Flash Drives in Future Apple Laptops?

danscript writes "Samsung hopes that falling prices for flash-memory chips will mean solid-state memory can eventually replace hard-disk drives in Apple PowerBooks and iBooks as well as other devices, Macworld UK is reporting. The benefits? - silent; less power; reliable and faster."

Comments

Flash

[Zlib pt]

"I'm sorry sir. You can only install OSX 10 times. Then you ran out of read/write operations"

Re:Flash

[FidelCatsro]

That's not really a worry any more , modern flash memory has a substantially greater number of read/ write cycles.
IIRC the numbers are good enough that they would probably live as long if not longer than your average laptop HDDs .

Re:Flash

[advocate_one]

IIRC the numbers are good enough that they would probably live as long if not longer than your average laptop HDDs.

the only people who might possibly run into the write limit would be Gentoo users...

Re:Flash

[Daniel Dvorkin]

While I'm sure that there are a number of different engineering approaches underlying the development, it seems to me that one of the requirements for anything to be called "flash memory" is (or ought to be) that it doesn't require continuous power to maintain its data. I certainly hope they're not talking about battery-powered RAM, because I'd really hate to lose everything on my main drive just because a battery wore out.

My impression is that the speed of USB thumb drives and hard drives is about equivalent, which leads me to think that USB (even 2.0) is the bottleneck, not the drive itself. But determining this would take rigorous testing, of course. Certainly the potential speed of solid-state devices is much, much higher. The hoops that hard drive engineers have to jump through these days to get acceptable speed relative to the rest of the computer are just insane.

Re:Flash

[TCM]

That's true if each block of the disk was written equally often. AFAIK, unless you use special file systems, that's not the case with common file systems today. Let's say you write one file once and only read it from then on. The access time would get updated each time, wearing out a single block faster than the rest.

Re:Flash

[WonderSnatch]

I'm guessing these drives would have a sort of "wear leveling" just like they have in most compact flash cards.

The wear leveling works by keeping a table of what physical flash is mapped to what address. The trouble comes when power is yanked whilst the table is in the middle of an update.

Brett

Re:Flash

[MaynardJanKeymeulen]

Thumb drives are slower, let's do a test:

[root@pentie4 /]# hdparm -t /dev/sde2

/dev/sde1:
Timing buffered disk reads: 78 MB in 3.04 seconds = 25.69 MB/sec
[root@pentie4 /]# hdparm -t /dev/sde1

/dev/sdc1:
Timing buffered disk reads: 36 MB in 7.43 seconds = 4.84 MB/sec

with sde my USB2 harddisk and sdc a USB2 thumbdrive.

Re:Flash

[WonderSnatch]

In fact, here's a white paper from Sandisk.

Brett

Re:Flash

[alienw]

Not true. Modern Flash cards have wear leveling built into the hardware. The address the computer sees is actually the logical address that the controller translates to a hardware-specific address. The controller automatically adjusts addresses to spread out the writes and to detect and remap bad blocks (this should occur without loss of data, since blocks only go bad when you write to them).

Re:Flash

[Paradise Pete]

last time I chekced my USB flash drive was a shitload slower than my hard disk. ;)

I don't think you can measure speed in shitloads. Say, you didn't have anything to do with that Kessel Run timing, did you?

(Yeah, I saw the smiley.)

Re:Flash

[Shanep]

What's happening Flash?

Gordon's alive!!!

Re:Flash

[Just Some Guy]

Thumb drives are slower

...at sequential reading. Wanna run the same test with bonnie++ or another benchmark that slams the drive with random accesses? I'll bet the near-zero seek penalty on solid state media makes up for quite a bit of its currently mediocre sequential access.

Imagine a database where you're writing millions of tiny blocks of data all over the place. Within reason, fast seeks are about as important as fast IO.

Re:Flash

[RovingSlug]

Which USB stick do you have? The Lexar JumpDrive Lightning reads at 22 MB/s. ArsTechnica review.

Re:Flash

[tji]

Maybe this is the reason that they are pitching it as a HDD replacement... Flash has been around for a LONG time, but it had major limitations.

From the article: The SSD's performance rate exceeds that of a comparably sized HDD by more than 150 percent. The storage disk reads data at 57 MegaBytes per second (MBps) and writes it at 32MBps.

