Slashdot Mirror
No More Rebooting?
blankmange writes: "This headline caught my eye: 'The End of Computer Rebooting.' Seems that there has been some new developments in memory technology: The new thin-film technology that could give rebooting the boot is based on resistor logic rather than the traditional transistor logic used in most PCs and other memory-enabled devices. It also is considerably faster than current memory systems and holds the promise of reducing the time required to transfer and download multimedia content and other massive files. This is great news, but what am I going to do with the extra hour or so a day?"
I've been waiting for this technology for over five years...when??? When??
When it's done, of course! (Please don't sue me Id Software)
So I guess this puts a big damper on Microsoft Tech Support. "I don't know what to do, please restart your computer."
So if I can't reboot, how am I supposed to recover from Windows crashes?
You are running windows ? right ?
.I would write a signature but I don't remember one.
My Win2k boxen are stable enough to be up for months without a reboot. What I need is a box that I can leave on 24-7 and not have to worry about energy consumption. These things are expensive to leave awake all day. Seriously. Do the math.
Life is the leading cause of death in America.
How else are we supposed to fix things when everything just stops working...
Rebooting is always a great way to fix things.... they even used it one of the star trek's once.
And how the hell is windows supposed to work?
But we can pretty much do this at the moment by using the various suspend and hibernate options. Ok, so it's a different technology but the effect is the same. But nobody not using a laptop ever does.
Sig is taking a break!
Like your computer, you need downtime (sleep, walking the dog, eating, etc).
If you are an avid computer user, you may only get your downtime when your computer is rebooting. This is especially true in workplaces where people are "chained" to their computers trying to finish a project, etc.
Those ergonomics posters on the wall do very little to get an average 'puter user to take care of themselves.. reboots served some of this purpose.
(Maybe that is why windows crashes so much - it's Bill Gates' gift to the employee!)
In any case, perhaps all offices should institute a staggered mandatory 15 minute inactivity period every couple of hours for each active computer.
Goat sex free since 2001
Easy -- get a second job, 'cause its probably going to cost all the extra cash you have to get stuff built with this, just because of the patent licensing rights...
"But remember, most lynch mobs aren't this nice." (H.Simpson)
-- Joe
This seems like a bad bad title. This stuff is persistent RAM, so it won't help you if you need to reboot after recompiling a kernel. Also, this article doesn't mention getting rid of POST...most computers stuff some data into memory and check it to make sure your ram is still working. Would be kind of hard to check your memory if it has stuff in it that you don't want to lose.
The Right Reverend K. Reid Wightman,
It just talks about memory that doesn't lose state when you hit the power button on your PC.
We've got to invent perfect software that can run forever without needing to be restarted, first.
Well, rebooting is MUCH more that just recovering the memory content!!
We could easily dump the memory contents onto the hard drive straigh away and we are not doing it (except in laptops, but even there it doesn't always work) This is because rebooting reinitializes various devices and takes care of the time jump (i.e. crons, anacrons, etc). The more complicated your system is the less likely it is that you are going to survive without booting.
Also, computers are now 1000 times faster than 10 years ago and they take much more time to boot (DOS did it in seconds on 286).
Windows has to be rebooted not because of the type of memory used, but the way it handles it. You could have the quickest, most efficient memory around and Windows would still make a pigs ear of dealing with it.
Unix boxes use exactly the same type of memory as PC's but the kernel is rather better at managing it...
Still, this memory would be great in a little Network Walkman - better than the 64Mb in my one!
Heck, I thought that was a given...
That's an extra hour for UT (Tactical Ops mod, of course)!
----------
Aww, FSCK!
Come to the University of Mars! Classes starting soon!
What the hell am I gonna do when the boss comes in and my puter is off and I tell him, oh im just rebooting.
Second, isnt resistor logig analog, and not binary (transistor) ?
Sig went tro...aahemmm.....fishing........
C'mon now...I've seen 286's that took 10 seconds to just count up 640K of RAM during the POST!
