If you fail over your HA systems every Sunday at 02:00 (or whatever time is safe...)
(voice of tech ignorant executive)
"We can't be down then. We have remote workers that want to do things at that time."
"The overtime for that window is too expensive, and we can't do it during production hours. We'll just assume you planned carefully."
"You just told me part of the reason that system is so expensive is that it is much less likely to fail. Well, we're not paying for a spare."
And after hearing that, I want to duct tape those fucking executives to their $1500 chair and let them watch while I take a powder-actuated nailer to their precious Mercedes S550.
Why the rage? Just spell out very clearly (and in writing) exactly what will happen if component X fails, and the cost to implement redundancy now. When component X fails and the company loses Y dollars of revenue and the CEO comes to you, just pull out the email and say "I tried to design redundancy but he wouldn't spend the money".
It worked for me when I tried to get money for a spare battery cabinet on our primary UPS. I told my boss that if a single battery in the string fails during a power failure, we'll lose most of the server room. That happened during a power outage -- the UPS went offline, we lost power to most of the equipment (we were able to scrounge up enough money for redundant power to some core network devices, but nearly all of the servers went down). Everything came up again after the generator restarted, then went down again after utility power came back because of a delay in transfer from generator to utility. My boss screamed at me, I told her I warned her it could happen and pulled out the email where I outlined exactly that scenario (except for the part where everything tripped off again during the switch back to utility power.
I had a new battery cabinet installed within 2 weeks, *and* a redundant UPS for most servers.
Cheap solution problem? Possibly. Infrastructure design fail? Possibly, but not likely. Couldn't critique it without seeing their setup, but it sounds like they designed some redundancy in. I wonder what kind of "memory card" failed. From the description, it sounds like it might be a cache module.
Uhh, if they designed in redundancy and it failed, isn't that the definition of "Infrastructure design fail"?
Regardless of whether or not Xeon's offer better performance per watt, there has to be a point where more money buys you better performing hardware per watt, so it still seems that the contest boils down to which team has the vendor with the most $$ or best performing hardware.
I can't find past results online, do they give prices for the systems in the contest results?
I'm not sure that 13 amps at 240VAC makes any more sense than 26 amps @ 120VAC - if they wanted to make it a single circuit, they could just provide a 30 amp 120VAC circuit.
Where do they mount them on the car? All my my GPS devices need a clear line of sight to the sky in order to work, even tree cover is enough to block the signal, so I don't see how the device can be mounted under the car and still operate? If they do need to be mounted with at least a small patch antenna with visibility to the sky, it seems that would make them easier to detect.
That makes this contest less interesting. It seems like the team with the best vendor contacts wins, regardless of how "smart" they are.
If team A's vendor is willing to pay $4,000 each for the top of the line 8 core Xeon's with 10Gig ethernet interconnects, but team B's vendor only supports 4 core Xeon's and only supplies GigE network, then it seems like team B loses no matter what they do.
Or are they judged on something besides raw performance?
True, but the point of the competition isn't to show off the vendor, or even the hardware; it's to show off how easy it is for undergraduate students, using commodity hardware, to construct and run a cluster that has more computing power than the fastest supercomputers from a decade ago at a fraction of the TCO.
If that's the goal, then why isn't cost of hardware included as part of the submission criteria?
And really, what is "commercial hardware"? If I can cram 36 Pico-ITX store bought motherboards into a 1U storebought case using standard power supply hardware connected with standard off the shelf wire, does that count as commercial off the shelf?
Or does "commercial" mean I have to buy a complete pre-built server from an established vendor?
From the project website (http://sc10.supercomputing.org/?pg=studentcluster_rules.html)
The computational hardware (processors, switch, storage, etc.) must fit into a single rack. All components associated with the system, and access to it, must be powered through the two 120-volt, 20-amp circuits, (each with a soft limit of 13 amps) for a total of 26 amps, provided by the conference. Power to each system will be provided via metered power distribution units The equipment rack must be able to physically hold these metering power strips.
This makes it even harder since theyir hardware has to be power balanced between the two power strips. They'll have to come up with some dynamic load balancing between cluster nodes based on power consumption. I guess dual power supplies might help (do dual power supplies draw perfectly balanced power between both power inputs?), but at a loss of power efficiency.