A reliable drive, at 40+ GB sizes, with that kind of performance would be great for laptops.. Silent operation, low power usage, potentially more resiliant (no head crashes). Bring it on.

Re:Flash

[Hal_Porter]

Here's how it works on NOR flash.

The flash is broken up into Erase Units, which as the name suggests is the smallest block you can erase at any one time. IDE hard disks have a small fixed sectorsize of 512 bytes, smaller than one EU.

Imagine a 12KB flash with a 2KB Erase Unit and 6 units. One of these is bad - this can seen by the absence of a metadata signature for example.

You could use this as an IDE disk with 6KB capacity, since some of the erase units needs to be spare at all times.

Each EU can hold 3 IDE sectors, some metadata including the signature, and then 3 integers saying which logical block the physical block holds.

The EU starts off erased, all ones. Let's write a fileystem,

EU 1
lookup table contains (1,2,3)
IDE sector 1
IDE sector 2
IDE sector 3
EU 2
lookup table contains (4,5,6)
IDE sector 4
IDE sector 5
IDE sector 6
EU 3
lookup table contains (7,8,9)
IDE sector 7
IDE sector 8
IDE sector 9
EU 4
lookup table contains (-1,-1,-1)
spare - every bit in this EU is one
EU 5
lookup table contains (-1,-1,-1)
spare - every bit in this EU is one
EU 6
bad

Now imagine we need to update sector 1. We can use one of the spare blocks to store the data. The old copy can be left, but marked as unused by setting its lookup table entry to 0 - programming all the remaining bits.

EU 1
lookup table contains (0,2,3)
IDE sector 1 (obsolete data)
IDE sector 2
IDE sector 3
... no change, damn lameness filter
EU 4
lookup table contains (1,-1,-1)
IDE sector 1 (latest copy)
spare (all ones)
spare (all ones)
... no change

Now imagine that you need to write the next two blocks, 2 and 3. That way all the blocks in the first EU will be obsolete, and you can erase it.

EU 1
lookup table contains (-1,-1,-1)
spare (all ones)
spare (all ones)
spare (all ones)
... no change
EU 4
lookup table contains (1,2,3)
IDE sector 1
IDE sector 2
IDE sector 3
... no change

If you're not lucky enough to get all the blocks written at any one time, you need to compact by copying the block with the most obsolete blocks into one of the spare EU's. Obvously you can skip the obsolete blocks - you just copy the ones that are used and mark the rest as spare.

So far I've talked about one lookup table, the one in the flash which gives you the logical block which each physical block contains. This is the inverse of the lookup table you want for reading an arbitrary block, but this inverseness gives it the useful property that it only needs to be updated only once per erase cycle - some of the bits in an entry are programmed when the block is written, and the rest are programmed when the data becomes obsolete.

If you want to read logical block N, it's useful to have another lookup table which gives you the mapping logical block to physical block. This will need to be updated a lot - once for each write of a block. But it can be generated from the lookup table in the flash at insertion time, and kept in Ram.

There are some corner cases obviously - like the bad bits in the metadata area or whole EU's going bad, but there's usually a fair bit of space in an EU for metadata, and a reasonable number of spare EUs. Plus, if the flash has always been used like this, the bits should all wear out at the same time, which is obviously not something you can work around. But the trick to avoiding the problem you mention is to store an _inverted_ lookup table on the flash.

Re:Flash

[big daddy kane]

theres also yaffs which requires less memory and runs faster on NAND flash chips, which is the type of memory that would be used for that kind of storage. NOR is slower has larger pages, it's usually used as ROM chip.

Won't Happen

[Anonymous Coward]

Now that the idea has been leaked Apple won't do it even if they were planning to.

Re:Won't Happen

[rednip]

No No you misuderstood. Someone leaked that they were NOT going to swich to intel so they had to

Well, I've heard from a 'very' reliable source in Apple, that they will not be giving a free notebook to anyone with the Slashdot nick of "rednip".

Too Bad pn Junctions cost more than magnets

[Metaphorically]

I remember talking to a guy at Radio Shack about flash-based drives and how this was going to be the new option back in 1992. I think they were calling it a "hard card." Looking back, it was probably the same thing as PCMCIA Flash drive. That's the precursor to Compact Flash cards for you young'uns.

