Apple setup in Ireland to appease an EU mandate regarding labor content in their products. This sort of thing was also done by the US to Japanese TV manufacturers. It's a time-honored way to get manufacturers to spend a little money in the country they're selling into-- to avoid getting hit with taxes and duties.
However, very little manufacturing goes on in the Caymans, the BWI, and Luxembourg. They're tax havens, just like Cheney's Halliburton likes to use. Getting back that revenue would mean the end of corporate welfare, and that's unlikely to happen soon.
So call a shovel a spade and don't weasel word about the implications.
It might be possible that your WiFi AP has support for Channel 14-- not found usually on US gear, unless you flash the AP with the International Version (hint). The second thing you can do is to get cheapo antennas to put on the AP (~$10 at a local Frys or eBay) to boost the output in a particular direction. My suggestion is to boost it in a direction away from adjacent neighbors.
And while 802.11a sounds good as it probably has little interference, the hardware is a bit expensive compared to 802.11b/g/n.... N is nice if you can find cheaper hardware and it's the first time that I finally put away the Ethernet cables and went truly wireless around my house. YMMV.
Which is also the fun of a 64-bit CPU: you can do huge amounts of math with a fat mantissa. FPUs don't even have to be accessed for a great deal of non-geometric math, thus speeding things up if you've got a good math lib with your compiler-- and the compiler was written well.
Spreadsheets in the early days even ran on Atari 800s..... Long Live SuperCalc!
It's not the operating system that pays, it's the sum of platform applications and infrastructure where Microsoft makes money, coupled to the students that know nothing else but Windows. The tip of the iceberg is Windows. The cash cows are Office, SQL Server, Exchange, and the add-ons, upgrades, and other platform products.
You don't 'lose' to Linux, you lose revenue that represents lots of infrastructure, server licenses, CALs, and so on.
There are few professional organizations that can do an end-to-end Linux infrastructure for educational needs (including school administration software costs) but the list is growing, if by populism alone.
Part of Microsoft's loss is the horrible security problems of 1998-2007, as they're less than before. That damage hurt Microsoft-- coupled to support costs for the products. Macs have always been a fractional part of the educational market, and Apple's done a lot to damage their own relationships with schools-- but students love them.
Microsoft has a lot to learn about love, rather than feigning leadership.
Seismic data is a bit different than transactional business data and I'm guessing it's subject to non-lossy codecs and other warehousing techniques that are just plain different.
I'll bet there's a way to compress that data and squish it into something that's more manageable.... especially if it's of a low sample rate. But I really don't know.....
My rationale: while no one's been looking, drive prices have (sorry to use this word) crashed. But drive reliability is getting really good. I don't see having to have two drives, one an enterprise 'branded' drive, another an 'OEM' drive, and still a third, a 'consumer' drive. You might pay for overnight service, but the quality differences aren't very much. A high speed SCSI or SAS or FC-modded drive is still damn fast in any one of the aforementioned categories and getting cheaper by the week. Tape drives are proprietary, slow, not randomly accessible, and require a lot of human intervention.
Throwing 5TB into a server mosh pit and crunching it until you get some meaning is a huge dataset, and unusual. Yet arrays can do this easily, and as they're arrays, often quickly. I'm guessing that a non-tape solution is viable, but I'm not experienced in such datasets and how they're processed.
USB write times are horrible. SSDs are getting faster, but there's a need to RAW that makes them slow, and the method that's used to write isn't like rotational media.
There are reasons to consider things like virtualization, FastScale, and other methods to cut down on the datasets. Then you need to have what's really needed in terms of audit, legal, and compliance needs. Then you factor reliability and availability.
The old maxim of online, near-line, and offline storage doesn't hold water these days. Random access and lack of human intervention and media management costs just push the fulcrum the other way now.
The place for tape technology is in a suitable recycling facility where it belongs. Go ahead: ask the hard questions to your vendors about what drives they use, the origin of them, and how they're different. Go ahead. Then do the checking on your own by pulling out some spares and finding out from Seagate, IBM, Hitachi, etc more info about those drives. Then find out how much the label cost, and how much that means to you.
You mistakenly believe that consumer grade drives have different MTBFs than OEM or 'enterprise' drives. What you're paying for is a vacation cruise for your sales rep.
I'm fully aware of differing needs for differing applications. I firmly believe all can be done with hard drives and even (in the future) SSD drives.