High technology is not likely to solve something that is actually a social problem. People in your country are accustomed to driving any way they like with impunity. Installing speed cameras in a few places is not likely to change their behavior. More likely, your camera is going to end up getting run down by a car (or with a bullet hole through the lens if it's mounted away from the side of the road). Are these tickets going to be legally enforceable under your country's legal system?
If you're determined to use technology to solve the problem, instead of trying to use a high tech method to punish drivers to alter behavior (especially when the punishment comes long after the infraction), look into low-tech methods like installing speed humps to reduce speed (which are not the same as speed *bumps* which are actually felt less the faster the car is traveling).
There are many other traffic calming measures that can be employed, like narrowing streets, using traffic circles, etc. These have the advantage that they are completely passive and work against all drivers, not just the ones honest enough to want to avoid a ticket.
Maybe, but those rural areas create the food that the cities need to house and feed their populations. When you increase the costs of those areas, you greatly affect the cost of city life. Cities are also far, far more subsidized then any rural area is. The roads needed to truck in supplies, heavily subsidized food programs, and greatly disproportionate distribution of state tax income as well as federal aid.
Increasing costs to live in rural areas will only slightly increase food prices to city dwellers. There are around 2 million farmers in the US, to 250M people, so the ratio of farmers to consumers is around 100:1.
If a farmer's expenses increase 100%, then the affect on food prices is just around 1%.
A 1% increase in raw food prices would probably increase the average city dweller's food prices 0.1% - 0.5%.
But you talk as if the farmer is growing food as a convenience to the city dwellers -- he grows food because it's his job. Without anyone in the city to buy his product, the farmer will have no job.
This won't work unless you retrofit vehicles with standard attachment points or indeed build new vehicles designed for this purpose, because the loading, securing, unsecuring, and unloading process is too arduous otherwise. You could do it if you only used Mercedes, since they have those semi-standardized jacking points, but even that is not really what I had in mind.
Unless you're doing a scientific survey, it may very well *appear* to be just as many overweight people in Europe, when you look at the numbers, the US is well in the lead.
So, unless you're actually counting people, it may be hard to see the difference between the 23% in the UK and the 30% in the US, but still it's a large difference.
Lucky for you that you have so many choices. I have only a single effective choice. And it's not like I live in some rural area.
I live in a large residential neighborhood in San Francisco and my choices are either Comcast or 1.5mbit/sec (or less) DSL. There are several DSL carriers to choose from, but they all deliver on the same copper and due to my distance to the CO, they are all limited to 1.5mbit.
My primary use for internet is for Netflix streaming, and I can't stream HD @ 1.5mbit, so that leaves me with a single carrier to choose from (Comcast).
I really can't complain about Comcast (so far). I typically get 15 - 20mbit/second for sustained downloads and they've been reliable, I haven't had a known internet outage in over 6 months. But, if they wanted to force me to use their own streaming service, without Net Neutrality they could throttle back the bandwidth to Netflix.
So where's the typo on the "Pedestrians use walks not roads" sign? A walk is " a place designed for walking" according to the dictionary. It doesn't seem like a sign should follow full grammatical sentence structure since it's supposed to be a brief message directing people on what to do. How else should they have worded the sign, while still keeping it terse and to the point? They can't say "sidewalks", because that looks like a national park sign, and not all "walks" are sidewalks. Nor can they say "trails" since not all paths are trails. Maybe they could have said "path", but is a sidewalk the same as a "path"? Maybe they could have said "Pedestrians must use pedestrian pathways, not roads" but that doubles the cost of the sign and gives very little added benefit.
Am I missing something else that makes the sign incorrect?
Re:Don't f* with the IT guy like at restaurant you
on
Child Porn As a Weapon
·
· Score: 1
How did you go from reading "Janitor" to equating that with the time you worked at a restaraunt, presumably as a waiter?
Yeah, food service sucks. That's why you got a degree and a better job. All us IT guys are just making jokes, not personally attacking your history.
