I appreciate where you are coming from and I cannot comment on AIX or HPUX, but I can comment on Solaris. Sun has been fascist about binary backword compatibility.
I wrote a 32 bit SCSI target driver in the early-90s using the (then deprecated) Solaris 2.3 ABI and it worked flawlessly until the early 2000s when we were required to switch to a 64 bit kernel, a switch we had delayed since the 90s.
The company still runs a version of Oracle compiled for Solaris 2.2 (1992?) and it works flawlessly, except it has a bug in its use of log archiving that was masked by earlier versions of the UFS. They still run the 14-year old 32 bit Oracle binaries on a 64 bit Solaris 8 box.
Backword compatibility is critical in the enterprise, and honestly, much of the OSS world ignores this.
Number of rewrites on solid state storage: ~1 million.
Number of rewrites on a laptop hard disk: Until the drive mechanism dies.
Hope you don't do a lot of swapping on your solid state flash hard drive.
Why does this myth refuse to die? These do "wear leveling" which moves the writes around the flash and means that you would need to write the whole drive one million times.
Let's do some math. One million writes of 32 GB equals 32,000,000 GB, or 32 PB.
Suppose you average 10 MB/s of writes the whole time your laptop is in use (good luck pulling that off). You would have 3.2 billion seconds of use, or 101 years of continued use.
Let's see a hard drive take that kind of pounding.
- batteries. You can run a heavy, hot laptop for about 2-3 hours before needing a lengthy recharge. You can run a handheld PDA for roughly the same amount of time;
A Nokia 9500 runs a full workday (and then some) on a single charge. Each newer device gets better, and soon all devices will carry enough juice for the average Joe.
- data input. The desktop/laptop has this amazing invention, the full size keyboard, that lets us enter tons of information more quickly and accurately than any other method.
While I agree, there are plenty of applications that do not require a full-size keyboard.
- storage. Desktops start at 40G of permanent storage and go up to terabytes. Nothing else can compare. What's more, our storage needs are growing, not shrinking. We're not going to switch to Pocket PC/Phone/consoles that have maybe a 10G memory card or a 30G hard disk and give up our 250 giggers.
This is the trend you miss. Flash storage is going up at a much faster rate than storage appetites. Flash storage is doubling annually. At some point in the next few years, flash will provide enough storage for most people's personal data.
- connectivity. A desktop is on DSL or Cable or T1 or dial-up and is a reliable way to access the internet. Handheld devices have to be in range of a wireless hub or in network for cellular connections. The widely available connectivity for broadband handheld devices simply doesn't exist yet.
You hit the nail on the head: yet.
As storage is further miniaturized and as voice input and battery technology improve, we will doubtless see a displacement of casual desktop/laptop use with handhelds, such as Blackberry-style email reading and Palm/PPC-style organizer functions, but for heavy lifting, the desktop will remain.
And each year, the number of applications that fall into the category of "heavy lifting" will diminish until the number is so small that no one really cares.
I doubt IT departments will encourage wholesale adoption of this over local storage solutions.
There will, of course, be resistance from the old school crowd, just like there is with any new, outsourced service. The advantages to this are mind-numbing, including:
Uniform clients
Centralized management
Lower overall cost
Centralized upgrades
Your provider upgrades the software/hardware in one place, and they get the joy of maintaining it, backing it up, ensuring availability, etc.
All you have to do is make sure the desktops are suitable, pay a per-desktop/per-month fee and everything is done. Your company can go back to focusing on whatever its core business is.
This transition is inevitable and has already started.
These guys might disagree with your assessment of it being for "dinky" apps.
At the risk of turning this into a Postgresql discussion, Yahoo! has some serious data integrity issues. Often, checking your e-mail does not cause the "new e-mail" indicator to go out for up to a minute.
This is just an annoyance for Yahoo!, but would be the death-knell of any banking or inventory management application. Imagine a bank where transactions were delayed and posted in a random order. With a constant transaction flow, there would never be a moment in time when you could actually see what the status of the bank accounts are -- scary stuff.
Parent is absolutely correct about Symbian, which has already dominated the smart mobile device market (PDA and smart phones).
Look at the numbers and you will see that Symbian twice outsells Microsoft and Palm combined.
The whole Palm vs. MS debate is like Wendy's fighting Burger King and both pretending McDonald's doesn't exist.
