Not all employees. Just those "who knowingly support an illegal action". If Suzy the receptionist, Bill the janitor and Jake the help-desk guy know nothing about the illegal actions they are clear.
What if they don't know said actions are illegal ? It's highly unlikely any employee under the level of VP or so has a good enough understanding of corporate law to know whether or not something is illegal at a given point in time (and more than arguable that they should not need to - after all, knowing such things certainly isn't part of their job description).
It shouldn't be limmited to upper management, it should be for ANYONE in the company.
Because clearly Suzy the receptionist, Bill the janitor and Jake the help-desk guy have not only in-depth knowledge of, but extensive influence over, the decision making process of executive management and therefore deserve to share their fate.
I suppose you think when people are executed any relatives they have closer than a second cousin should go down with them as well ?
Microsoft, on the other hand, had near complete control on the operating system market up until the resurgence of Macs in the last few years, though their control is still very strong, and it's still nearly complete in the corporate market.
At which point in the last, say, two decades has there not been at least one functionally equivalent alternative to the Windows platform ?
Incorrect. Apple is not a monopoly (in no small part because you can't be a monopoly until a court rules you as one).
ABUSING your monopoly position, THAT is illegal. Read the linked wikipedia article, they go into length describing this. It's amazing how many people think monopolies are illegal...
That's because it's basically impossible to be a monopoly and _not_ be "abusive", since a) you can be a monopoly without actually knowing about it and b) normal and typical business practices (eg: special pricing schemes and sole vendor agreements) suddenly become "illegal".
The weird thing here is that this is actually a case of a patent working the way that patents are supposed to work:
I thought patents were supposed to protect implementations, not ideas ? How is Microsoft developing their own implementation of the same idea an example of "a patent working the way that patents are supposed to work" ?
One reason might be, because if it's anything like the typical American Resume I see (one, maybe two pages of very high-level summary), it's unlikely to have any genuinely useful information on it _anyway_.
One thing employers should do more often is talk about salary up-front, so prospective employees don't waste a lot of time with testing and interviewing just to get an insulting low-ball offer.
But that would be impossible. After all, how is the employer supposed to know what the lowest offer you'll accept is ?:)
Why would I want to get to it from anywhere? When I'm away from my desk, it means I am FREE from my work! I just don't understand people who want to be grunting slaves 24/7. If you've got an excuse to not be working, then take it.
Crazy talk, I know, but some people use computers for personal things as well as work.
You do if you want to compete with what "the cloud" is offering.
I can have the phone autojoin my home wifi and use KIO Slaves on my desktop. On the road, I can run the programs on the phone. When I need more power, I can use X11 to remote home.
Which means you need to maintain a working environment "at home" you can get back to. That's the point of "the cloud" - so normal people don't need to do that.
Make a pretty UI and it's good enough for normal people.KIOSlave ? Rsync ? You're already well outside the bounds of "normal people".
Replying to myself, I'd just like to specify I'm against cloud computing. I'm saying I'm going to have the data on the phone, not access web services with it.
Properly synchronising data between multiple devices is a massive PITA.
On Windows you need special tricks, because Windows doesn't like to be installed there; it works on some netbooks with "special" BIOS. I think the specialness is at least partly from their EFIness but I'm just kind of firing in the dark here. I have a 4G Surf and an Aspire One, both will allegedly play this trick. Actually, I have a DT Research DT366 which seems to have some sort of USB disk emulation mode also.
Basically, it just needs to appear to the OS as a "fixed disk" rather than a "removable disk".
I agree, but why would Intel want to use flash memory for this?
Because, according to TFA, it's 1/4 the price of DRAM.
This is essentially a cache, which means it's going to get a lot of reads and writes. Under those circumstances, the flash memory's going to wear out relatively quickly and unless it's easily replaceable it means everyone's going to need to buy new motherboards every year.The SLC flash they're talking about will almost certainly last longer than the hard drives it is caching.
Ergo, it must be bad idea, there's obviously something wrong with it, everything is fine, they made the right call and would do so again tomorrow.
When making the comparison they are (their system to Dell, EMC, NetApp, et al) pretty much _everything_ is wrong with it - the design is riddled with catastrophe-multiplying SPOFs and performance would be dismal by every relative measure.