It wasn't new then and it isn't new now. Is it time? Sure. It's long overdue and I'd love to see solid state drives suddenly become financially feasable.

I doubt it's going to happen though because it seems like the cost of the magnetic materials used in disc platters will always be low and a solid state memory cell (flash, ram, eeprom, whatever) takes a couple transistors. The price of both drops, but hard drive price per GB (or MB, TB, whatever) always drops faster because of the lower transistor count.

Re:Too Bad pn Junctions cost more than magnets

[yttrstein]

SCSI, SATA and ATA controllers found on each and every hard drive made take quite a bit more than a couple transistors.

Re:Too Bad pn Junctions cost more than magnets

[KD5YPT]

The point he's making is that the controller on harddrive have a relatively fix amount of transistor, regardless of the harddrive's space. Flash mem, however, needs a set of transistor for every bit of data it stores.

Re:Too Bad pn Junctions cost more than magnets

[FidelCatsro]

Sometimes good tech just takes years to catch on and neigh on useless legacy crap takes forever to die off. Just look at the fact that FDDs are still clinging on tooth and nail in the PC world.
LPT ports are still used for printers , serial ports still come as standard on most motherboards .

I think the principle advantage is in the power consumption for laptops, Batteries are really expensive , if you use your Laptop on the road a lot then it has two principle advantages , lower power consumption and its far far more robust to damage due to things like dropping .
In this field of the ultra compact laptops i would say that the Power consumption and durability will outweigh the per GB cost .
If the write/erase cycles are now within the realms that samsung is considering offering these drives for commercial sale then i would defiantly want one in my laptop , i could easily cope with 15GB on a portable device for work.

Re:Too Bad pn Junctions cost more than magnets

[pedantic bore]

No, the "hard card" was a self-contained hard drive that could be installed, like a peripheral card, directly into a mobo slot. Nice idea for old cases that didn't have expansion drive bays. Saw lots in stores, but never saw many in actual computers...

On the other hand, you might be thinking of bubble memory... which was going to be the next big thing in the early 80's. Still waiting on that one.

Re:Too Bad pn Junctions cost more than magnets

[shine-shine]

The price of both drops, but hard drive price per GB (or MB, TB, whatever) always drops faster

This isn't to say that solid state drives won't become popular. If I can get a 80GB solid state drive for the price of a standard 400GB HD, I'd go for it. I think this is exactly what'll happen. As capacity grows, it becomes less and less important for people to have the largest HD on the market. Sure, many people have the need for large drives (video editing, pr0n, etc.), but most can do just fine with tenth the size of a modern HD. Especially when it comes to laptops.

Re:Too Bad pn Junctions cost more than magnets

[Metaphorically]

That's the one. I must be bad with names; bubble memory is what I call those flashbacks of what my paycheck looked like 5 years ago.

Still keeps me up some nights.

... if you listen close, you can hear the quiet sobbing

Re:Too Bad pn Junctions cost more than magnets

[epiphani]

You missed the boat.

Samsung to release 16GB IDE flash drive in august.

I want them in my servers.

Re:Too Bad pn Junctions cost more than magnets

[pete-classic]

Hard cards were the original IDE drives. Before hard cards you had to have an actual disk controller plugged into your bus, then cabled to the drive. IDE, in it's original implementation, was just an ISA slot re-configured as a pin header connector with some unneeded pins (such as all the IRQs except for 14 or 15) removed.

With all the advances that IDE has taken, it is still a simple interface, not a disk controller.

Hard cards didn't last long, but they're an important mutation in the evolution of the modern PC.

-Peter

Re:Too Bad pn Junctions cost more than magnets

[Cylix]

Serial is by far not even close to dead.

I deal with so much equipment that has a life cycle in the 30+ years that it's just amazing by comparison. So in another 10 years when more long lived technology has died off we might see the end of serial, but there is simply too much non-pc equipment in the world to do away with it.

Nearly every electronic requires a serial port or some form there of.

It's slow, clunky and definatelynot the newest thing, but it's a good standard and far too many things in this world require it.

Now, think about the devices in your home and ask yourself how many can do serial communications. (Hell even my cable box does)

Re:Too Bad pn Junctions cost more than magnets

[waynelorentz]

Back somewhere near 1989/1990 I cobbled together an XT from scavenged parts that included a hardcard. The thing was a full-length card and was very heavy.

