will of it's citizenry. Any government which is serving the interests of the US in any way, isn't really a democracy. All our meddling in other countries, no matter how well intentioned, undermines peace and security there. This is true, not just in theory, but it's visible in practice in almost every country the CIA has tried to advance the "American" agenda. The disasters they have wrought have killed millions and subverted democracy all over the world and very likely even here in the US. Our government shouldn't, and arguably doesn't have the authority to do these things and yet they keep happening.
Just look at the cold war for example: We were so worried about Russia spreading communism to the world that we were willing to commit atrocities to keep it from happening. Look a few years later and not only are they not trying to push communism on other countries, they've abandoned it themselves. Now.. instead of committing atrocities to save a bad thing from happening.. looking back all we really did was commit atrocities. In fact, contrary to the goal of making the world safe for us, these middlings have made the world a very unsafe place for us. This is the thing the war-mongers don't want people to see. They want you to see everything in terms of either we keep fighting until we win or we give up and they win. A battle of ideas isn't like that though. Often, the harder you fight the worse you make everything.
The 3 major Republican candidates are trying to sell me protection from nebulous bad guys who are out to get me. I'm not buying it. They are ending up using that power
Theres a lot of revolutionary features in parrot. Its unpopular so its fun to bash it but I expect the
The same way people who bash Linux just don't get it, the people who bash parrot just don't get it. Peel back the covers and you'll be impressed.
C might be "high level" by some definition of the word but where I come from C is a low level language one step up from assembly. Calling C a high level language with Java, C# and Ruby around is like calling a 4 story building tall when theres a 400 story skyscraper next to it.
.net programs require the .net framework to be installed to be used. Its my understanding that one of parrot's goals is to be able to create standalone OS binaries that don't need a giant environment be installed to run. This may seem like a small issue, temporary and easy to fix, but its not.
.net 4.2 includes this great new feature that will make your little tool that you want to run on the 4000 machines you administrate that much better but they all have 3.5 installed. Unfortunately.. 4.2 has a bug that breaks an important application everyone uses. Even more unfortunately, you don't figure that out until you install it on those 4000 machines so you can run your utility. You're fired.
.net limitation which many people dismiss as unimportant but deeply limits the utility of the languages in creating code in wide-spread use on a machine. A library that is used by every application on the system (networking library, registry type library, etc) that is created in a way that its memory is unsharable is extremely inefficient in comparison to one that is sharable. With java (and I believe .net) a fundamental design principal is that each process's memory is completely isolated from all other processes (for the components written in .net and java). This leads to deep inefficiencies in memory use but more importantly in cache use and probably accounts for some of the "molasses" feeling of those environments.
Say, for example,
The issue I brought up about shared memory for shared libraries is another
Yes.. but.. (from what I understand) unlike .net and java, you will be able to compile binary versions of your applications for distribution that don't rely on a giant, bloated, pre-installed runtime *and* they aren't compiled to bytecode at program launch. Also, the register based parrot engine underneath is supposed to be much faster than the purely stack based .net and java VMs.
.net can't use this since their libraries aren't implemented on a level where that is possible. Parrot probably could but I'm not sure if it does. Does anyone know the answer to that?
Binaries that launch and run at the speed of C with the ability to write in a high level language? Sounds like the future of programming to me.
The big question I have is will multiple programs be able to share memory to store a commonly used library like C can. One of the things I think causes big inefficiencies in modern languages is the way they don't share memory well. If I have a C library that has, say, 5 megs of pure code in it and I'm running 10 copies of a program that uses it, all 10 can use the same 5 meg copy with only parts that are written to localized to the application. With high level languages, I need to use 50 megs of memory for each. Can Parrot share like C does and have all copies of the program reference the one copy stored in memory? Java and
Think about it for a second... If a highway has a carrying capacity of 400 cars per minute and you try to push 500 cars per minute through it how many cars per minute will end up sitting and waiting somewhere?
People seem deeply perplexed by this problem of "who is responsible" for them having to sit in traffic and for some reason always expect it to be some drivers fault. Unless the traffic jam you are in lasts about 10 seconds, or there is a real accident, then odds are the trigger event for the slowdown is irrelevant. The slowdown was inevitable.
Your road isn't big enough. It's operating at near or over saturation. Any road that is being pushed over its saturation point will buckle and traffic will stop, plain and simple. The trigger for the slowdown can be a person hitting the breaks, a momentary breakdown, even someone deciding they need to play it safe and leave a little more room between them and the next car can start the wave that ends up in a traffic jam. But the question isn't how the wave starts.. that question is meaningless. The question is why doesn't it dissipate like all the other waves and that answer is saturation. If the road wasn't saturated, the wave that got started when the guy hit his breaks too hard would have lasted a few cars and then dissipated as people worked around it and space between cars was compressed. Traffic would have absorbed the shock and kept moving.
Instead of tinkering with exotic breaking systems why not simply build more and bigger roads to accommodate the traffic. Better yet.. make tax incentives for companies who have employees work from home so less people are actually on the road.