The availability systems I've engineered were through through, with audit and compliance and low MTTR in mine. It scales like crazy, and even more cost-effectively every month that hard drive costs go down-- and they do, every month since about 1988 in terms of cost per GB stored through a five year life cycle.
My decades of experience are unable to be exposed here. My years of battling the farce of overpriced, underperforming availability systems is the crux of my posts.
In 1982, your tape format was proprietary. Maybe, if you're lucky, the tape hasn't oxidized. Maybe, if you're lucky, you can still read the ink on the label.
Or, perhaps you put the archive into a new SAN cache when you copied over the data set and bindery.
There might be media in Wang format, or maybe Lanier, or a hard-sectored 8". 3740 format tapes can still be read, unbelievably.
But you're citing 1982, and now that the tape is 26yrs old, was found in some forsaken warehouse, and now how are you going to get at the data? Now that there's an Internet, and online storage is not only feasible but desirable, rethinking all of this takes a measure of disbelief coupled to courage. That's how SANs are working, virtualized infrastructure, and highly audit-compliant availability systems are constructed. Newspapers are also dying; can't wait for Liquid Paper to become the analog, just like I waited for disks to drop in price to make a practical alternative to the hideous nature of tape archiving.
We don't CLOAD anymore. We don't use punch cards. We got rid of punch-tape. Now let's get rid of tapes.
If you design a sane availability system, you can obviate tape and all the crap behind it. It's all about availability, audit, and a systematic approach to provisioning, archiving, and lifecycle.
The procedure isn't to connect and disconnect SAS or SATA cables, rather to pipe the data across network wires, in-band or out-of-band. Having a shadow network to perform backups, provision machines, re-slab servers and so on (especially VMs!) is the goal.
Tape media is fraught with difficulties, tho not as much as when we dealt with DC600s and 3740's. The cute little cartridges today are nice, and wonderfully fast, and still lack random accessibility, require physical manipulation (and are more subject to loss, theft, and environmental disasters), are proprietary to specific vendor solutions, and while more compatible with various OS blends than ever before, are still a hassle worth retiring, IMHO.
Oh-- do we ever agree about that. But who says you have the data in just one building? My snapshots and mirrors are 13mi from where I sit. The backups for those are nine states away. Another mirror of those is in Canada. Cost? Not that much, but my payloads aren't the size of other organizations.
And listen: there's not one freaking tape. Not one. And it's randomly accessible fully, and the MTTR for primary systems is under eight minutes from click to verified. You, too, can do it.
And have you noted the difference between "consumer grade" hard drives and others? Do you realize the madness in MTBF statistics used by the various hard drive vendors? Do you track the failure modes between OEM and "consumer grade" drives? Do your research and put your money where you believe. I do.
Yes, the silver bullet is to drive it right through the heart of your expensive heaping honking tape library. Maybe garlic will help, too.
You can data dupe (2x cost). But you can also just say no, and start the procedure that gets you alternate storage, randomly accessible, easily tested, highly portable, and without proprietary media and infrastructure for about the same price as drive+tape+wiretime+hotelcost+human_intervention costs.
This is neither immoral or silver bullet-mentality. This is real and now. The sooner we shake tape, the better off we are. I've spent toooooooooo many hours watching jukeboxes shuffle stuff around to find out a vendor screwed up archiving, blew their compression algorithm, had a crappy library transport, or whatever-- all proprietary. Days of my life have been spent watching these things; they should be recycled at the earliest opportunity for the madness that they are.
You're welcome to pay for some EMC's salesperson's golf trip if you want. Enterprise class equipment certainly costs a bit more, but you're often paying for luxury where utility will do nicely.
Seagate bulk SAS drives 1tb barracudas are about $220 in oem bulk.... less if you want ata/300s. Compressed snapshots are fine in my experience, and better still, just in a nice, easy to slab hypervisor mountable.... or any other format that fits the bill. Delta snapshots are fine, too, depending on the transactional nature of the data. If it's transactional, then there are lots of great techniques to have multiple concurrent data instances that are easily coalesced when needed. The trouble: remembering to do fire drills.
Fine. Get the cartridges, but what about the capital cost minus depreciation of the drive? What about random access?
Now weigh those against an inexpensive jbod frame with a 2gb FC backplane. What's the write speed of LT vs a tasty little GB SAS drive? Rackspace? You can put a dozen into about 4U. Cooling? Although I'll grant you green cost, the random accessibility out-classes the seek time and tape insertion by a human cost dramatically. Stable media? Tape? Sometimes. Shelf space?