Why yes, I did post this with a Symbian phone.
If you are deploying a system that will be around for several years, Sparc Solaris lets you upgrade hardware and software with minimal hassles and downtime.
Try bringing a 1998 Linux install in a 24x7 environment and upgrading the disks, CPU, memory, mobo, networking, the OS, apps and libraries without breaking everything that has been tacked on to the system in the past 7 years. This is done all the time with Solaris.
If the lifespan of your app/server is only 18 months, then none of this matters.
Re:Read the fine print for your savings and checki
on
Tracking Your Taxes
·
· Score: 5, Informative
Banks make good money selling your financial information to "related buisnesses".
Sorry to burst your bubble. I work closely with dozens of banks and credit unions on this very topic. GLB inspires more neurotic fear in bankers than anything else I have seen in some time.
By the way, "related businness" means sharing information with other companies that must be there to support the bank, like disaster recovery companies, records archiving companies, etc.
Whether or not you believe it, "related businesses" simply cannot use your information for anything other than performing their service for the bank.
The closest a bank can get to profiting from your personal information is using it to offer services. A bank may notice that you have a high credit card balance and offer you a HELOC, it may notice that you have a high savings balance and offer its CDs, it may notice that your car loan is getting paid down and offer a pre-approved loan for a newer car, etc.
Other than that, your information is strictly off limits.
... there are no Operating System APIs that IE uses that are not documented on MSDN as part of the platform SDK and available to other browsers and any other software that runs on Windows..
Then how is it that...
IE is part of the Windows Operating System so that parts of the OS and other applications can rely on the functionality and APIs being present.
These two statements seem to contradict each other.
Not necessarily. I read those comments as:
IE does not use infrastructure unavailable to other applications
IE becomes infrastructure that other applications can depend on
Parent is correct. Has anyone on slashdot ever tried to install unsigned software on a Symbian device?
It is littered with warnings and confirmation screens. Anyone who got this virus had to endure the installation process confirmations. It is worse than a EULA.
I find that I lack sympathy for a user who repeatedly selected "ok" and "continue" after being warned that this software cannot be verified -- software that arrived unsolicited.
It takes a whole new kind of inattention to allow this virus to spread.
The real advantage to this is when you need the headroom. If 5% of the time your organization needs massive computing power, then you need to maintain that overhead the other 95% of the time when you don't need it.
With a utility solution you pay only for what you use, but have the headroom to spike when you need it. Very cool.
Commercials. If advertisers know their audience won't see their commercials, they lose incentive to invest in advertising, and the networks lose money.
You may be on to something here. They probably have a hard time selling commercial space on the internet if they cannot say who is watching what. Nielsen ratings allow the networks to sell ads based on hard evidence of viewership, but...
What if Tivo teamed up with a network to offer shows via broadband, complete with ad injection? You could watch what you want regardless of your local lineup. The network could get feedback from Tivo on what people are watching which shows, broken down by demographics.
The only downside to this that I can think of is that a network trying this out would alienate its current last-mile providers: TV stations, cable or satellite companies, which have probably demanded some sort of exclusive.
Supporing internet downloads might be a contract violation or worse, piss off the last-mile providers enough that they drop the network completely, leaving the network only with the nascent internet market.
Now that I think about it, I can see why they just want the internet thing to go away.
This is not a winner. My cell phone is used to call people and for them them to call me. It is not a PDA, that is a separate device. It's also not a web browser, a camera, nor a music player. These devices are separate for a reason: so I can use them all only when I want, so I can upgrade them when I feel like, and if one of them breaks (or gets lost) then they are not all gone. Also, I can buy the individual devices much more easily because I'd buy only 1 a month, rather than having to buy the very expensive all-in-1 device; and who says they'll even have all the right features anyway?
If you don't want one of these phones, simply don't buy one. There are plenty of "basic phones" that do what you want.
The problem for the carriers is that they are trying to preserve existing revenue, increase revenue per customer and add new customers.
There is a segment of the market shouting, "Just give me a basic phone that works everywhere at a decent price and you will have my business."
The problem with this group is that they don't help the carriers reach their growth objectives. The only way to attract these folks is to lower prices (killing margins) or install many more towers (killing margins). Worse, there is no loyalty in this group. If the carrier does manage to attract them, then these subscribers will defect en masse to the next carrier that offers slightly lower prices or slightly better coverage.
it's been around for years and it cost under $2 a ride
I love these posts from the "mass transit cures all ills" fanatics.