He was just pointing out that it's not true that IT folks deal with more shit. Neither literally nor figuratively. When the toilet overflows and seeps into the server room, IT isn't called in to mop it up. When someone pukes on their keyboard, IT's role is to walk down with a new keyboard after the janitors clean up the mess.
Granted, the janitor isn't going to get called at 3am because Exchange is down, but I'd rather log on at 3am to clean out the porno-spam that filled up my disks than clean up the disgusting mess someone left on the bathroom floor. Besides, at least IT has control in building a reliable solution that won't go down, but Janitorial staff can't do anything to keep an 8 year old kid from puking all over the carpet.
Don't work at that company until they get some more professional IT staff.
and word is getting around that the admins who were given access to his Outlook account have found personal things that are embarrassing at best
If any of my IT admins revealed any personal information about a mailbox they'd been given access to, they'd be looking for a new job right now. If they can't stop blabbing about someone's sex-change operation, why should I think that they can keep any confidential business data a secret?
Technically, it's a permit, not a license. And just like you need to get a permit when you add a room onto your house, some localities require a permit for pools. While everyone sees it as a way to make money, the permit does have specific safety requirements about fencing the pool, suction inlets, and has to be signed off by an electrician.
Given the number of pool deaths that happen in my area (seems like 3 - 6 every summer), it sounds like reasonable requirements.
This permit is only required for pools deeper than 24", so your kids wading pool that you bought from Walmart isn't going to need a permit, but if you want to put up a 3 foot deep above ground pool, you'll have to get a permit and follow the regulations on fencing, etc.
But it wasn't available on AT&T until March -- 3 months after it became available on T-Mobile. By then, the the impending release of the HTC Incredible (a better phone than the Nexus one - made by the same company) was know, further eroding the potential market.
The biggest lesson Google learned is that without the marketing muscle of a major cell phone carrier, they aren't going to sell phones.
I agree with the other posters that doubt that it has to do with consumers seeing the phone first.
There are three reasons why I didn't buy the Nexus one:
It was on T-mobile rather than being on Verizon (or even AT&T)
Buying an unsubsidized phone gives very little benefit - it's not even clear that the plan is cheaper if I buy an unsubsidized phone.
No keyboard (but that's just my personal preference)
Now, when the Motorola Droid was released, I bought it on opening day -- I didn't even see the phone until the sales rep took my phone out of the box to activate it. I bought my last 2 phones online and didn't see them in person until they arrived. I'm definitely in the target market of the Nexus One, but my decision to not buy the phone had nothing to do with whether or not I could see it first.
I wouldn't say that Google Voice does a decent job, most of the voice mails left for me on Google Voice come out like this:
Hey when you get this is a I'm via out what you know what it's Johnathon of the bad idea. So maybe we can meet that I don't know about that. And I do that, but what I thought I had, but that's also the Damon now, but I thought I had bought the house. I had to get out of the night okay and give me a call. bye bye.
Except for the "bye bye" at the end, none of it is close enough to the actual message to be useful. The actual message said nothing about Johnathan, any meeting, Damon, or a house.
'The only remaining question will then be: why on earth did we squander so much money by not thinking this way until now?'"
The reason is that for 90% of businesses, the software and processes that could actually manage migrating unused data off to where it's not on primary storage but still accessible is so expensive and complex so as to not be worth it.
My company has about 10TB of corporate fileserver data. It's all sitting on SATA disks in a big NAS (well, used to be "big", nowadays it's "small"). Much of that data is important to the company, but may rarely be needed. While I could purchase a tiered storage system with fast FC drives for recently used data, slower SATA for nearline storage and tape for rarely used data plus software to manage it all, in reality it's cheaper to just keep adding SATA shelves to keep it all online. No one wants to pay for a librarian to manage Documentum or other such product to enable us to move unused data offline.
Plus there's the fact that personally, I don't trust tape for "offline" storage -- if it's not spinning and scrubbed, then it may not be readable. I still use tape for offsite disaster recovery, but would feel better if I could replicate data offsite.