Around the same time I also fell into a computer that used bubble memory. It was a GRiD Compass. I remember being amazed that I could unplug it and it would remember my word processing documents for months and months with no power. Back then, this was a big deal.

As an aside, the GRiD also had a label silk screened on the bottom of it listing the nine countries the computer was legal to be used in, and taking it to any other country was a violation of some law or other. That was one of many really weird things about that machine. I got it from someone tight with the Reagan administration who said they were standard issue for the Secret Service at the time.

faster writes?

[nblender]

They must be talking about some other kind of flash than anything I've used... I routinely rewrite 128MB-512MB CF cards for an embedded product and it's nowhere near the speed of a laptop disk. Maybe they're thinking some sort of RAM cache.

Re:faster writes?

[syukton]

What interface are you using? USB? ATA? other? Maybe you're encountering a bottleneck that has little to do with the card technology itself?

Re:faster writes?

[Metaphorically]

Write speed for flash is slower than read speed for flash, but making a direct comparison like that can be tricky because of differences in the technology.

Flash memory has to be erased before it can be written and has an inherent minimum erase block size. This has made designers put some write buffering in the card (at least in the older PC Card devices, which afaik are just bigger versions of CF cards). That write buffering is one layer of caching, then there can be additional layers of read and write caching in the driver.

Generally those caches improve normal use of the device, but in cases like yours you blow right through all those caches since you're rewriting the whole card.

btw, that caching is why you have to tell Windows to eject the device before you pull it out. Your write might not have actually finished even though you can do other things to the drive in Explorer.

Reliable? Don't think so.

[Anonymous Coward]

Show me a flash drive that survives a couple of million write cycles, and I might consider using a flash drive instead of a normal hard drive.

Re:Reliable? Don't think so.

[laffer1]

So you must not use swap space?

Seriously, its possible for a hard drive to have a lot of writes. Log files, swap space, patches, virus patch updates, etc. A lot of writes happen on modern computers. I've even read articles about people booting computers with flash memory as routers but they had to make sure /var and /tmp were memory disks because it would kill the flash memory in 3 months. (logs alone!)

Re:Reliable? Don't think so.

[Dolphinzilla]

here you go:

>5,000,000 Write/Erase cycles, unlimited Read

http://www.m-systems.com/content/Products/product. asp?pid=34

Re:Reliable? Don't think so.

[kesuki]

I've even read articles

What are you doing on /. then?

but seriously, prior to wear leveling (ie: antique 32MB cards, early controllers/drivers etc..) this was a valid statement, however cards back then only had 10k or fewer rewrite cycles. the rewrite cycles have gone up 10 fold, and wear leveling makes sure that data that is 'rewritten' is actually written on a portion of the disc that was previously determined to be a low write, freely alloctabale block, and they always keep spare low write sectors available. it's even possible for a hard drive such as this to Warn users of impending drive failure. BTW when a block on a 'flash' based drive 'fails' it falls back into read only mode. only the data it attempted to write must now be written somewhere else.

Also, to get back on track the MTBF for most platter based hard drives is something along the lines of 50,0000 hours for a really top notch drive, reading/writing the entire disc will on average take ~2 hours, requardless of capacity, because that's how long it takes the data heads to travel the entire platter surface area... so statistically, you have 25k write cycles before failure on a conventional hard drive. so how is that inferior to flash memory? and remember, this is APPLE Macintosh running of a FreeBSD derived kernel. put the right amount of ram in the laptop, and it will _never_ use swap space.

extra bonus

[justforaday]

The extra bonus: Apple gets to sell you a new one after 1,000 or so boots... : p

Re:extra bonus

[rokzy]

who is booting a Mac 1000 times?

you're obviously a Windows user.

Re:extra bonus

[justforaday]

Install OS 9 on it and it won't even last 6 months... : p

Has MRAM finally arrived?!?

[Rattencremesuppe]

From TFA:

The drives are also typically lighter and can read and write data faster than conventional drives.

