"Assembly" is a proper noun, specifying a specific language.
Actually, it doesn't specify "a specific language" - it refers to a class of languages. I'm still fairly fluent in maybe half a dozen different assembly langauges, and could probably get by in another half dozen or so if I had to. (And no, I'm not counting different syntax for the same architecture, I'm actually talking about different target platforms)
i believe a more appropriate measure of the 'bloat' (i.e. useless functions) or the size of any software package is through function point analysis
I recall many years ago, a PHB (this is long enough ago that nobody called them that yet) was talking about developer productivity metrics; he announced that the powers that be were considering either KLoC or Function Points. The guy sitting next to me said "I have no idea what function points are, but they've got to be better than KLoC". The remark made one of those wonderful whooshing sounds as it sailed straight over the PHB's head...
LOC is without question one of the easiest measurement (aside from total package size in bytes, which is nearly as uninformative)
+1 - Fundamental Law Of Physics. LoC's only redeeming feature as a metric of anything is that it's (relatively) easy to measure. Of course, there's the debate about "do we count comments", "do we count whitespace", "how do we count curly braces" - so it turns out it's actually NOT all that easy to measure. But don't let a PHB hear you speaking such heresy...
But since you've descended to (well, more "started with" rather than actually "descended to") ad hominem attacks, it's perfectly clear that there's no point in trying to have a reasonable discourse with you.
Maybe you missed the bit where I said "arrays of equal sizes", thus causing you to spout bullshit?
It smells to me like you're talking about a 4-way mirror; I don't see how a 4-way mirror gives you the same storage ("equal size") as a 4-drive RAID6 configuration. (and, as has been pointed out elsewhere, I don't think anything actually supports 4-way mirrors)
If you've got 4x1TB drives, you can have a 2TB RAID1 config, or a 2TB RAID6 config. The RAID1 config will be faster, sure - but in the face of 2 drive failures, there's a good chance you've lost half your data (and if your file system spans the whole array, that could potentially mean losing all of it). With RAID6, you'll still have it all there.
Given that RAID1 is pretty much for two disks only
No, RAID1 will work with any even number of disks. If you have an array with 4 drives in, you've got 2 pairs of mirrors. In the dual-failure case, you have two scenarios: 1) 2 "paired" drives fail. The data on that half is gone, but the other half is still mirrored, so you've only lost half your data. 2) 2 "unpaired" drives fail - you survived, you just don't have mirrors any more.
I fail to see how any form of RAID will prevent data loss when you lose two disks.
Broadly speaking, it's down to having better data distribution. You've got an extra set of parity data to rebuild the array from. Wikipedia is your friend. As the number of drives in the array increases, RAID6 gets even more attractive than RAID1 (From an cost & data integrity POV, at least - but RAID1 will generally give you better performance)
Of course, it's theoretically possible to treat half or a RAID1 pair as a "hot backup" - just pull it out, plug in a fresh drive, and cart it off to use as your off-site. I wouldn't recommend that as your primary backup strategy though...
What would you recommend then for a 30user non-profit business with a 136gb scsi drive and tape backups they have been running (but not realizing that they weren't completely backing up everything)? Thats the crap I just inherited...I was thinking moving to 2 sata drives in raid 1(the scsi drive sounds like it's gonna die soon) and fixing the bloated folders so that tape backups will work again. Differential appended tape backups M-R and a normal backup on friday after the workday seems like the best idea.
I'd probably go along with your Raid 1 solution as the primary storage.
For the backups, I'd probably forget tapes altogether - I'd go for a couple of cheap-ish, easily portable NAS enclosures. Run your proposed backup cycle (i.e. weekly full backup, dcaily incrementals) direct onto one of them overnight, then during the day mirror that onto the other one. Somebody takes the second unit home at night as an off-site. (Obviously, three or more enclosures would be preferable, since it means that at least one of your off-sites is always actually off-site!)
Every so often (maybe monthly), back up the NAS to DVD as a belt-and-braces measure. It all kind of depends on how important the data is, and how much they can afford to spend.
The chances that your RAID will have a double failure causing your data to be lost are just about the same as the chances that your RAID will have a single failure and your tape backup also has a failure.
Only if you assume a single backup tape. Anyone who's even half-way serious about keeping their data intact will have multiple backup tapes. (Once upon a time we called it "grandfather-father-son" - but these days we tend to go for more generations than that). For our critical systems, we have multiple copies of *every* backup (two in the data centre for ease of restoration, and a minimum of four offsite - 2 each in 2 locations). Even on databases that are replicated globally.
You can back up your RAID, but that'd be like backing up your backup tapes.
Exactly like backing up your backup tapes - an eminently sensible strategy if your data means anything to you.
I'm a little less fastidious about it when it comes to backing up my data at home - but every time I burn a backup DVD, I burn at least two copies of it. Once bitten, twice shy.
Oh come on. Do you have 12TB of home data? Seriously? And if you do, it's not that hard to have another another 12TB of external USB drives at some relatives place.
Not all of us have relatives, you insensitive...[URE]
RAID 5, as well as RAID 6 is nothing more at an attempt to add some amount of redundancy without sacrificing too much space. Go RAID 1 instead with the same number of disks.