In the article they don't even seem to advocate a tape backend, just SAS disks on the front end and SATA for nearline, but I don't see the point in that -- I've got a measured 98% read to write ratio -- SATA with a good NAS gives me more than enough performance for corporate fileserver needs, why would I want to pay the price premium to put a fast SAS cache in front of my SATA disks? SAS or FC gives me about 3X the IOPS at 8X the cost of SATA.
Perhaps in a large company it makes more sense to move to tiered storage, but in the 1000 user range, I just don't see the benefit.
(Also, god help you if you put a database server on RAID 5... goodbye performance! RAID 10 or bust.)
Only if your database is writing fast enough to overflow the battery backed cache on your RAID controller. As long as your writes fit in the cache before they get flushed to disk, then IOPS's of the drive don't really matter.
And if the parent poster really is running his databases on 3 SATA drives + $100 SATA controller, I'd bet that his database isn't busy enough to warrant RAID1+0 let alone an SSD. His 3 disk RAID volume gives him the performance he needs at a good pricepoint.
Of the 70 databases I manage (a mix of SQL/Server, Oracle, and MySQL), none of them are busy enough to make switching to SSD's cost effective -- most get along quite well on a shared SAN LUN on RAID-DP (RAID-6) - with only 2GB of cache shared among them.
But my hard disk already uses ECC to check for bit errors at the block level, so I'm already paying for ECC on the hard drive. And the SATA path to the hard drive also uses some CRC or ECC checking.
While my video card may have a flipped bit, it's not going to silently corrupt my data
I've seen estimates ranging from 1 flipped bit per month to 4 flipped bits per day for 4GB of RAM -- do you have evidence that bit flips are rare aside from anecdotal evidence that people aren't reporting corruption all the time?
I can hold my Android phone any way I want to and I don't see any loss of signal strength, so apparently it *is* possible to design a working phone within current SAR limits...
I wasn't aware that government regulation forced BP to cut corners and undertake a risky well closure strategy, ignoring evidence that the BOP was damaged. Which regulation was that?
I went to Dell's site and configured a few Dell Desktops (non-ECC) and Workstations (with ECC), and prices were similar for comparable systems. Though the Workstations that supported ECC didn't support many low-end processors, so if i didn't want ECC and didn't care about processor performance I could have gotten a desktop for about 60% of the price of the cheapest workstation with ECC. But I didn't see a 5x increase for ECC.
Slashdot Mirror
If you fail over your HA systems every Sunday at 02:00 (or whatever time is safe...)
(voice of tech ignorant executive)
"We can't be down then. We have remote workers that want to do things at that time."
"The overtime for that window is too expensive, and we can't do it during production hours. We'll just assume you planned carefully."
"You just told me part of the reason that system is so expensive is that it is much less likely to fail. Well, we're not paying for a spare."
And after hearing that, I want to duct tape those fucking executives to their $1500 chair and let them watch while I take a powder-actuated nailer to their precious Mercedes S550.
Why the rage? Just spell out very clearly (and in writing) exactly what will happen if component X fails, and the cost to implement redundancy now. When component X fails and the company loses Y dollars of revenue and the CEO comes to you, just pull out the email and say "I tried to design redundancy but he wouldn't spend the money".
It worked for me when I tried to get money for a spare battery cabinet on our primary UPS. I told my boss that if a single battery in the string fails during a power failure, we'll lose most of the server room. That happened during a power outage -- the UPS went offline, we lost power to most of the equipment (we were able to scrounge up enough money for redundant power to some core network devices, but nearly all of the servers went down). Everything came up again after the generator restarted, then went down again after utility power came back because of a delay in transfer from generator to utility. My boss screamed at me, I told her I warned her it could happen and pulled out the email where I outlined exactly that scenario (except for the part where everything tripped off again during the switch back to utility power.
I had a new battery cabinet installed within 2 weeks, *and* a redundant UPS for most servers.
Cheap solution problem? Possibly. Infrastructure design fail? Possibly, but not likely. Couldn't critique it without seeing their setup, but it sounds like they designed some redundancy in. I wonder what kind of "memory card" failed. From the description, it sounds like it might be a cache module.
Uhh, if they designed in redundancy and it failed, isn't that the definition of "Infrastructure design fail"?