Let's see how well some of us can operate with a bus (assuming it goes anywhere near my home and work, which it does not):
* Drop off and pick up kids at school and daycare?
Nope.
* Get groceries for my family of six?
Riiiight. I will just take the shopping cart brimming full with 12 bags of stuff up the small bus stairs and sit on my seat.
* Rush to get a sick child from school?
No.
* Take same sick child to doctor?
No.
* Pick up dry cleaning?
No.
* Take child to evening school event?
No again.
* Parent - Teacher conferences?
No.
* Take child to soccer? To his friend's birhtday party? To get a haircut?
No again.
* Visit a client?
Nope.
* Work late?
No.
* Out to dinner with my wife?
No.
On and on...
So, I guess if I had no family and no clients, lived alone, worked at my job only when the bus was running and could buy groceries only for me, then yes, a bus might make sense, if it came near my work or home.
Other than the above minor limitations, you are right, the bus is for everyone.
Their revenue stream and rights to artistic integrity end when they reach the consumer. At that point it's my signal in my device and as long as I don't rebroadcast it they need to leave me the hell alone.
While IANAL, I suspect that if you engage in activity that breaks their business model, they are well within their rights to exclude you from their business.
Movie theatres usually lose money on movie tickets and recover the money from the concession stand. Theatres can (and do) legally restrict patrons from bringing in their own food.
Some websites (slashdot?) give away the web service and make money from advertising. Some websites can (and do) legally restrict bots from their site.
Your estimate should actually be five to six weeks.
Yeah, I realized after I had posted it that I had an order of magnitude error in my math.
It doesn't change much, though, because many devices will withstand 1 million write cycles instead of the 100,000 I used in the calculations, giving the same result.
But a hard drive needs a FAT or something similar, which is generally going to live in one spot on the disk-- you're going to hit the max writes on that segment fairly quickly.
IIRC, the cards use a level of indirection so that the same sector gets put in different physical places. You may write to sector 11843 over and over, but it actually gets written to a different physical spot each time.
I thought the problem with flash was a limited number of write cycles (10,000-100,000?). With this thing rated at up to 25,000 IOPS, is would seem that they might not last all that long (4 seconds?).
That may be true, but modern stuff does "wear leveling" which spreads out the writes. In essence, you would have to write the whole card 100,000 times. For a 1 GB card, that's 100 TB of writes. At a sustained 30 MB/s of writes (which is probably very rare), it would take three million seconds, or one year of constant writing, just to kill the card, which brings us to...
It should be fairly simple to put together a "CF-raid" drive for way less than $1K/GB.
I am amazed that this has not happened yet. For applications that do lot's of critical writing in tiny blocks (think Oracle log files), this would, speedwise, destroy any spindle-based competitor at a low cost.
The truth is that flash technology is increasing at a rate faster than the needs of typical server computing, which makes it inevitable that these will take over the datacenter some day.
That's like a car manufacturer saying, "We've installed a mechanism which will keep you from opening the hood if your intention is to upgrade the engine, because we want you leasing and buying new expensive cars very soon."
Or it is more like a car manufacturer saying, "The engine is tuned to handle X. If you put a blower or nitrous on it and ramp up the output, you are on your own, no warranty. If you buy a 200 HP Hondaccord, we will ensure that it has the actual 200 HP engine, and not a modified 120 HP engine that will fail sooner."
I would bet this is only to stem companies selling overclocked boxes and advertising them at the overclocked speed. Neither Intel nor the customer wins in this scenario. They are probably just stopping fraud.
Thank God, finally someone understands! I was recently benchmarking Turtle Logo against some highly optimized Opteron-specific assembly code for an application that calculates PI to the 400,000th digit.
I was surprised to see that Logo was slower, but I think I just didn't optimize it enough. I am tuning the Logo floating point code right now, and Logo should stomp the hell out of the assembly code "Real Soon Now."
There are lots of examples where a converged device was horrid. There are lots of examples where an independent device was horrid.
Where there are synergies from convergence, things converge, and always have.
To the crowd that says, "Just give me a cellphone that works!" I tell you that there ain't enough of you to drive phone purchasing trends, so you will be ignored. Companies (rightly) focus on what they think will drive the bottom line.
