Slashdot Mirror

Music for a phone keypad

and another

it's full of them.

No, that's not how MTBF for hard drive is calculated. And that's also not the "average" hours of an "average" drive can expect if someone gets an "average" hard drive.

See here for an extensive discussion, http://forums.storagereview.net/index.php?showtopic=18811 I'm sure there's more info in that forum.

The problem with the hard drive MTBF is the way the manufacturers measure and come up with those numbers.

I'm pretty lazy, but you can search around or go here http://forums.storagereview.net/index.php to find out how they exaggerate the numbers to ridicules proportions.

Besides, a lot of their forum regulars are fanatical Windows supporters, to the point of claiming "...in smp scaling windows is better than linux, and linux is better than freebsd. for vms windows is better than freebsd, and freebsd is better than linux.", so I'd take what they say with a grain of salt.

not as much as you might think.

HDD Power consumption

IDE/SATA drives only draw about 7-13W Idle/read&write, 15K SCSI drives a bit over 20W read&write.

Spin up might be a problem, but I'd assume you'd want to use cards that supported staggered sinpup on a setup that large.

So, yes 16 HDDs can pull quite a bit of power, about 300W for top end SCSI solutions. Though you wouldn't be thought of as particularly bright if you entrust a setup like that to a basic quality desktop PSU. And the quality of supplies you'd be using with a high end storage array like that (ie something in the N+1 redundant Zippy line) have been availible at well over 1000W for a while.


I think a 1000W PSU in a standard EAXT setup is massive overkill. I really have a hard time thinking of a workstation / stand alone server setup that would be too much for quality 500-600W PSUs to handle right now.
Anandtech reviewed a 4 CPU dual core Opteron setup from SUN while back, it only drew about 600W.

Not everyone agrees with PC World that Earthlink is the best, even for the average joe (but then, what's new?).

I actually thought that this was common knowledge--that Windows Server 2003 with ASPX was faster than Linux/Apache with PHP, or that Server 2003 was generally faster with static content. (I admit, I only glanced over the article, and Adobe Acrobat's search tool is the worst of crap, so sue me if it didn't mention ASPX).

1)ASP (not ASPX) are fairly flaky and recent versions are roughly comparable to, but slower than, PHP4 (not sure about 5), in general.

2) Windows is not very good at creating new processes quickly. This is why CGI (not fastCGI) in the platform is so glacially slow.
Let's have an example. Let's say that you make a dynamic webpage in which all content is generated by a C++ CGI program. Ignoring database access for the time being, since that dilutes the example, on Windows, the website would be MUCH slower than the same website written in ASPX, even though the actual execution time of the C++ program is shorter (assuming a competent C++ coder).
This is because for each request, Windows must create a new process (the CGI program), and destroy the process when the request is complete.
While the execution time is low, the process management overhead dwarfs the actual page runtime, because Windows doesn't do that sort of thing quickly. This is why CGI has long been blacklistedon Windows systems by good web devs, and this is one reason that Apache 1.x was such a dog on Windows. Apache 1.x creates a new Apache process for each request.

Now Linux, on the other hand, creates processes about as fast as it creates threads, which is to say, really damn fast. Apache 1 has always worked just fine on Linux (and indeed most Unix systems) because the overhead of creating a process, while significant, isn't slower than a dead slug stuck in frozen molasses like it is on Windows.
Apache 2.x allows requests to be served by a thread or a process, or a number of processes that each create several threads (any Apache gurus please correct me if any of this is off).
It follows that this isn't a big deal on Linux (because process creation isn't really much slower than thread creation), but is a very big deal on Windows.

Windows has ASPX, which is Microsoft's marketing term for the use of the .NET framework for web content delivery (get it--the 'X' makes it sound cool. Or something). .NET is compiled, and ASPX needs neither process nor thread creation. Like any .NET application, ASPX can run sort of close to native speeds (native + lots of wrapper overhead + generic memory management overhead and such.)

Yet Apache is still back here creating a process or thread for each and every request (note that there are some ways to speed things up. FastCGI comes to mind, but I don't want to get into the gory details that I don't know enough about). This is not the brightest way to do it in terms of performance, but then, Apache appears to have been designed for universality and configurability over raw throughput.

It is unwise to hold the attitude that Apache can't be beaten by IIS, especially when IIS is optimized for one platform--by the vendor of that platform. Apache isn't even the fastest on Linux. Take a look at Zeus webserver. It serves circles around Apache on any platform it supports--including Penguin land.
In fact, Zeus uses a technique called SendFile() which, oddly enough, is strikingly similar Microsoft's own TransmitFile() API. Hmm.