Regardless of whether or not Xeon's offer better performance per watt, there has to be a point where more money buys you better performing hardware per watt, so it still seems that the contest boils down to which team has the vendor with the most $$ or best performing hardware.
I can't find past results online, do they give prices for the systems in the contest results?
I'm not sure that 13 amps at 240VAC makes any more sense than 26 amps @ 120VAC - if they wanted to make it a single circuit, they could just provide a 30 amp 120VAC circuit.
Where do they mount them on the car? All my my GPS devices need a clear line of sight to the sky in order to work, even tree cover is enough to block the signal, so I don't see how the device can be mounted under the car and still operate? If they do need to be mounted with at least a small patch antenna with visibility to the sky, it seems that would make them easier to detect.
That makes this contest less interesting. It seems like the team with the best vendor contacts wins, regardless of how "smart" they are.
If team A's vendor is willing to pay $4,000 each for the top of the line 8 core Xeon's with 10Gig ethernet interconnects, but team B's vendor only supports 4 core Xeon's and only supplies GigE network, then it seems like team B loses no matter what they do.
Or are they judged on something besides raw performance?
True, but the point of the competition isn't to show off the vendor, or even the hardware; it's to show off how easy it is for undergraduate students, using commodity hardware, to construct and run a cluster that has more computing power than the fastest supercomputers from a decade ago at a fraction of the TCO.
If that's the goal, then why isn't cost of hardware included as part of the submission criteria? And really, what is "commercial hardware"? If I can cram 36 Pico-ITX store bought motherboards into a 1U storebought case using standard power supply hardware connected with standard off the shelf wire, does that count as commercial off the shelf? Or does "commercial" mean I have to buy a complete pre-built server from an established vendor?
The computational hardware (processors, switch, storage, etc.) must fit into a single rack. All components associated with the system, and access to it, must be powered through the two 120-volt, 20-amp circuits, (each with a soft limit of 13 amps) for a total of 26 amps, provided by the conference. Power to each system will be provided via metered power distribution units The equipment rack must be able to physically hold these metering power strips.
This makes it even harder since theyir hardware has to be power balanced between the two power strips. They'll have to come up with some dynamic load balancing between cluster nodes based on power consumption. I guess dual power supplies might help (do dual power supplies draw perfectly balanced power between both power inputs?), but at a loss of power efficiency.
High technology is not likely to solve something that is actually a social problem. People in your country are accustomed to driving any way they like with impunity. Installing speed cameras in a few places is not likely to change their behavior. More likely, your camera is going to end up getting run down by a car (or with a bullet hole through the lens if it's mounted away from the side of the road). Are these tickets going to be legally enforceable under your country's legal system?
If you're determined to use technology to solve the problem, instead of trying to use a high tech method to punish drivers to alter behavior (especially when the punishment comes long after the infraction), look into low-tech methods like installing speed humps to reduce speed (which are not the same as speed *bumps* which are actually felt less the faster the car is traveling).
There are many other traffic calming measures that can be employed, like narrowing streets, using traffic circles, etc. These have the advantage that they are completely passive and work against all drivers, not just the ones honest enough to want to avoid a ticket.
Maybe, but those rural areas create the food that the cities need to house and feed their populations. When you increase the costs of those areas, you greatly affect the cost of city life. Cities are also far, far more subsidized then any rural area is. The roads needed to truck in supplies, heavily subsidized food programs, and greatly disproportionate distribution of state tax income as well as federal aid.
Increasing costs to live in rural areas will only slightly increase food prices to city dwellers. There are around 2 million farmers in the US, to 250M people, so the ratio of farmers to consumers is around 100:1.
If a farmer's expenses increase 100%, then the affect on food prices is just around 1%.
A 1% increase in raw food prices would probably increase the average city dweller's food prices 0.1% - 0.5%.
But you talk as if the farmer is growing food as a convenience to the city dwellers -- he grows food because it's his job. Without anyone in the city to buy his product, the farmer will have no job.
This won't work unless you retrofit vehicles with standard attachment points or indeed build new vehicles designed for this purpose, because the loading, securing, unsecuring, and unloading process is too arduous otherwise. You could do it if you only used Mercedes, since they have those semi-standardized jacking points, but even that is not really what I had in mind.