AFAIK, flash memory reads data faster than a hard drive, but writing is slow as hell because of the long block erase cycles,

does samsung have a new technology for flash chips?
or do they eventually use MRAMs? :-D

Flash Buffer

[Lockelator]

It certainly wouldn't hurt to have a 1 gb flash buffer to lessen wear and tear on the HD.

Re:Flash Memory?

[Mr. Maestro]

And under certain other conditions, people have been known to burst in to flames. Oh wait, no they haven't. Come on, seriously, do you have ANY idea how many flash drives are currently brought on Airplanes? I've brought my digital camera, MP3 Player, and thumbdrive, all which have flash memory/drives. No fires.

Silence is golden, but is flash big enough?

[G4from128k]

We know that the black turtle-necked one hates noisy machines and I agree with him. I configured an old Powerbook 190cs to boot from a CF card in the PCMCIA slot -- wonderfully silent and much faster than booting from the HD. Of course on that old machine, the OS, a couple of applications, and some files fit nicely in only a 4 MB flash memory. In contrast, OSX, modern apps, and files will need 1024 times that space (4 GB) at a minimum and tens of GB if the person has even a modest collection of media files.

I can only hope that Samsung's technology roadmap (16 GB by 2006, 100 GB by 2008) is correct although I wonder how HD technology will have evolved over those same years.

18Gb prototype info here:

[Idou]

http://www.physorg.com/news4220.html

Flash still has lots of room to grow

[eldawg]

The recent Samsung announcements were for 16 gig drives for this year. Considering that most laptops are being pushed with 100+ gig HDs, Flash still have some ways to go.

Re:Flash still has lots of room to grow

[Anita Coney]

Not every one is concerned about massive drive sizes. There are plenty of people who would choose the battery saving advantages of flash drives in their laptops.

I'm of the opinion that laptops should be as small and energy efficent as possible. I just don't get the point of using them desktop replacements. If you want something as huge and powerful as desktop, buy a friggin' desktop. If you want something portable, buy something portable.

I mean, what's the point of a "portable" computer if you have to plug it in all the time?!

Re:Flash still has lots of room to grow

[BenjyD]

what's the point of a "portable" computer if you have to plug it in all the time?!

I agree that there is always a place for small laptops, but a desktop replacement laptop has many advantages:

• If you need to move office, just pick it up and go. I need to go to the US occasionally for work - when I do, I can just find an empty desk and plug in my complete development environment without sacrificing processor speed.
• One desktop replacement laptop is much cheaper than one decent desktop and one small laptop, and avoids the hassle of keeping two systems in sync.
• They are generally quieter
• They use less power

I use a 15.4" widescreen Pentium-M Sonoma laptop for all my work and it's great: 3hrs battery time, easily fast enough and I can work pretty much anywhere there's internet access.

Re:Flash still has lots of room to grow

[Ancil]

Considering that most laptops are being pushed with 100+ gig HDs, Flash still have some ways to go.

Unless you consider that most 100 GB hard drives are about 15% full...

Re:Flash still has lots of room to grow

[daikokatana]

The point of using a laptop as a desktop, in our case, is very simple.

You take it to work, work 8 hours straight (plugged in), take it home and you're able to continue to work. Easy. And because it's laptop, you can work during your commute as well.

Re:Flash still has lots of room to grow

[bwalling]

I mean, what's the point of a "portable" computer if you have to plug it in all the time?!

Given the complete scarcity of electrical outlets, I'd have to agree with you!

Re:Data recovery?

[Rattencremesuppe]

Flash chips have a interface similar to RAM chips (address / data bits, chip select, write enable, ...) If your filesystem is corrupt, you can still read the data contents byte-wise,

Well, of course he wants it to happen

[intmainvoid]

The most recent IDC data ranks Samsung as the world's number one producer of flash chips

So of course "Hwang Chang-Gyu, president and CEO of Samsung's semiconductor business" wants his company's technology to take over from hard drives. That's very different to Apple saying it will happen.

Two drives: one flash, one magnetic

[AdamInParadise]

I've been thinking about this for a long time. What about using a flash drive for the important stuff (OS+user docs) and a hard drive for the unimportant stuff (divxes, CD backups, you name it)? Basically, the hard drive would be powered down most of the time, bringing down noise and heat, therefore driving up the reliability of the whole system. That's certainly possible with every kind of computer out there, but it would be better with specific OS support. For example, the OS could transparently copy your data back and forth between both drives, like the iPod does (with RAM instead of Flash).