SAN fabric is dirt these days. You can get a nice Silkworm and a cheap-but-reliable SAS backplane for dirt as well. Perhaps a couple of GBICs.... or some handy-dandy fiber cables (also dirt these days) and you're in business. Or, put up a 10dot network off your public-face grid, and just use iSCSI. No need to use tape anymore. Get out of the reality distortion field, but do the right thing by testing what you have and doing drills to ensure that whatever you have, works and is a procedure understood by all.
Mirrors are fine, just snapshot them and store them offsite regularly. Do delta backups as needed but close-in for fast restoration.
There is no rational justification for tape anymore, what with the cost per TB stored on hard disks now under $130, total $$. Random accessibility unless you're stalling a subpoena, is just mandatory on backup media.
I'm generally on the side of NPR and its content, except that it only rarely touches on the needs of minorities. There's a lot of WASP-ish content that's focused to YUPpie types, rather than trying to reach out to a larger community.
In terms of news content, there's a lot of belabored discourse, some of which is good, some beats a dead horse... perturbative.
Hey-- what is perfect? There are millions of answers because there are millions of people, all with different needs and skills sets.
Microsoft may give customers what they want, but they're insular and incestuous, and ride roughshod over anyone in their paths. Their convictions and settlements across the planet are a matter of fact. In giving customers what they want, they may have cornered the market illegally, or simply stolen what they gave to their customers.
Linux is only now becoming civilian friendly. Apple's done a reasonable job but portends as much proprietary hubris as Microsoft does. Apple gets great points for having dogmatic and high quality standards, but they're imperfect, too. They all are.
Were I to predict a final outcome, the 'Windows Franchise' becomes worth less and less, as Microsoft's infractructure slowly falls out of sync. Ask Oracle what happens then-- it nearly killed them.
Slashdot Mirror
Good to know.
That's all right, you may not have to worry if it's a 1TB Seagate drive. They self-brick.
Sadly, it's best just to physically destroy the drive after use. I suggest a two-year old child just after its nap ought to do the trick.
Apple setup in Ireland to appease an EU mandate regarding labor content in their products. This sort of thing was also done by the US to Japanese TV manufacturers. It's a time-honored way to get manufacturers to spend a little money in the country they're selling into-- to avoid getting hit with taxes and duties.
However, very little manufacturing goes on in the Caymans, the BWI, and Luxembourg. They're tax havens, just like Cheney's Halliburton likes to use. Getting back that revenue would mean the end of corporate welfare, and that's unlikely to happen soon.
So call a shovel a spade and don't weasel word about the implications.
It might be possible that your WiFi AP has support for Channel 14-- not found usually on US gear, unless you flash the AP with the International Version (hint). The second thing you can do is to get cheapo antennas to put on the AP (~$10 at a local Frys or eBay) to boost the output in a particular direction. My suggestion is to boost it in a direction away from adjacent neighbors.
And while 802.11a sounds good as it probably has little interference, the hardware is a bit expensive compared to 802.11b/g/n.... N is nice if you can find cheaper hardware and it's the first time that I finally put away the Ethernet cables and went truly wireless around my house. YMMV.
Which is also the fun of a 64-bit CPU: you can do huge amounts of math with a fat mantissa. FPUs don't even have to be accessed for a great deal of non-geometric math, thus speeding things up if you've got a good math lib with your compiler-- and the compiler was written well.
Spreadsheets in the early days even ran on Atari 800s..... Long Live SuperCalc!
It's not the operating system that pays, it's the sum of platform applications and infrastructure where Microsoft makes money, coupled to the students that know nothing else but Windows. The tip of the iceberg is Windows. The cash cows are Office, SQL Server, Exchange, and the add-ons, upgrades, and other platform products.
You don't 'lose' to Linux, you lose revenue that represents lots of infrastructure, server licenses, CALs, and so on.
There are few professional organizations that can do an end-to-end Linux infrastructure for educational needs (including school administration software costs) but the list is growing, if by populism alone.
Part of Microsoft's loss is the horrible security problems of 1998-2007, as they're less than before. That damage hurt Microsoft-- coupled to support costs for the products. Macs have always been a fractional part of the educational market, and Apple's done a lot to damage their own relationships with schools-- but students love them.