Imagine hearing from someone, "All I want is a frickin' car! I don't need a recliner, a stereo system, HVAC, a roof, attractive paint, seat belts, air bags, electric starter or any of that other frilly crap. All I want is a car that goes from A to B." You, my friend, are in the minority and always have been.
Personally I don't subscribe to a newspaper. I bought one on Saturday though, and as far as I'm aware I can keep as long as I want and read it as many times as I want without having to pay anybody anymore money.
Well said. You could also buy the latest DVD of your favorite movie, or rent it for far less.
I subscribe to cable, my cell phone, power company, etc because I get continuing service from them which I can't buy outright as it isn't an asset. And last time I went on holiday I stayed in a hotel as it was much cheaper than buying a house and then selling it again at the end of the week.
Another good point. Yes, you could buy all of those things, but renting it is still much cheaper, or free, as Sun is suggesting.
I do, however, like buying computers and software outright as they are assets which I feel I should be able to buy once and use as many times as I want, like I do my TV, my stereo, my microwave. If kept going to the same place on holidy, maybe I'd buy a house there. Be nice if I at least had the option, wouldn't it?
You still have the option. The difference is, that soon we will also have the option at getting the hardware for "no extra cost." You can buy and I can rent, or vice versa, so we are both happy.
You may have a point in what you say, but next time I'd at least make sure the examples you use are relevant.
Yes, I do have a point, and despite your implication to the contrary, the examples are relevant.
I don't think the free market, specially normal consumers, will like subscription based goods. They want to pay once and then own the thing they paid for, not pay all the time they use it. Even if you have to "buy it once again" every few years.
Yeah, I think you're right. Nobody will rent a hotel room, get a cell phone plan, subscribe to a cable or dish, subscribe to a magazine or newspaper, buy utilities or subscribe to an ISP.
It is much cheaper and better to purchase a house every time you travel, build your own cellular network, make your own premium TV service, write your own newspaper and magazine, build your own power plant/waste processing plant/water treatment plant/landfill and roll your own ISP.
Slashdot Mirror
I appreciate where you are coming from and I cannot comment on AIX or HPUX, but I can comment on Solaris. Sun has been fascist about binary backword compatibility.
I wrote a 32 bit SCSI target driver in the early-90s using the (then deprecated) Solaris 2.3 ABI and it worked flawlessly until the early 2000s when we were required to switch to a 64 bit kernel, a switch we had delayed since the 90s.
The company still runs a version of Oracle compiled for Solaris 2.2 (1992?) and it works flawlessly, except it has a bug in its use of log archiving that was masked by earlier versions of the UFS. They still run the 14-year old 32 bit Oracle binaries on a 64 bit Solaris 8 box.
Backword compatibility is critical in the enterprise, and honestly, much of the OSS world ignores this.
Number of rewrites on solid state storage: ~1 million.
Number of rewrites on a laptop hard disk: Until the drive mechanism dies.
Hope you don't do a lot of swapping on your solid state flash hard drive.
Why does this myth refuse to die? These do "wear leveling" which moves the writes around the flash and means that you would need to write the whole drive one million times.
Let's do some math. One million writes of 32 GB equals 32,000,000 GB, or 32 PB.
Suppose you average 10 MB/s of writes the whole time your laptop is in use (good luck pulling that off). You would have 3.2 billion seconds of use, or 101 years of continued use.
Let's see a hard drive take that kind of pounding.
The trend is already happening.
- batteries. You can run a heavy, hot laptop for about 2-3 hours before needing a lengthy recharge. You can run a handheld PDA for roughly the same amount of time;
A Nokia 9500 runs a full workday (and then some) on a single charge. Each newer device gets better, and soon all devices will carry enough juice for the average Joe.
- data input. The desktop/laptop has this amazing invention, the full size keyboard, that lets us enter tons of information more quickly and accurately than any other method.
While I agree, there are plenty of applications that do not require a full-size keyboard.
- storage. Desktops start at 40G of permanent storage and go up to terabytes. Nothing else can compare. What's more, our storage needs are growing, not shrinking. We're not going to switch to Pocket PC/Phone/consoles that have maybe a 10G memory card or a 30G hard disk and give up our 250 giggers.