Regards

Re:Two drives: one flash, one magnetic

[morgan_greywolf]

Actually, you might be right.

I've got 1 GB of RAM on my Fedora Core machine and even though I use GNOME heavily and even use the machine to compile source (even kernel source) and I gotta say that the machine hardly ever has to go to swap. Right now I've got OOo, Firefox and a couple of terminal windows going and even opening files and such in OOo, this is my vmstat output


[me@host ~]$ vmstat 5 5
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
r b swpd free buff cache si so bi bo in cs us sy id wa
1 0 5320 4996 364188 220788 0 0 8 11 45 15 16 20 64 0
0 0 5320 5012 364188 220788 0 0 0 8 1061 776 14 2 84 0
4 1 5320 4420 364364 221124 7 0 106 2 1178 1596 29 5 59 7
0 0 5320 3016 364568 222264 0 0 222 67 1228 1148 21 5 66 7
0 0 5320 3032 364568 222264 0 0 0 3 1139 1090 19 3 79 0


As you can see, very little swap activity; only 7 pages were swapped out and this machine has been up for 10 days.

Re:write cycles

[schnikies79]

The ipod mini uses a hard drive. 4gb in the 1st generation and 6gb in the 2nd. The iPod shuffle uses flash.

Flash or SFF drives...

[Wdomburg]

I don't doubt flash may make some headway in the ultraportable market, but the advances in microdrive technology promise escalating capacity with reduced power consumption. Toshiba's already announced an 80GB drive in a 1.8" form factor, drawing around 1.4W and Hitachi has been talking up plans for a 20GB drive in a 1" form factor.

whats so new about this ?

[Dolphinzilla]

http://www.m-systems.com/content/Products/product. asp?pid=34

M-Systems has been providing fast FLASH based 2.5" laptop drives in the 1 GB to 128 GB range for a while - while they are god awful expensive, they do work very well and I have used them in several mission critical applications. My hope is that Samsung can get the price point down by an order of magnitude (or two)

Re:Flash Memory?

[MyLongNickName]

I googled. No such thing. Sorry.

Uhmmm...

[Cinquero]

... and what makes Apple notebooks (!) so special that flash drives only fit into that brand?

Server for Showgirls

[suso]

Maybe this isn't write cycles, but when I was at ApacheCon 2001, I met the guy who setup the webserver for the Showgirls (movie) website. He had the server right there and it used a 32MB flash drive for storage. That's a lot of read cycles.

Re:Flash Memory?

[ettlz]

...requires you to wear an oxygen mask at higher altitudes. In which case, even "mustache wax" can be a fire starter.

Is this something the RAF found out to the detriment of several pilots?

Another solution from Gigabyte

[eldimo]

Gigabyte is preparing an interesting solution. AnandTech give us a brief overview as seen in the last Computex:
http://www.anandtech.com/tradeshows/showdoc.aspx?i =2431&p=5

Basically they just use ordinary DDR and use a pack of batteries to keep the data when the computer is powered down. The batteries have a maximum life of 16 hours. So this is for enthusiasm that leave their computer always on. I wouldn't install an OS on this since in case of long power failre you would loose eveything, but I really wish I could have one so that I could install Battlefield 2 on this. It certainly would lower the very long load time. :)

the number of writes is 10,000...

[Anonymous Coward]

It used to be higher, (up to 100,000), but new MLC flash has lower numbers. Note that the 1,000,000 numbers you read is low-density NOR flash, not the NAND flash a hard drive would be made of.

You must wear level, so the real life of the drive is basically 10,000*num sectors writes. A sector is 128KB or so, depending on the flash type.

This seems like a lot until you realize that often you write sectors over and over. Also, due to the large sector/page size of flash, you end up doing multiple writes when you think you are doing a single one. For example, if you write to a file in 4 chunks, 32K at a time, it uses up 4 of your writes. It might be possible to remove this with intelligent caching, but you're gonna need a lot of RAM for the caching.

Honestly, this is just an idea that isn't ready yet. Flash is too slow to write right now. The life is decent. Reads work well.

Re:the number of writes is 10,000...

[rpdillon]

Flash is too slow? The article I read said it was much faster, but maybe they were only dealing with reads.