For their unique purposes, it might work well (although I have my doubts for the long term), but the systems they are comparing themselves to are just so much better at what they do it's not even funny.
I don't think RAID6 is going to be all that slow across 15 SATA drives. Bonus if the controller has acceleration to speed up parity calculations for faster writes.
The "overhead" of parity calculations is not a bottleneck in any remotely modern (<10 years old) system.
The CPU in the average budget _laptop_ machine of today calculates parity multiple times faster than the ASICs on even high-end RAID controllers.
If we lowball cheap drives at 300,000 hours MTBF, he'll see an average of two failures per month. It might take him $200 and an hour to recover each failed drive.
Personally, I'd be _extremely_ interested in seeing those numbers, because from my back-of-the-envelope calculations, it would take anywhere from 2-4 weeks to rebuild the arrays in each node (ie: in case of disk failure, or as part of initial commissioning), and on the order of 1-2 weeks to replicate all the data on a single node (in case they decided to write off an entire node).
My gut feeling, based on the above, is that the only thing keeping these guys from a catastrophic data loss event is luck (or they've already had some, but the customers weren't high profile enough to matter). Either that or each pod's worth of data costs a hell of a lot more than ~$180k to store (the power and cooling costs alone would be a small fortune).
Slashdot Mirror
Not all employees. Just those "who knowingly support an illegal action". If Suzy the receptionist, Bill the janitor and Jake the help-desk guy know nothing about the illegal actions they are clear.
What if they don't know said actions are illegal ? It's highly unlikely any employee under the level of VP or so has a good enough understanding of corporate law to know whether or not something is illegal at a given point in time (and more than arguable that they should not need to - after all, knowing such things certainly isn't part of their job description).
It shouldn't be limmited to upper management, it should be for ANYONE in the company.
Because clearly Suzy the receptionist, Bill the janitor and Jake the help-desk guy have not only in-depth knowledge of, but extensive influence over, the decision making process of executive management and therefore deserve to share their fate.
I suppose you think when people are executed any relatives they have closer than a second cousin should go down with them as well ?
Microsoft, on the other hand, had near complete control on the operating system market up until the resurgence of Macs in the last few years, though their control is still very strong, and it's still nearly complete in the corporate market.
At which point in the last, say, two decades has there not been at least one functionally equivalent alternative to the Windows platform ?
CORRECT!
Incorrect. Apple is not a monopoly (in no small part because you can't be a monopoly until a court rules you as one).
ABUSING your monopoly position, THAT is illegal. Read the linked wikipedia article, they go into length describing this. It's amazing how many people think monopolies are illegal...
That's because it's basically impossible to be a monopoly and _not_ be "abusive", since a) you can be a monopoly without actually knowing about it and b) normal and typical business practices (eg: special pricing schemes and sole vendor agreements) suddenly become "illegal".
And you're surprised that people don't understand the idea of a well-designed, single-function device?
They understand it quite well, they just find more value in only having to carry around one device instead of a dozen.
The weird thing here is that this is actually a case of a patent working the way that patents are supposed to work:
I thought patents were supposed to protect implementations, not ideas ? How is Microsoft developing their own implementation of the same idea an example of "a patent working the way that patents are supposed to work" ?
If you can't read a resume why did I write it?
One reason might be, because if it's anything like the typical American Resume I see (one, maybe two pages of very high-level summary), it's unlikely to have any genuinely useful information on it _anyway_.
One thing employers should do more often is talk about salary up-front, so prospective employees don't waste a lot of time with testing and interviewing just to get an insulting low-ball offer.
But that would be impossible. After all, how is the employer supposed to know what the lowest offer you'll accept is ? :)
You want me to work during the interview? You can pay me.
How is taking a test during the interview any more "work" than every other part of the interview ?
Why would I want to get to it from anywhere? When I'm away from my desk, it means I am FREE from my work! I just don't understand people who want to be grunting slaves 24/7. If you've got an excuse to not be working, then take it.
Crazy talk, I know, but some people use computers for personal things as well as work.
But that's just it, you don't need to sync.
You do if you want to compete with what "the cloud" is offering.