This is the trend you miss. Flash storage is going up at a much faster rate than storage appetites. Flash storage is doubling annually. At some point in the next few years, flash will provide enough storage for most people's personal data.
- connectivity. A desktop is on DSL or Cable or T1 or dial-up and is a reliable way to access the internet. Handheld devices have to be in range of a wireless hub or in network for cellular connections. The widely available connectivity for broadband handheld devices simply doesn't exist yet.
You hit the nail on the head: yet.
As storage is further miniaturized and as voice input and battery technology improve, we will doubtless see a displacement of casual desktop/laptop use with handhelds, such as Blackberry-style email reading and Palm/PPC-style organizer functions, but for heavy lifting, the desktop will remain.
And each year, the number of applications that fall into the category of "heavy lifting" will diminish until the number is so small that no one really cares.
There will, of course, be resistance from the old school crowd, just like there is with any new, outsourced service. The advantages to this are mind-numbing, including:
Your provider upgrades the software/hardware in one place, and they get the joy of maintaining it, backing it up, ensuring availability, etc.
All you have to do is make sure the desktops are suitable, pay a per-desktop/per-month fee and everything is done. Your company can go back to focusing on whatever its core business is.
This transition is inevitable and has already started.
These guys might disagree with your assessment of it being for "dinky" apps.
At the risk of turning this into a Postgresql discussion, Yahoo! has some serious data integrity issues. Often, checking your e-mail does not cause the "new e-mail" indicator to go out for up to a minute.
This is just an annoyance for Yahoo!, but would be the death-knell of any banking or inventory management application. Imagine a bank where transactions were delayed and posted in a random order. With a constant transaction flow, there would never be a moment in time when you could actually see what the status of the bank accounts are -- scary stuff.
Parent is absolutely correct about Symbian, which has already dominated the smart mobile device market (PDA and smart phones).
Look at the numbers and you will see that Symbian twice outsells Microsoft and Palm combined.
The whole Palm vs. MS debate is like Wendy's fighting Burger King and both pretending McDonald's doesn't exist. Why yes, I did post this with a Symbian phone.
I have run a Solaris system since the 2.3 days.
If you are deploying a system that will be around for several years, Sparc Solaris lets you upgrade hardware and software with minimal hassles and downtime.
Try bringing a 1998 Linux install in a 24x7 environment and upgrading the disks, CPU, memory, mobo, networking, the OS, apps and libraries without breaking everything that has been tacked on to the system in the past 7 years. This is done all the time with Solaris.
If the lifespan of your app/server is only 18 months, then none of this matters.
Banks make good money selling your financial information to "related buisnesses".
Sorry to burst your bubble. I work closely with dozens of banks and credit unions on this very topic. GLB inspires more neurotic fear in bankers than anything else I have seen in some time.
By the way, "related businness" means sharing information with other companies that must be there to support the bank, like disaster recovery companies, records archiving companies, etc.
Whether or not you believe it, "related businesses" simply cannot use your information for anything other than performing their service for the bank.
The closest a bank can get to profiting from your personal information is using it to offer services. A bank may notice that you have a high credit card balance and offer you a HELOC, it may notice that you have a high savings balance and offer its CDs, it may notice that your car loan is getting paid down and offer a pre-approved loan for a newer car, etc.
Other than that, your information is strictly off limits.
Then how is it that
IE is part of the Windows Operating System so that parts of the OS and other applications can rely on the functionality and APIs being present.
These two statements seem to contradict each other.
Not necessarily. I read those comments as:
That did not make me a sick pervert: I'm a engineer now
The distinction between sick pervert and engineer being what?
Ducks...
Parent is correct. Has anyone on slashdot ever tried to install unsigned software on a Symbian device?
It is littered with warnings and confirmation screens. Anyone who got this virus had to endure the installation process confirmations. It is worse than a EULA.
I find that I lack sympathy for a user who repeatedly selected "ok" and "continue" after being warned that this software cannot be verified -- software that arrived unsolicited.
It takes a whole new kind of inattention to allow this virus to spread.
I mean really.
According to the numbers, http://www.canalys.com/pr/2005/r2005012.htm smartphones sell more each quarter than PDAs sell all year.
The smartphone market is still young and growing at a blistering pace, while the PDA market is in decline.
Mod parent up.