Since Wi-Fi is generally 11-54 mbit they're only talking 1-5 gbit. The mentioned use is for video so it sounds like they are trying to connect displays to devices that generate output, i.e. replacing a monitor cable. For comparison DVI is 3.7 gbit, DVI-D 7.4. Most likely they are talking about the 1-2 gbit range since if it was in the 5gbit range they'd probably have said so instead of 100x wifi. That data rate would only be useful for low-resolution displays like HDTVs, not for general purpose computer monitor use. The devices would likely need to be close to each other due to the high frequencies. It sounds like they may be targeting removing the cable requirements home theater systems or something similar.
Personally.. I like cables for hooking up video. Wireless is buggy, snoopable, power hungry, and hard to set up (with 4 transmitters and 4 receivers, how to you configure what displays where?) Cables, while bulky and sometimes annoying have an incredibly easy UI. Plug one end here, the other end there, the things are connected. Want to change it? plug the wire in somewhere else.
Linux needs to be able to dislodge Microsoft Office in order to become a viable option for large-scale replacement of Windows in the corporate environment. Open Office can handle most of the document creation work and Firefox/Thunderbird do a great job at handling web browsing and email.
What is needed now is something that can integrate with an Exchange server's calendaring and also integrate with a robust open-source calendering server system to replace Outlook's calendaring functionality with an open, standards based system. We need to embrace and extend Outlook.
I've been saying it for 7 years now... 2010 is the year. This patent thing may put it off a few years but I doubt it. Every year Linux gets better and better, closer and closer.
Of the 235 probably 150 will be trivial to write around or just be thrown out as obvious.. most of the rest will take significant time and effort to work around but that work will eventually be done. I suspect somewhere around 10 will cause features to be removed from Linux distributions but they'll be features we can live without. During the process the patent system will hopefully be reigned in.
Any way this ends up it will be an interesting and ultimately healthy fight for FOSS.
Schools are government organizations and if you are a child, the schools rules are forced upon you. A government organization removing people's free speech rights is a horrible idea and quite correctly prohibited by the constitution.
Administrators like to think of the students like stores think of customers... here's how things are, take it or leave it. The problem is, there is no "leave it". Truancy laws say you have to go somewhere and essentially no children have the financial means to chose to go somewhere else.
If you take away a kids right to free speech during school hours you haven't really taken it away, you've just pushed it to non-school hours. This seems reasonable. If you take it away at home, however, you've removed it completely.
Repression of free speech is unconstitutional because it is a stupid idea. Inevitably it leads to the opposite of what it intends. Bad thoughts and ideas fester in private and die when the light of many eyes are shown upon them.
Just because something affects a school doesn't mean school administrators have control over it. They have tons of tools at their disposal, ones that will actually work. Try those. Get creative.
The metric system is better for people doing calculations.
The imperial system is better for people who cook, fix, build and actually do things.
I'm tired of scientists whining about how everybody doesn't make their math easier, especially when computers do 99% of their math for them anyway.
The English units are intuitive. 1 cup, 1 pint, 1 teaspoon are designed around quantities that are easy to visualize. 350 milliliters is designed for easy math but it's hard to visualize 350 1 milliliter things.
Also, in the English measurement system you find systems divisible by 3 and 4 which are very handy for building things. Metric is divisible by 2, 5 and 10 which is useful for doing math.
The metric system was designed before computers to make certain math in science easier. That goal is outdated. Metric units are inferior now.
The question as to what takes a road a saturation and pushes it over the edge is meaningless. It can be anything. Someone trying to go too fast, someone hitting the breaks a bit too hard, someone changing lanes.. any little blip can start things off. As many people have noted, once the road is at saturation, eventually something will push it over the limit.
The real question here is "How do we stop traffic jams?"
The answer is simple: don't saturate the road. Trying to avoid traffic jams on saturated roads is like trying to dodge raindrops.
A road below saturation has room enough between cars to handle little impulses. The impact of someone slowing traffic will be absorbed and dampened by the extra space instead of amplified. Roads that are built with excess capacity do this naturally so one way to achieve this is to simply make roads bigger. Most roads are built so far below the load that would naturally be put on them though that it would require a massive (2-3x) increase in their capacity to make this work.
A similar solution is to simply have more highways. Right now they only account for about 2% of the roads in the US and handle about 25% of the traffic. Taking land to build roads is always controversial but every second you sit in a traffic jam should bring you one second closer to understanding why it has to happen.
Other methods would include things like blocking access to the highway when it is near capacity. There have been failed attempts at this in California but they weren't designed to keep the roads under the saturation point so they were doomed to fail. A correct solution would have to outright deny access to people if the road is near saturation until it dropped below. This solution would probably be more annoying than the problem though.
I had a similar problem with drive caching swapping out all my running programs. I have a linux box where I back up one drive to another each night at 3am. The machine has 2 gigs of RAM but when I came in the next day it had swapped out all my programs to use it for disk cache and never swapped them back in. Every time I clicked anything in a program theres was nasty delay as it loaded that functionality off of disk. The UI was horribly sluggish for a while and I kept finding new things that were swapped out as the day went on. It was very frustrating.