I thought flash was slow, too, but then I reconsidered. Since I only ever use flash over a USB connection, I assumed it was the flash that was slow, when it might have been the connection.

Anyone used flash on anything but USB? Not sure what native speed would be...

Re:the number of writes is 10,000...

[Shanep]

due to the large sector/page size of flash, you end up doing multiple writes when you think you are doing a single one. For example, if you write to a file in 4 chunks, 32K at a time, it uses up 4 of your writes. It might be possible to remove this with intelligent caching, but you're gonna need a lot of RAM for the caching.

That is interesting. I know modern CF cards employ wear levelling within the cards themselves, but I was previously leaning on BSD's soft-updates and noatime to prolong to life of the CF cards in a few servers and firewalls. Maybe some slices should also have FFS file systems with a 128kb block size to limit any block to no more than one file. I'm off to see if FFS in OpenBSD can use a block size of 128kb. Assuming this would actually cause 128kb block sized writes?

Honestly, this is just an idea that isn't ready yet. Flash is too slow to write right now. The life is decent. Reads work well.

Yes I must say that I find the high speed flash cards I use, which are directly plugged into my motherboards IDE controllers via passive IDE-CF converters (no electronics, just different connectors at either ends of the converters), are slow to write.

I have Lexar Pro series 80X CF's and San Disk Ultra II CF's. The Lexar's are WAY faster than the San Disk's BTW. But still they're pretty slow. I use them in Sun Ultra 10's and 5's, which have really slow IDE interfaces, but the CF's are slower than these and are also slow in various PC's I've tried them in.

# dd bs=64k count=1024 if=/dev/zero of=/tmp/64MB.bin
1024+0 records in
1024+0 records out
67108864 bytes transferred in 54.374 secs (1,234,192 bytes/sec)

# dd bs=128k count=512 if=/dev/zero of=/tmp/64MB.bin
512+0 records in
512+0 records out
67108864 bytes transferred in 54.322 secs (1,235,385 bytes/sec)

# dd bs=64k count=1024 if=/dev/zero of=/var/log/pf/64MB.bin
1024+0 records in
1024+0 records out
67108864 bytes transferred in 4.043 secs (16,594,884 bytes/sec)

# dd bs=128k count=512 if=/dev/zero of=/var/log/pf/64MB.bin
512+0 records in
512+0 records out
67108864 bytes transferred in 3.980 secs (16,858,338 bytes/sec)

Here I have written 64Mb to a San Disk Ultra II, first with 64kb chunks and then with 128kb. I did this because I usually use 64kb chunks and thought that 128kb would be faster for CF. Turned out to not be so, unless dd is writing 128kb chunks but the writes are being committed to the file system at no greater than the filesystems block size at a time.

The two last dd's are the same thing but being written to a 120GB Seagate 7200 RPM PATA drive. Both the CF and HDD are directly connected to the on board IDE controllers of a Sun Ultra 10, each seperately as masters without slaves.

The IDE controllers on Sun Ultra 5/10 motherboards are garbage BTW. They can get the best out of the CF cards I have, but certainly not the Seagate HDD, which does about 2-3 times better transfer rates in some of my PC's.

I feel compelled now to do these tests again but to raw devices instead of to files within a filesystem. If the results were that 128kb block size filesystem is likely to be many times faster than a filesystem with much smaller blocks, I might be inclined to build filesystems with 128kb blocks on my CF based machines. But I think I would be limited to 1024 files on a 128Mb slice. Which might not be workable.

Re:the number of writes is 10,000...

[RzUpAnmsCwrds]

Your results are not indicative of flash performance - CF is simply not that fast. I frequently get 10MB/sec with my USB 2.0 SD card reader and generic PQI 1GB SD card.

Flash can be *very* fast. Remember, you can interleave many flash chips using RAID-like techniques without the cost of having multiple disk assemblies.

Re:the number of writes is 10,000...

[SmittyTheBold]

I'd be concerned that setting up 128K blocks would be counter-prodcutive, since you'd have a hard time getting your your blocks to align perfectly with the drive's hardware blocks. After all, you lose a few K here for the partition table, another few bytes there for filesystem info...

Re:the number of writes is 10,000...