The real advantage to this is when you need the headroom. If 5% of the time your organization needs massive computing power, then you need to maintain that overhead the other 95% of the time when you don't need it.
With a utility solution you pay only for what you use, but have the headroom to spike when you need it. Very cool.
Commercials. If advertisers know their audience won't see their commercials, they lose incentive to invest in advertising, and the networks lose money.
You may be on to something here. They probably have a hard time selling commercial space on the internet if they cannot say who is watching what. Nielsen ratings allow the networks to sell ads based on hard evidence of viewership, but...
What if Tivo teamed up with a network to offer shows via broadband, complete with ad injection? You could watch what you want regardless of your local lineup. The network could get feedback from Tivo on what people are watching which shows, broken down by demographics.
The only downside to this that I can think of is that a network trying this out would alienate its current last-mile providers: TV stations, cable or satellite companies, which have probably demanded some sort of exclusive.
Supporing internet downloads might be a contract violation or worse, piss off the last-mile providers enough that they drop the network completely, leaving the network only with the nascent internet market.
Now that I think about it, I can see why they just want the internet thing to go away.
I'll feed the troll...
This is not a winner. My cell phone is used to call people and for them them to call me. It is not a PDA, that is a separate device. It's also not a web browser, a camera, nor a music player. These devices are separate for a reason: so I can use them all only when I want, so I can upgrade them when I feel like, and if one of them breaks (or gets lost) then they are not all gone. Also, I can buy the individual devices much more easily because I'd buy only 1 a month, rather than having to buy the very expensive all-in-1 device; and who says they'll even have all the right features anyway?
If you don't want one of these phones, simply don't buy one. There are plenty of "basic phones" that do what you want.
The problem for the carriers is that they are trying to preserve existing revenue, increase revenue per customer and add new customers.
There is a segment of the market shouting, "Just give me a basic phone that works everywhere at a decent price and you will have my business."
The problem with this group is that they don't help the carriers reach their growth objectives. The only way to attract these folks is to lower prices (killing margins) or install many more towers (killing margins). Worse, there is no loyalty in this group. If the carrier does manage to attract them, then these subscribers will defect en masse to the next carrier that offers slightly lower prices or slightly better coverage.
No wonder carriers don't cater to this crowd.
it's been around for years and it cost under $2 a ride
I love these posts from the "mass transit cures all ills" fanatics.
Let's see how well some of us can operate with a bus (assuming it goes anywhere near my home and work, which it does not):
* Drop off and pick up kids at school and daycare?
Nope.
* Get groceries for my family of six?
Riiiight. I will just take the shopping cart brimming full with 12 bags of stuff up the small bus stairs and sit on my seat.
* Rush to get a sick child from school?
No.
* Take same sick child to doctor?
No.
* Pick up dry cleaning?
No.
* Take child to evening school event?
No again.
* Parent - Teacher conferences?
No.
* Take child to soccer? To his friend's birhtday party? To get a haircut?
No again.
* Visit a client?
Nope.
* Work late?
No.
* Out to dinner with my wife?
No.
On and on...
So, I guess if I had no family and no clients, lived alone, worked at my job only when the bus was running and could buy groceries only for me, then yes, a bus might make sense, if it came near my work or home.
Other than the above minor limitations, you are right, the bus is for everyone.
Their revenue stream and rights to artistic integrity end when they reach the consumer. At that point it's my signal in my device and as long as I don't rebroadcast it they need to leave me the hell alone.
While IANAL, I suspect that if you engage in activity that breaks their business model, they are well within their rights to exclude you from their business.
Movie theatres usually lose money on movie tickets and recover the money from the concession stand. Theatres can (and do) legally restrict patrons from bringing in their own food.
Some websites (slashdot?) give away the web service and make money from advertising. Some websites can (and do) legally restrict bots from their site.
The list goes on and on...
Your estimate should actually be five to six weeks.
Yeah, I realized after I had posted it that I had an order of magnitude error in my math.
It doesn't change much, though, because many devices will withstand 1 million write cycles instead of the 100,000 I used in the calculations, giving the same result.
But a hard drive needs a FAT or something similar, which is generally going to live in one spot on the disk-- you're going to hit the max writes on that segment fairly quickly.
IIRC, the cards use a level of indirection so that the same sector gets put in different physical places. You may write to sector 11843 over and over, but it actually gets written to a different physical spot each time.