As far as I'm concerned, RAID 5 really has no redeeming features (it's slow, not particularly safe, but lulls people into a false sense of security).
From a data integrity perspective, though, RAID6 is a better solution than RAID1.
Given arrays of equal sizes, with RAID6 your data can survive the loss of *any* two disks; with RAID1, if you lose two disks which happen to be a mirrored pair, then you're hosed.
But, as you point out, RAIDn doesn't really qualify as "carefully protected"
> If you read some of the documents regarding the NSA and AES it doesn't look like it would be that hard for them to do it.
I've read the article you've linked to - and I don't see *ANY* reference to AES in there (I see a reference to dual EC generators, but that's not something you could describe as a cornerstone of AES). If there were a problem with AES itself, I'd expect a cryptographer of Schneier's standing to be shouting it from the rooftops.
If you know anything about security, then you'll know that one of the problems with putting a backdoor in, is dealing with the consequences when somebody else finds out how to use it. If you're even half-way serious about security, you assume that somebody WILL find out how to use it. Remember that the US Government certifies 256-bit AES for use on information classified "top secret", which implies to me that they think it's reasonably secure.
The problem Bruce describes with the NSA-specified EC generator, on the other hand, is probably real (I haven't studied the maths behind it, but I respect him enough as a cryptographer and mathematician to take him at his word). But saying something is "possible" is a very different proposition from saying something "would not be hard to do".
Throw that in and you have $1.5 billion, no problem. Lehman's don't own the building on Bank Street, they lease(d) it from Songbird. So it's not actually part of the deal.
Slashdot Mirror
"Assembly" is a proper noun, specifying a specific language.
Actually, it doesn't specify "a specific language" - it refers to a class of languages.
I'm still fairly fluent in maybe half a dozen different assembly langauges, and could probably get by in another half dozen or so if I had to. (And no, I'm not counting different syntax for the same architecture, I'm actually talking about different target platforms)
i believe a more appropriate measure of the 'bloat' (i.e. useless functions) or the size of any software package is through function point analysis
I recall many years ago, a PHB (this is long enough ago that nobody called them that yet) was talking about developer productivity metrics; he announced that the powers that be were considering either KLoC or Function Points. The guy sitting next to me said "I have no idea what function points are, but they've got to be better than KLoC". The remark made one of those wonderful whooshing sounds as it sailed straight over the PHB's head...
LOC is without question one of the easiest measurement (aside from total package size in bytes, which is nearly as uninformative)
+1 - Fundamental Law Of Physics.
LoC's only redeeming feature as a metric of anything is that it's (relatively) easy to measure. Of course, there's the debate about "do we count comments", "do we count whitespace", "how do we count curly braces" - so it turns out it's actually NOT all that easy to measure. But don't let a PHB hear you speaking such heresy...
And you don't even have to change the lettering!
Not even if you measure it in Killer Libraries Of Congress...
You could try:
DIVIDE SLOC BY 1000 GIVING KLOC.
*sigh* Arithmetic really isn't your strong point, is it?
Drive 0 mirrors Drive 1 = 1 TB
Drive 2 mirrors Drive 3 = 1 TB
1+1=2, so a total of 2TB. Really, look it up
But since you've descended to (well, more "started with" rather than actually "descended to") ad hominem attacks, it's perfectly clear that there's no point in trying to have a reasonable discourse with you.
This is what you're looking for.
Maybe you missed the bit where I said "arrays of equal sizes", thus causing you to spout bullshit?
It smells to me like you're talking about a 4-way mirror; I don't see how a 4-way mirror gives you the same storage ("equal size") as a 4-drive RAID6 configuration. (and, as has been pointed out elsewhere, I don't think anything actually supports 4-way mirrors)
If you've got 4x1TB drives, you can have a 2TB RAID1 config, or a 2TB RAID6 config. The RAID1 config will be faster, sure - but in the face of 2 drive failures, there's a good chance you've lost half your data (and if your file system spans the whole array, that could potentially mean losing all of it). With RAID6, you'll still have it all there.
Given that RAID1 is pretty much for two disks only
No, RAID1 will work with any even number of disks. If you have an array with 4 drives in, you've got 2 pairs of mirrors. In the dual-failure case, you have two scenarios:
1) 2 "paired" drives fail. The data on that half is gone, but the other half is still mirrored, so you've only lost half your data.
2) 2 "unpaired" drives fail - you survived, you just don't have mirrors any more.
I fail to see how any form of RAID will prevent data loss when you lose two disks.
Broadly speaking, it's down to having better data distribution. You've got an extra set of parity data to rebuild the array from. Wikipedia is your friend. As the number of drives in the array increases, RAID6 gets even more attractive than RAID1 (From an cost & data integrity POV, at least - but RAID1 will generally give you better performance)
Of course, it's theoretically possible to treat half or a RAID1 pair as a "hot backup" - just pull it out, plug in a fresh drive, and cart it off to use as your off-site. I wouldn't recommend that as your primary backup strategy though...