[mr_exit]

He was saying the Compact Flash interface is the slow part. Pro SLR cameras are now coming with SD and CF slots now because SD flash cards are one of the fastest interfaces around at the moment (memory stick pro is also very fast) CF just can't cut the mustard anymore.

Note that they use the same memory chips inside the cards, just as SCSI and IDE hard drives are often the same inside but having a good interface can make all the difference.

There is a reason the solid state video cameras use SD cards over CF, and its not just for the size difference.

Re: Plus Hard Card history

[johu]

Back when Plus Hard Card was new thing there were no such thing as IRQ 14 or 15 on typical PC. IBM PC was still running on 4,77MHz and using 8-bit slots with IRQ's 2-7. It was not even possible to install it to 16-bit slot of IBM AT because 16-bit part of ISA slot was blocked by frame of card. Of course most AT systems had also 8-bit slots for this particular reason so installation was still possible.

They didn't use IDE-like drives either. Controller part of board was fairly complex with multiple large chips.

There were lot of clones that were'nt as plug-and-play as original Plus Hard Card was. Clones were just 8-bit MFM/RLL (and later 16-bit MFM/RLL/IDE/ESDI/SCSI) controllers mounted to metal plate with normal 3,5" HDD.

Quantum bought Hard Card manufacturer eventually as someone already wrote.

Hard Cards from various manufacturers lasted quite long. I'd say around 10 years eg. from 1985 to mid 90's. I considered 10 years pretty long time in computer industry.

There's link to photo of original card. Text talks about smallest being 20 MB, but first model was actually 10 MB.
http://incolor.inebraska.com/bill_r/hardcard.htm

depends on your flash drive

[johnpaul191]

i know for digital cameras there is enough difference with some brands of flash memory that it will effect how long you wait between shots (at least for cameras without internal memory to buffer it). if you figure the transfer of one JPEG is that noticeable, then transferring real data would matter too.

i guess if you have the buffer of internal memory in your camera, you will not notice. so the cheaper, slower flash cards are effectively the same.

Interesting but unlikely...

[Bones3D_mac]

While this certainly sounds interesting, I can't see Apple committing entirely to flash drives until they hit the 80-100GB point.

However, one thing I can see Apple doing is giving the user 8-10GB of high speed flash memory to use in tandem with a standard hard drive, in which the user can install the OS and their primary applications. The benefit to this, is that it could make the system faster, while allowing it to conserve power at the same time. (The only time the hard drive is accessed is to either write data, or read user-selected data / secondary applications.)

Flash disks that look like mag disks

[nadador]

They're out there in the embedded market, where your option is paying more for a flash disk or having your spinning mag plates fly apart because of shock/vibe.

http://www.m-systems.com/Content/Markets/Embedded. asp

As others may have noted, there are different kinds of flash, some that have good write performance, some that have good read performance, and some that have both.

And if you want to pay, you can get an Ultra320 flash disk:

http://www.m-systems.com/Content/Products/product. asp?pid=41

Re:Flash Memory?

[Hal_Porter]

This has to be the best troll ever.

I googled for "Flash omium potential" and no hits. Google being google, I'm sure your post and this discussion will be in the index soon, so no worries - soon googling for "Flash omium potential fire hazard" will turn up enough hits for people to believe its a real problem.

Even better you said

"These conditions include higher levels of oxygen, and the like commonly found on airplanes."

Firstly, flash memory, like all IC's is sealed. Levels of oxygen around it can't affect it. Secondly, airplanes don't have "higher levels of oxygen" as far as I know. Thirdly, I work with flash memory a lot, and I've never heard of this scare story.

And then the killer

"Does this mean that the use of iBooks and PowerBooks will be banned on airplanes?"

There's no reason for them to be, but you're working hard on it. I must say, I like idea of me taking out my cheap plastic Fujitsu laptop, and yuppie next to me taking out his ultra hip (and ultra expensive) Powerbook only to be told he can't use it because of some bullshit firehazard while I work away.

Powerbooks are a firehazard BTW - their cases are made out of a material which is fire hazard. Even worse, terrorists may ignite them with a blow torch deliberately. The Department of Homeland Security must protect the homeland and band them now.

Re:Flash Memory?

[Anonymous Coward]

http://en.wikipedia.org/wiki/Spontaneous_human_com bustion