Microsoft has a lot to learn about love, rather than feigning leadership.
Seismic data is a bit different than transactional business data and I'm guessing it's subject to non-lossy codecs and other warehousing techniques that are just plain different.
I'll bet there's a way to compress that data and squish it into something that's more manageable.... especially if it's of a low sample rate. But I really don't know.....
My rationale: while no one's been looking, drive prices have (sorry to use this word) crashed. But drive reliability is getting really good. I don't see having to have two drives, one an enterprise 'branded' drive, another an 'OEM' drive, and still a third, a 'consumer' drive. You might pay for overnight service, but the quality differences aren't very much. A high speed SCSI or SAS or FC-modded drive is still damn fast in any one of the aforementioned categories and getting cheaper by the week. Tape drives are proprietary, slow, not randomly accessible, and require a lot of human intervention.
Throwing 5TB into a server mosh pit and crunching it until you get some meaning is a huge dataset, and unusual. Yet arrays can do this easily, and as they're arrays, often quickly. I'm guessing that a non-tape solution is viable, but I'm not experienced in such datasets and how they're processed.
USB write times are horrible. SSDs are getting faster, but there's a need to RAW that makes them slow, and the method that's used to write isn't like rotational media.
There are reasons to consider things like virtualization, FastScale, and other methods to cut down on the datasets. Then you need to have what's really needed in terms of audit, legal, and compliance needs. Then you factor reliability and availability.
The old maxim of online, near-line, and offline storage doesn't hold water these days. Random access and lack of human intervention and media management costs just push the fulcrum the other way now.
The place for tape technology is in a suitable recycling facility where it belongs. Go ahead: ask the hard questions to your vendors about what drives they use, the origin of them, and how they're different. Go ahead. Then do the checking on your own by pulling out some spares and finding out from Seagate, IBM, Hitachi, etc more info about those drives. Then find out how much the label cost, and how much that means to you.
You mistakenly believe that consumer grade drives have different MTBFs than OEM or 'enterprise' drives. What you're paying for is a vacation cruise for your sales rep.
I'm fully aware of differing needs for differing applications. I firmly believe all can be done with hard drives and even (in the future) SSD drives.
The availability systems I've engineered were through through, with audit and compliance and low MTTR in mine. It scales like crazy, and even more cost-effectively every month that hard drive costs go down-- and they do, every month since about 1988 in terms of cost per GB stored through a five year life cycle.
My decades of experience are unable to be exposed here. My years of battling the farce of overpriced, underperforming availability systems is the crux of my posts.
In 1982, your tape format was proprietary. Maybe, if you're lucky, the tape hasn't oxidized. Maybe, if you're lucky, you can still read the ink on the label.
Or, perhaps you put the archive into a new SAN cache when you copied over the data set and bindery.
There might be media in Wang format, or maybe Lanier, or a hard-sectored 8". 3740 format tapes can still be read, unbelievably.
But you're citing 1982, and now that the tape is 26yrs old, was found in some forsaken warehouse, and now how are you going to get at the data? Now that there's an Internet, and online storage is not only feasible but desirable, rethinking all of this takes a measure of disbelief coupled to courage. That's how SANs are working, virtualized infrastructure, and highly audit-compliant availability systems are constructed. Newspapers are also dying; can't wait for Liquid Paper to become the analog, just like I waited for disks to drop in price to make a practical alternative to the hideous nature of tape archiving.
We don't CLOAD anymore. We don't use punch cards. We got rid of punch-tape. Now let's get rid of tapes.
We disagree.
If you design a sane availability system, you can obviate tape and all the crap behind it. It's all about availability, audit, and a systematic approach to provisioning, archiving, and lifecycle.
The procedure isn't to connect and disconnect SAS or SATA cables, rather to pipe the data across network wires, in-band or out-of-band. Having a shadow network to perform backups, provision machines, re-slab servers and so on (especially VMs!) is the goal.
Tape media is fraught with difficulties, tho not as much as when we dealt with DC600s and 3740's. The cute little cartridges today are nice, and wonderfully fast, and still lack random accessibility, require physical manipulation (and are more subject to loss, theft, and environmental disasters), are proprietary to specific vendor solutions, and while more compatible with various OS blends than ever before, are still a hassle worth retiring, IMHO.
WIth no drives to restore onto, whatcha gonna do?
What about persistence, bleed through, good old oxidation, and leaving them in the back seat of a taxi?