What would you recommend then for a 30user non-profit business with a 136gb scsi drive and tape backups they have been running (but not realizing that they weren't completely backing up everything)? Thats the crap I just inherited...I was thinking moving to 2 sata drives in raid 1(the scsi drive sounds like it's gonna die soon) and fixing the bloated folders so that tape backups will work again. Differential appended tape backups M-R and a normal backup on friday after the workday seems like the best idea.
I'd probably go along with your Raid 1 solution as the primary storage.
For the backups, I'd probably forget tapes altogether - I'd go for a couple of cheap-ish, easily portable NAS enclosures. Run your proposed backup cycle (i.e. weekly full backup, dcaily incrementals) direct onto one of them overnight, then during the day mirror that onto the other one. Somebody takes the second unit home at night as an off-site. (Obviously, three or more enclosures would be preferable, since it means that at least one of your off-sites is always actually off-site!)
Every so often (maybe monthly), back up the NAS to DVD as a belt-and-braces measure. It all kind of depends on how important the data is, and how much they can afford to spend.
The chances that your RAID will have a double failure causing your data to be lost are just about the same as the chances that your RAID will have a single failure and your tape backup also has a failure.
Only if you assume a single backup tape. Anyone who's even half-way serious about keeping their data intact will have multiple backup tapes. (Once upon a time we called it "grandfather-father-son" - but these days we tend to go for more generations than that). For our critical systems, we have multiple copies of *every* backup (two in the data centre for ease of restoration, and a minimum of four offsite - 2 each in 2 locations). Even on databases that are replicated globally.
You can back up your RAID, but that'd be like backing up your backup tapes.
Exactly like backing up your backup tapes - an eminently sensible strategy if your data means anything to you.
I'm a little less fastidious about it when it comes to backing up my data at home - but every time I burn a backup DVD, I burn at least two copies of it. Once bitten, twice shy.
Oh come on. Do you have 12TB of home data? Seriously? And if you do, it's not that hard to have another another 12TB of external USB drives at some relatives place.
Not all of us have relatives, you insensitive...[URE]
RAID 5, as well as RAID 6 is nothing more at an attempt to add some amount of redundancy without sacrificing too much space. Go RAID 1 instead with the same number of disks.
As far as I'm concerned, RAID 5 really has no redeeming features (it's slow, not particularly safe, but lulls people into a false sense of security).
From a data integrity perspective, though, RAID6 is a better solution than RAID1.
Given arrays of equal sizes, with RAID6 your data can survive the loss of *any* two disks; with RAID1, if you lose two disks which happen to be a mirrored pair, then you're hosed.
But, as you point out, RAIDn doesn't really qualify as "carefully protected"
Hibernation has worked on every single Microsoft machine I have ever tried, laptop, desktop, XP, Vista, every single time I did it.
You've never had a Sony laptop, have you?
People make mistakes too.
To err is human. To really foul things up, you need a computer.
I fail to see the point of shaking it.
With a bit of luck, it'll go off in your hand - thus saving the flight... ;-)
> If you read some of the documents regarding the NSA and AES it doesn't look like it would be that hard for them to do it.
I've read the article you've linked to - and I don't see *ANY* reference to AES in there (I see a reference to dual EC generators, but that's not something you could describe as a cornerstone of AES). If there were a problem with AES itself, I'd expect a cryptographer of Schneier's standing to be shouting it from the rooftops.
If you know anything about security, then you'll know that one of the problems with putting a backdoor in, is dealing with the consequences when somebody else finds out how to use it. If you're even half-way serious about security, you assume that somebody WILL find out how to use it. Remember that the US Government certifies 256-bit AES for use on information classified "top secret", which implies to me that they think it's reasonably secure.
The problem Bruce describes with the NSA-specified EC generator, on the other hand, is probably real (I haven't studied the maths behind it, but I respect him enough as a cryptographer and mathematician to take him at his word). But saying something is "possible" is a very different proposition from saying something "would not be hard to do".
> Why is your encryption key any different from the safe/door you have?
It isn't. I'll just stand back and watch them break my 256-bit AES...
You think nobody's ever confessed to something they didn't do under torture?
I'd say a false confession qualifies as "making up sh*t"
Must remember to think in Russian when using the USSR version.
Don't you mean:
"Must remember to think in Russian when using FireFox?"
Throw that in and you have $1.5 billion, no problem.
Lehman's don't own the building on Bank Street, they lease(d) it from Songbird. So it's not actually part of the deal.
And when the master server that knows where all those little pieces are goes down, you are still without your data.
One would assume that this company is competent enough to use highly available, clustered systems to avoid a SPOF.
Umm, yeah. And, presumably, in multiple geographically disparate locations. But that still kind of misses the point of:
I have been looking for a way to securely store my files online without being tied to a single vendor
Doesn't matter how good their DR capability is, it's still a single vendor...
To be replace by "Logan 5" and "Jessica 6", no doubt ;)
...is Lucas. If he hadn't made the movie, then nobody would be able to download it.
(Does this stunning display of illogic qualify me for a multi-million$ job at the MPAA?)
> 8 years computer programming experience
Frikkin' newbie.
> iTunes still exist.
You can download a hour's worth of music in considerably less time than it takes to download a hour's worth of movie.