Or just use nylon straps to strap down the wheels: http://www.uscargocontrol.com/Towing-Auto-Hauling/Wheel-Nets-Auto-Car-Tie-Downs/Side-Mount-Wheel-Net-W-Cam-Buckle-2-Ratchets-and-Flat-Hooks Mount them on plates or rods that can be attached anywhere along the deck to allow for different sized cars. Every car has wheels and these tie-down straps are already in widespread use on flatbed trailers. Experienced crew can tie down a car within a few minutes.
Unless you're doing a scientific survey, it may very well *appear* to be just as many overweight people in Europe, when you look at the numbers, the US is well in the lead.
Here are the top 5 leaders (losers?) in Obesity:
The full list is here
So, unless you're actually counting people, it may be hard to see the difference between the 23% in the UK and the 30% in the US, but still it's a large difference.
Lucky for you that you have so many choices. I have only a single effective choice. And it's not like I live in some rural area.
I live in a large residential neighborhood in San Francisco and my choices are either Comcast or 1.5mbit/sec (or less) DSL. There are several DSL carriers to choose from, but they all deliver on the same copper and due to my distance to the CO, they are all limited to 1.5mbit.
My primary use for internet is for Netflix streaming, and I can't stream HD @ 1.5mbit, so that leaves me with a single carrier to choose from (Comcast).
I really can't complain about Comcast (so far). I typically get 15 - 20mbit/second for sustained downloads and they've been reliable, I haven't had a known internet outage in over 6 months. But, if they wanted to force me to use their own streaming service, without Net Neutrality they could throttle back the bandwidth to Netflix.
So where's the typo on the "Pedestrians use walks not roads" sign? A walk is " a place designed for walking" according to the dictionary. It doesn't seem like a sign should follow full grammatical sentence structure since it's supposed to be a brief message directing people on what to do. How else should they have worded the sign, while still keeping it terse and to the point? They can't say "sidewalks", because that looks like a national park sign, and not all "walks" are sidewalks. Nor can they say "trails" since not all paths are trails. Maybe they could have said "path", but is a sidewalk the same as a "path"? Maybe they could have said "Pedestrians must use pedestrian pathways, not roads" but that doubles the cost of the sign and gives very little added benefit.
Am I missing something else that makes the sign incorrect?
How did you go from reading "Janitor" to equating that with the time you worked at a restaraunt, presumably as a waiter?
Yeah, food service sucks. That's why you got a degree and a better job. All us IT guys are just making jokes, not personally attacking your history.
He was just pointing out that it's not true that IT folks deal with more shit. Neither literally nor figuratively. When the toilet overflows and seeps into the server room, IT isn't called in to mop it up. When someone pukes on their keyboard, IT's role is to walk down with a new keyboard after the janitors clean up the mess.
Granted, the janitor isn't going to get called at 3am because Exchange is down, but I'd rather log on at 3am to clean out the porno-spam that filled up my disks than clean up the disgusting mess someone left on the bathroom floor. Besides, at least IT has control in building a reliable solution that won't go down, but Janitorial staff can't do anything to keep an 8 year old kid from puking all over the carpet.
and word is getting around that the admins who were given access to his Outlook account have found personal things that are embarrassing at best
If any of my IT admins revealed any personal information about a mailbox they'd been given access to, they'd be looking for a new job right now. If they can't stop blabbing about someone's sex-change operation, why should I think that they can keep any confidential business data a secret?
Technically, it's a permit, not a license. And just like you need to get a permit when you add a room onto your house, some localities require a permit for pools. While everyone sees it as a way to make money, the permit does have specific safety requirements about fencing the pool, suction inlets, and has to be signed off by an electrician.
Given the number of pool deaths that happen in my area (seems like 3 - 6 every summer), it sounds like reasonable requirements.
This permit is only required for pools deeper than 24", so your kids wading pool that you bought from Walmart isn't going to need a permit, but if you want to put up a 3 foot deep above ground pool, you'll have to get a permit and follow the regulations on fencing, etc.