Oh-- do we ever agree about that. But who says you have the data in just one building? My snapshots and mirrors are 13mi from where I sit. The backups for those are nine states away. Another mirror of those is in Canada. Cost? Not that much, but my payloads aren't the size of other organizations.
And listen: there's not one freaking tape. Not one. And it's randomly accessible fully, and the MTTR for primary systems is under eight minutes from click to verified. You, too, can do it.
Truly, you've been listening to too many sales guys.
And have you noted the difference between "consumer grade" hard drives and others? Do you realize the madness in MTBF statistics used by the various hard drive vendors? Do you track the failure modes between OEM and "consumer grade" drives? Do your research and put your money where you believe. I do.
Yes, the silver bullet is to drive it right through the heart of your expensive heaping honking tape library. Maybe garlic will help, too.
You can data dupe (2x cost). But you can also just say no, and start the procedure that gets you alternate storage, randomly accessible, easily tested, highly portable, and without proprietary media and infrastructure for about the same price as drive+tape+wiretime+hotelcost+human_intervention costs.
This is neither immoral or silver bullet-mentality. This is real and now. The sooner we shake tape, the better off we are. I've spent toooooooooo many hours watching jukeboxes shuffle stuff around to find out a vendor screwed up archiving, blew their compression algorithm, had a crappy library transport, or whatever-- all proprietary. Days of my life have been spent watching these things; they should be recycled at the earliest opportunity for the madness that they are.
You're welcome to pay for some EMC's salesperson's golf trip if you want. Enterprise class equipment certainly costs a bit more, but you're often paying for luxury where utility will do nicely.
Seagate bulk SAS drives 1tb barracudas are about $220 in oem bulk.... less if you want ata/300s. Compressed snapshots are fine in my experience, and better still, just in a nice, easy to slab hypervisor mountable.... or any other format that fits the bill. Delta snapshots are fine, too, depending on the transactional nature of the data. If it's transactional, then there are lots of great techniques to have multiple concurrent data instances that are easily coalesced when needed. The trouble: remembering to do fire drills.
Fine. Get the cartridges, but what about the capital cost minus depreciation of the drive? What about random access?
Now weigh those against an inexpensive jbod frame with a 2gb FC backplane. What's the write speed of LT vs a tasty little GB SAS drive? Rackspace? You can put a dozen into about 4U. Cooling? Although I'll grant you green cost, the random accessibility out-classes the seek time and tape insertion by a human cost dramatically. Stable media? Tape? Sometimes. Shelf space?
SAN fabric is dirt these days. You can get a nice Silkworm and a cheap-but-reliable SAS backplane for dirt as well. Perhaps a couple of GBICs.... or some handy-dandy fiber cables (also dirt these days) and you're in business. Or, put up a 10dot network off your public-face grid, and just use iSCSI. No need to use tape anymore. Get out of the reality distortion field, but do the right thing by testing what you have and doing drills to ensure that whatever you have, works and is a procedure understood by all.
Nope.
Mirrors are fine, just snapshot them and store them offsite regularly. Do delta backups as needed but close-in for fast restoration.
There is no rational justification for tape anymore, what with the cost per TB stored on hard disks now under $130, total $$. Random accessibility unless you're stalling a subpoena, is just mandatory on backup media.
I'm generally on the side of NPR and its content, except that it only rarely touches on the needs of minorities. There's a lot of WASP-ish content that's focused to YUPpie types, rather than trying to reach out to a larger community.
In terms of news content, there's a lot of belabored discourse, some of which is good, some beats a dead horse... perturbative.
The problem is that Chris Matthews and others claim their journalists, and in fact, they're entertainers.
Hey-- what is perfect? There are millions of answers because there are millions of people, all with different needs and skills sets.
Microsoft may give customers what they want, but they're insular and incestuous, and ride roughshod over anyone in their paths. Their convictions and settlements across the planet are a matter of fact. In giving customers what they want, they may have cornered the market illegally, or simply stolen what they gave to their customers.
Linux is only now becoming civilian friendly. Apple's done a reasonable job but portends as much proprietary hubris as Microsoft does. Apple gets great points for having dogmatic and high quality standards, but they're imperfect, too. They all are.
Were I to predict a final outcome, the 'Windows Franchise' becomes worth less and less, as Microsoft's infractructure slowly falls out of sync. Ask Oracle what happens then-- it nearly killed them.