I ended up adding a new cron job at 6am that turns off swap space and then turns it back on. Everything is back in real RAM by the time I get to the machine.
I keep seeing these statistics thrown around, but they just don't ring true.
There's a reason!
The thing they are looking at is "assets" calculated by how much you have - how much you owe. Under this model a very wealthy couple who together make $300,000 per year, own a million dollar home, drive $70,000 cars, have a nanny, a maid, take expensive vacations, buy anything they could possibly want and are by most measures very wealthy could be considered having nothing. The fictional couple above, if they are young or not saving properly for retirement, may have debt exceeding their assets in which case they would be considered as having no net worth at all. To this study they would be "poor" which they are most definitely not.
The problem with using net assets as a measure of wealth is people use these numbers in discussions of societal fairness and equity. These numbers are completely meaningless in that context. The only thing its really appropriate for doing is discussing how well people are saving for retirement because that's the primary reason to have money pooling in the way that would show up here. People use these numbers to point out economic inequity and imply things about standard of living which are completely inappropriate.
someone please add some tags to this article:
lies, damnlies, statistics
Most of WoW is fun and not bottable. People who think it's bottable generally don't understand that all the interesting difficult game content happens after you hit the level cap.
Blizzard has historically done a great job at catching the bad guys without catching innocent people. They probably caught some people who didn't deserve it but they were generally fair. This latest round wasn't like that at all. It would appear that they are currently targeting wow-glider and anti-afk macros so aggressively, they are using unreliable tactics prone to false positives.
I've been paying attention to this latest wave of bannings because my guild's main tank got hit and he has no idea why. I've talked to him at length and tried to figure out possible triggers. Nothing we've come up with yet would explain things. The most likely candidates include logging in from friends machines but he's talked to them and nobody seems to have installed any of the types of cheats that I know about.
Right now Blizzard's justice system is operating in a black box. They hide behind email auto-responders that claim to have "reinvestigated" their "extensive in-game logs" and technicians who handle appeals but don't seem to be given any details beyond "cheat detected". We don't know what their evidence is. We have to trust that when they ban it's legitimate.
Legally, Blizzard hides behind WoW's terms of service that says basically that they can yank your account anytime they feel like it with or without a valid reason. This is only part of the agreement between Blizzard and us though. Before investing hundreds and thousands of hours into characters that they can rip away at any time we expect them to be fair about doing so. We expect them to have a good reason and reasonable evidence and until now, they have upheld that.
In this latest round of bannings Blizzard banned unfairly. They banned innocent people based on flimsy faulty evidence and called them cheaters and lairs when they complained. The evidence that this was happening was there before this announcement but this is irrefutable proof. Unbanning the people who got banned for using Linux is a good start but what about all the other people who were innocent and got banned? Are we to believe that these are the only unfair bannings? The only mistake made happened to be in a tight active community with commercial representatives? Doubtful. I want to see Blizzard chow down on a big piece of humble pie this Thanksgiving and start actually looking at the evidence against people and un-ban the innocent victims of their latest dragnet banning scheme.
Blizzard usually does a great job at balancing protecting it's customers and protecting the gaming experience. This latest round of bannings stands in stark contrast to that. Someone, somewhere obviously got overzealous and decided to ban based on flimsy faulty evidence. I won't have any trust in the fairness of Blizzard's system until I see more people get un-banned.
Technology has made it increasingly easy to avoid interacting with other people. When was the last time you sat on your porch and chatted with a neighbor?
Like it or not technology is connecting people more, not less. We connect in different ways but those ways are easier, safer, more convenient and often more fun. How much porch sitting do you really think has been lost? Does it really compare with 10s of millions of people hanging out, having fun, helping each other and making friends in MMORPGs? How about the young couple with a baby at home who can't so much as catch a movie without hours of prep followed by worrying about what the person watching their kid is doing? For me, the kid goes to sleep at night and I start raiding.
Spending 4 hours with friends playing a game, joking around, playing music, making fun of each other, etc. It could be softball or WoW. More and more it's online games though since they are dramatically more accessible. I agree, there is something about real face to face friends lacking in online friends but there are upsides too.
Gamers may be divorced from the reality that OReilly cares about, but OReilly is also divorced from the reality that gamers are tuned into. Its intellectually lazy to whine about the destruction of what we had and ignore the value of what is being created in it's place.
If you haven't watched it yet, check out Hacking Democracy.
I like the idea of electronic voting. In the end it will be more secure than paper when it is done right but as it is now, security and tamper resistance haven't even been made a priority. If these things had the kind of security that you find in an X-Box it would be an enormous improvement.
They argue essentially "We cant tell you how the system works because if you knew that the system would be compromised." The problem is that some people *do* know how it works and other people can and have figured it out. Therefor The system is already compromised.
A system done right will be able to be watched very carefully. It will have an audit trail, it will find problems and when inspected carefully, catch people attempting to commit voter fraud. Any system that doesn't allow you to double check it is not tamper evident and shouldn't be used. Any system that comes out perfectly and finds no issues is probably not tamper evident.