I thought the problem with flash was a limited number of write cycles (10,000-100,000?). With this thing rated at up to 25,000 IOPS, is would seem that they might not last all that long (4 seconds?).
That may be true, but modern stuff does "wear leveling" which spreads out the writes. In essence, you would have to write the whole card 100,000 times. For a 1 GB card, that's 100 TB of writes. At a sustained 30 MB/s of writes (which is probably very rare), it would take three million seconds, or one year of constant writing, just to kill the card, which brings us to...
It should be fairly simple to put together a "CF-raid" drive for way less than $1K/GB.
I am amazed that this has not happened yet. For applications that do lot's of critical writing in tiny blocks (think Oracle log files), this would, speedwise, destroy any spindle-based competitor at a low cost.
The truth is that flash technology is increasing at a rate faster than the needs of typical server computing, which makes it inevitable that these will take over the datacenter some day.
That's like a car manufacturer saying, "We've installed a mechanism which will keep you from opening the hood if your intention is to upgrade the engine, because we want you leasing and buying new expensive cars very soon."
Or it is more like a car manufacturer saying, "The engine is tuned to handle X. If you put a blower or nitrous on it and ramp up the output, you are on your own, no warranty. If you buy a 200 HP Hondaccord, we will ensure that it has the actual 200 HP engine, and not a modified 120 HP engine that will fail sooner."
I would bet this is only to stem companies selling overclocked boxes and advertising them at the overclocked speed. Neither Intel nor the customer wins in this scenario. They are probably just stopping fraud.
Languages aren't "faster" or "slower"
Thank God, finally someone understands! I was recently benchmarking Turtle Logo against some highly optimized Opteron-specific assembly code for an application that calculates PI to the 400,000th digit.
I was surprised to see that Logo was slower, but I think I just didn't optimize it enough. I am tuning the Logo floating point code right now, and Logo should stomp the hell out of the assembly code "Real Soon Now."
Mod Parent up, it is dead on!
There are lots of examples where a converged device was horrid. There are lots of examples where an independent device was horrid.
Where there are synergies from convergence, things converge, and always have.
To the crowd that says, "Just give me a cellphone that works!" I tell you that there ain't enough of you to drive phone purchasing trends, so you will be ignored. Companies (rightly) focus on what they think will drive the bottom line.
Imagine hearing from someone, "All I want is a frickin' car! I don't need a recliner, a stereo system, HVAC, a roof, attractive paint, seat belts, air bags, electric starter or any of that other frilly crap. All I want is a car that goes from A to B." You, my friend, are in the minority and always have been.
Personally I don't subscribe to a newspaper. I bought one on Saturday though, and as far as I'm aware I can keep as long as I want and read it as many times as I want without having to pay anybody anymore money.
Well said. You could also buy the latest DVD of your favorite movie, or rent it for far less.
I subscribe to cable, my cell phone, power company, etc because I get continuing service from them which I can't buy outright as it isn't an asset. And last time I went on holiday I stayed in a hotel as it was much cheaper than buying a house and then selling it again at the end of the week.
Another good point. Yes, you could buy all of those things, but renting it is still much cheaper, or free, as Sun is suggesting.
I do, however, like buying computers and software outright as they are assets which I feel I should be able to buy once and use as many times as I want, like I do my TV, my stereo, my microwave. If kept going to the same place on holidy, maybe I'd buy a house there. Be nice if I at least had the option, wouldn't it?
You still have the option. The difference is, that soon we will also have the option at getting the hardware for "no extra cost." You can buy and I can rent, or vice versa, so we are both happy.
You may have a point in what you say, but next time I'd at least make sure the examples you use are relevant.
Yes, I do have a point, and despite your implication to the contrary, the examples are relevant.
I don't think the free market, specially normal consumers, will like subscription based goods. They want to pay once and then own the thing they paid for, not pay all the time they use it. Even if you have to "buy it once again" every few years.
Yeah, I think you're right. Nobody will rent a hotel room, get a cell phone plan, subscribe to a cable or dish, subscribe to a magazine or newspaper, buy utilities or subscribe to an ISP.
It is much cheaper and better to purchase a house every time you travel, build your own cellular network, make your own premium TV service, write your own newspaper and magazine, build your own power plant/waste processing plant/water treatment plant/landfill and roll your own ISP.