Think of it this way: Apache is to IIS as GCC is to ICC, at least in terms of performance and generality.
Intel's compiler (ICC) consistantly blows away GCC in terms of the performance and size of the compiled code, but GCC runs on just about anything with a CPU, can cross-compile, is free, doesn't pull any PHB evil tricks, and actually compiles things like the Linux kernel without pat

So AMD and some other companies did some benchmarks with unequal systems. And?
It would certainly have been less misleading to use an Intel laptop with a similar video card at least, but it isn't like you are going to see these benchmarks on television ads or anything. I doubt any big buyers will care about benchmarks of pre-release products anyway.

Granted, this is rather shady of AMD, but it doesn't even approach the raw evil of, say, a company joining BAPCO and systematically removing all benchmarks in their Sysmark tool in which AMD wins. No, that would be unprecedented in this industry's history.

It's also noteworthy that TheRegister has a partnership with Tom's Hardware in the U.S., and some editors of Tom's have been noted as being overtly biased towards Intel, though Tom's itself seems to be getting better, having articles like the used to--real tech info rather than the sensationalized, poorly written crap which had infested my once favorite hardware site.

Granted, both AMD and Intel are "evil" for-profit companies, but something like an unfair benchmark hardly brings tears to the eye when you consider some of the staggering bullshit actions of the past.

What it all comes down to is preference--The Turion is going to be a 64-bit chip (isn't it?) with the benefits of AMD64 mode (most of which involve the fact that it has double the general-purpose registers in the chip, and not from the fact that those registers are 64-bits wide). The Turion will likely outperform the Pentium-M in most test, like the Athlon64.
The Pentium-M, however, will perform just fine thankyou, and will drain less battery power and thus be in cooler-running laptops with better battery life.
I'd pick the Pentium-M myself, since to choose a product based on anything other than overall effectiveness/price ratio set is usually either fanboyism or poor research.

Let's google it : 'amd dual-core TDP

So that's 95W max TDP for the fastest AMD against 130W 'typical' TDP for intel's. That's abour 50%-100% more depending on the exact configuration & performances.

There's an interesting (as far as "new drive is bigger than old ones!" is interesting) thread on Storagereview.com which includes some insights as to how this thing is built, and why it uses lower-capacity platters than even Seagate's 400GB drives.

Itanium is being pushed increasingly into higher end computers. You know why Itanium is important?

Power970 cpu limit: 2-4 cpus
Opteron cpu limit: 8 cpus
Itanium cpu limit: 512 cpus.

For example that 2nd ranked "top500" computer is a 20 machine Beowolf style cluster. Each machine has 512 cpus.

What's next? Dual Xeon laptops with a car battery backpack accessory? :)

If this is real which is doubtfull it is probably a marketing trick. The drive manufactures proably make one drive and sell it as 3 different drives in different capacities.

Actually, this is exactly what they do. The difference, however, is that the lower-end (smaller) drives are identical except that they come with fewer platters. For example, a 160GB hard drive today likely has two 80GB platters, whereas an 80GB drive probably has one (though different combinations of different sizes are of course used, depending on when the hard drive was manufactured and other factors)

In some cases, a hard drive will be sold with a greater potential capacity than its available capacity. For example, a drive with two 60GB platters may be sold as a 100GB drive, the platters having been "short stroked". This has nothing to do with the absurd technique described in the Inquirer article, and I doubt that it is possible to recover the lost space.
Hard drives are the highest precision mechanical devices that most people have in their home--moreso than processors, high-end printer heads, or toasters. They are not something that you want to physically modify.

See the following highly informative and interesting (if you are a geek) posts by a Maxtor engineer:
Here
here
and here

If this is real which is doubtfull it is probably a marketing trick. The drive manufactures proably make one drive and sell it as 3 different drives in different capacities.

Actually, this is exactly what they do. The difference, however, is that the lower-end (smaller) drives are identical except that they come with fewer platters. For example, a 160GB hard drive today likely has two 80GB platters, whereas an 80GB drive probably has one (though different combinations of different sizes are of course used, depending on when the hard drive was manufactured and other factors)

In some cases, a hard drive will be sold with a greater potential capacity than its available capacity. For example, a drive with two 60GB platters may be sold as a 100GB drive, the platters having been "short stroked". This has nothing to do with the absurd technique described in the Inquirer article, and I doubt that it is possible to recover the lost space.
Hard drives are the highest precision mechanical devices that most people have in their home--moreso than processors, high-end printer heads, or toasters. They are not something that you want to physically modify.

See the following highly informative and interesting (if you are a geek) posts by a Maxtor engineer:
Here
here
and here

If this is real which is doubtfull it is probably a marketing trick. The drive manufactures proably make one drive and sell it as 3 different drives in different capacities.