The biggest lesson Google learned is that without the marketing muscle of a major cell phone carrier, they aren't going to sell phones.
Now, when the Motorola Droid was released, I bought it on opening day -- I didn't even see the phone until the sales rep took my phone out of the box to activate it. I bought my last 2 phones online and didn't see them in person until they arrived. I'm definitely in the target market of the Nexus One, but my decision to not buy the phone had nothing to do with whether or not I could see it first.
Hey when you get this is a I'm via out what you know what it's Johnathon of the bad idea. So maybe we can meet that I don't know about that. And I do that, but what I thought I had, but that's also the Damon now, but I thought I had bought the house. I had to get out of the night okay and give me a call. bye bye.
Except for the "bye bye" at the end, none of it is close enough to the actual message to be useful. The actual message said nothing about Johnathan, any meeting, Damon, or a house.
'The only remaining question will then be: why on earth did we squander so much money by not thinking this way until now?'"
The reason is that for 90% of businesses, the software and processes that could actually manage migrating unused data off to where it's not on primary storage but still accessible is so expensive and complex so as to not be worth it.
My company has about 10TB of corporate fileserver data. It's all sitting on SATA disks in a big NAS (well, used to be "big", nowadays it's "small"). Much of that data is important to the company, but may rarely be needed. While I could purchase a tiered storage system with fast FC drives for recently used data, slower SATA for nearline storage and tape for rarely used data plus software to manage it all, in reality it's cheaper to just keep adding SATA shelves to keep it all online. No one wants to pay for a librarian to manage Documentum or other such product to enable us to move unused data offline.
Plus there's the fact that personally, I don't trust tape for "offline" storage -- if it's not spinning and scrubbed, then it may not be readable. I still use tape for offsite disaster recovery, but would feel better if I could replicate data offsite.
In the article they don't even seem to advocate a tape backend, just SAS disks on the front end and SATA for nearline, but I don't see the point in that -- I've got a measured 98% read to write ratio -- SATA with a good NAS gives me more than enough performance for corporate fileserver needs, why would I want to pay the price premium to put a fast SAS cache in front of my SATA disks? SAS or FC gives me about 3X the IOPS at 8X the cost of SATA.
Perhaps in a large company it makes more sense to move to tiered storage, but in the 1000 user range, I just don't see the benefit.
(Also, god help you if you put a database server on RAID 5... goodbye performance! RAID 10 or bust.)
Only if your database is writing fast enough to overflow the battery backed cache on your RAID controller. As long as your writes fit in the cache before they get flushed to disk, then IOPS's of the drive don't really matter.
And if the parent poster really is running his databases on 3 SATA drives + $100 SATA controller, I'd bet that his database isn't busy enough to warrant RAID1+0 let alone an SSD. His 3 disk RAID volume gives him the performance he needs at a good pricepoint.
Of the 70 databases I manage (a mix of SQL/Server, Oracle, and MySQL), none of them are busy enough to make switching to SSD's cost effective -- most get along quite well on a shared SAN LUN on RAID-DP (RAID-6) - with only 2GB of cache shared among them.
But my hard disk already uses ECC to check for bit errors at the block level, so I'm already paying for ECC on the hard drive. And the SATA path to the hard drive also uses some CRC or ECC checking.
While my video card may have a flipped bit, it's not going to silently corrupt my data
I've seen estimates ranging from 1 flipped bit per month to 4 flipped bits per day for 4GB of RAM -- do you have evidence that bit flips are rare aside from anecdotal evidence that people aren't reporting corruption all the time?
I can hold my Android phone any way I want to and I don't see any loss of signal strength, so apparently it *is* possible to design a working phone within current SAR limits...
I wasn't aware that government regulation forced BP to cut corners and undertake a risky well closure strategy, ignoring evidence that the BOP was damaged. Which regulation was that?
I went to Dell's site and configured a few Dell Desktops (non-ECC) and Workstations (with ECC), and prices were similar for comparable systems. Though the Workstations that supported ECC didn't support many low-end processors, so if i didn't want ECC and didn't care about processor performance I could have gotten a desktop for about 60% of the price of the cheapest workstation with ECC. But I didn't see a 5x increase for ECC.