I can have the phone autojoin my home wifi and use KIO Slaves on my desktop. On the road, I can run the programs on the phone. When I need more power, I can use X11 to remote home.
Which means you need to maintain a working environment "at home" you can get back to. That's the point of "the cloud" - so normal people don't need to do that.
Make a pretty UI and it's good enough for normal people.KIOSlave ? Rsync ? You're already well outside the bounds of "normal people".
Replying to myself, I'd just like to specify I'm against cloud computing. I'm saying I'm going to have the data on the phone, not access web services with it.
Properly synchronising data between multiple devices is a massive PITA.
Please show me evidence of this. I have my doubts health organizations would be saying BMI>25 (or even 23 in Japan) is bad if it were not true.
Being fit and eating a sufficient range of food is _vastly_ more important than whether or not your BMI is within the magic range.
I'm assuming it is essentially paying a data center to host my data from my home system? Why in the hell would I even WANT to do that?
Because then it's trivially simple for you (more importantly, for people who aren't at all technologically inclined) to get at it from anywhere.
Is a 4GB buffer really capable of successfully buffering all that data?
Run some benchmarks before and after disabling the 16-32MB cache on your hard disk. You might be surprised.
In what way would it even compete with the SSD market?
When it offers equivalent benefits for a fraction of the cost.
On Windows you need special tricks, because Windows doesn't like to be installed there; it works on some netbooks with "special" BIOS. I think the specialness is at least partly from their EFIness but I'm just kind of firing in the dark here. I have a 4G Surf and an Aspire One, both will allegedly play this trick. Actually, I have a DT Research DT366 which seems to have some sort of USB disk emulation mode also.
Basically, it just needs to appear to the OS as a "fixed disk" rather than a "removable disk".
Oh there's a bunch of them.
So why is an additional cache between memory and disk "redundant" ?
I don't need any more, certainly.
Why not ?
Random I/O is essentially uncacheable.
I think you mean "unpredictable" here, not "random".
I agree, but why would Intel want to use flash memory for this?
Because, according to TFA, it's 1/4 the price of DRAM.
This is essentially a cache, which means it's going to get a lot of reads and writes. Under those circumstances, the flash memory's going to wear out relatively quickly and unless it's easily replaceable it means everyone's going to need to buy new motherboards every year.The SLC flash they're talking about will almost certainly last longer than the hard drives it is caching.
Exactly. I already have a disk cache. This solution is redundant.
How many caches are there between your CPU and RAM ?
Ergo, it must be bad idea, there's obviously something wrong with it, everything is fine, they made the right call and would do so again tomorrow.
When making the comparison they are (their system to Dell, EMC, NetApp, et al) pretty much _everything_ is wrong with it - the design is riddled with catastrophe-multiplying SPOFs and performance would be dismal by every relative measure.
For their unique purposes, it might work well (although I have my doubts for the long term), but the systems they are comparing themselves to are just so much better at what they do it's not even funny.
How long does it take to rebuild the array if a disk has to be replaced?
By my calculations, best case scenario is about ten days for the arrays connected to PCIe controllers and 25 days for the array on the PCI controller.
I don't think RAID6 is going to be all that slow across 15 SATA drives. Bonus if the controller has acceleration to speed up parity calculations for faster writes.
The "overhead" of parity calculations is not a bottleneck in any remotely modern (<10 years old) system.
The CPU in the average budget _laptop_ machine of today calculates parity multiple times faster than the ASICs on even high-end RAID controllers.
If we lowball cheap drives at 300,000 hours MTBF, he'll see an average of two failures per month. It might take him $200 and an hour to recover each failed drive.
Personally, I'd be _extremely_ interested in seeing those numbers, because from my back-of-the-envelope calculations, it would take anywhere from 2-4 weeks to rebuild the arrays in each node (ie: in case of disk failure, or as part of initial commissioning), and on the order of 1-2 weeks to replicate all the data on a single node (in case they decided to write off an entire node).
My gut feeling, based on the above, is that the only thing keeping these guys from a catastrophic data loss event is luck (or they've already had some, but the customers weren't high profile enough to matter). Either that or each pod's worth of data costs a hell of a lot more than ~$180k to store (the power and cooling costs alone would be a small fortune).