Actually, this is exactly what they do. The difference, however, is that the lower-end (smaller) drives are identical except that they come with fewer platters. For example, a 160GB hard drive today likely has two 80GB platters, whereas an 80GB drive probably has one (though different combinations of different sizes are of course used, depending on when the hard drive was manufactured and other factors)

In some cases, a hard drive will be sold with a greater potential capacity than its available capacity. For example, a drive with two 60GB platters may be sold as a 100GB drive, the platters having been "short stroked". This has nothing to do with the absurd technique described in the Inquirer article, and I doubt that it is possible to recover the lost space.
Hard drives are the highest precision mechanical devices that most people have in their home--moreso than processors, high-end printer heads, or toasters. They are not something that you want to physically modify.

See the following highly informative and interesting (if you are a geek) posts by a Maxtor engineer:
Here
here
and here

Taken from the afore-mentioned StorageReview thread:

Find it here

Mod down!!

Sorry, but this is complete bullshit.
Did aureal density technology increase to 200GB/platter overnight? No.

Please refer to this thread on StorageReview.com for more information.

Fujitsu's MPG-Series HDDs experience a similar problem (>90% failure rate due to defective chips; see The Register).

Fortunately, there is an unofficial tool available to recover all of your lost data.

I know it's not supposed to be a low-end RAID-card (however, it IS :>) like the mentioned ones, but probably you also considered this one. Don't even think about it. I shared my experiences on StorageReview, click here to read it.

The main problem w/ the 3ware 8500's is that they're actually 7500's w/ SATA converters (actually a hair slower due to the conversion overhead), and still PCI 64/33. A whole slew of PCI-X true SATA RAID controllers (one from 3ware is supposed to be out late this quarter) will be coming out in the next few months. Should be interesting.

The sequential transfer rates being quoted are likely RAID 5 for reads. It's not unheard of for modern U320 arrays to top 500MB/s in sequential reads and >20,000 IOPs. (the 3ware 7500-12 clocks in at 190MB/s and ~700 IOPs according to their benchmarks.

(note that PCI 64/33 maxes out at 264MB/s, PCI 64/66 at 528MB/s. With the recent 2.0 specs, PCI-X will take you up to a ridiculous 4.2GB/s)

For major speed, check out this Ram San. 700MB/s bandwidth and 200,000 IOP?!?! Yowza!

Microsoft Works.

Thank you, thank you. I'll be here all week.

I wrote this. I feel that the topic of hard disk humor is largely, and tragically underappreciated. :)

It was because of recurring problems with IBM drives that I ended up replacing them with Fujitsus about six months ago.

My system is all SCSI, all the time. As a result, I end up paying in the neighborhood of $200 for an 18G drive. With prices like that, failure is simply not acceptable. Some people say that all hard drives are crud and are going to fail, so one should simply plan for it. Well, then why are the hard drives in my 12-year-old Amiga still working fine?

After enduring my most spectacular failure to date, I resolved to change drive brands. A couple of years prior, I upgraded the drive in my laptop computer. The first drive I tried was an IBM Travelstar, and it made the most gawd-awful racket. I could hear the thing two rooms away over the fans in my main rig. So I sent it back and took a Fujitsu instead. It's been perfectly quiet and reliable ever since.

After this happy experience, I decided to put a couple of Fujitsu MAN3184MP SCSI drives in my main rig. So far, they have given me no trouble at all.

I can't imagine what the heck's going on in the hard drive industry to cause so many failures. I can only hope one of the manufacturers will spill the beans at some point.

Schwab

FYi this information has been on the StorageReview.com forums for about a week. There is a small discussion there.

Fluid Dynamic Bearing (FDB) motor ... Fujitsu got that working in 2000 "They've debuted fluid bearing motors in their MPF series. First introduced by Seagate in its 7200 RPM Medalist Pro, fluid bearings rapidly faded away as problems surfaced from the heat caused by leaked fluid."

As for the noise, the Fujitus MPD3084ATs I got in 1999 had very little. Oh well :)

Fujttsu may not be performance beasts, but for reliability, coolness, and quietness, they took the crown long ago.
--

Fluid Dynamic Bearing (FDB) motor ... Fujitsu got that working in 2000 "They've debuted fluid bearing motors in their MPF series. First introduced by Seagate in its 7200 RPM Medalist Pro, fluid bearings rapidly faded away as problems surfaced from the heat caused by leaked fluid."

As for the noise, the Fujitus MPD3084ATs I got in 1999 had very little. Oh well :)

Fujttsu may not be performance beasts, but for reliability, coolness, and quietness, they took the crown long ago.
--