OLD personal computers simply turned on and gave you the OK prompt from the ROM BASIC.
Seems like this innovation won't actually eliminate the need for rebooting, it will only keep your computer's state between power downs. This may or may not help you when your OS freezes.
The press release doesn't really go into detail, so I don't know how similar (or disparate) the respective IBM and Samsung solutions are. They do both have the same net effect for users: non-volatile main memory.
This is cool stuff, but what hasn't been said is that as long as operating systems and applications leak memory, there will be a need for reboots.
Ciao.
Last time I checked, downloading speed depended on your connection, not how fast your RAM goes. I'm sure my memory can handle more than 1.5 Mb/s but that's as fast an I can download, because that is the limit of my DSL line.
"Da ist ein Technölüst in mein Unterpanten!"
What's a reboot? :-)
My firewall has an uptime of 317 days, my vmware PC has 195, (all running Linux), but oh yeah my games pc, (windows 98), has to be rebooted every day.- OK it would be an advantage to those uptime challenged OS.
-- main(s){printf(s="main(s){printf(s=%c%s%c,34,s,34
What the article seems to be saying is that there could be a way of producing non-volatile memory which is so cheap, you'll be able to use NVRAM instead of ordinary RAM in your computers. But that depends on no further falls in RAM prices - I wouldn't bet on this technology taking over.
However, a cheap, fast non-volatile memory which can be written and read unlimited times could be a very useful supplement to RAM. Think journalling filesystems for example - put your ext3 journal in a 100Mbyte NVRAM device and you'd hardly need to touch the hard disk for hours at a time, given moderate use. (Eg notebooks could spin down the drive.) This is possible already, but NVRAM devices are relatively expensive and most PCs don't have them.
-- Ed Avis ed@membled.com
People: Read the frelling article. This isn't 'an end to rebooting', it's highspeed nonvolatile memory that could theoretically be used to replace mass storage and RAM simultaneously. Although this would speed up booting a bit, it would not obviate rebooting entirely.
In fact, on some OSen (cough, Windows, cough) it could be very dangerous - if there's only one copy of the OS code in this combination memory, you can't reboot and reload a fresh copy from disk - meaning bugs have a significantly greater probability of rendering your system unusable.
Sounds like fun, right?
--
Damn the Emperor!
This is what memory was like in the olden days, before DRAM took over from core memory. Apparently early DRAM computers had big piles of batteries in the bottom of the RAM cabinets to prevent the memory from failing due to a power cut.
(We're talking early 70s here; before my time)
Is a rare voluntary reboot really worth the unmentioned price?
This new type of memory won't eliminate the need of rebooting.
If your OS happens to crash, you'll still have to wait for it to re-initalize itself, i.e. reboot. With this new memory you just don't have to do it every time you switch your computer off and back on.
That doesn't help us one bit. :( sure it ends re-booting, but now the windows box is totally useless. (wait, this may be a Good Thing (TM) afterall)
I was hoping this was going to be a story about how they found a way to keep windows from crashing, instead its about memory that's non-volitile, so re-booting after a crash would be futile since it would come back up the same when you turn the power back on.
Did anybody else run into a 404 when you clicked "read more"? /. is slippin'...
Well, as my collegues have pointed out, I must say that the title was misleading. This has been thought about a lot, it's applications and implications, but without proper software, it won't eliminate the reboot. That's like saying a video card that can project 3d indisinguishable from reality, and which provides an api for its own ui, will eliminate the reboot.
It's just one more step -- hell, linux boxen are approaching record uptimes of years! (right?)
-------
boat
What about when W2K or NT or 95/98, etc. decides to not quite completely clear out of a particular area of memory? Will this plan still flush it out for me??? I hope so...if so...this COULD do alot for Windows stability.
So in other words, this is not something you're going to be seeing anytime soon. And by the time it is ready as a product, who knows what kind of memory/HD technology we'll be using, or what the price of these chips will be compared to the mainstream stuff...
Don't people WANT to be able to remove power from a system on occasion and start from scratch?
When your favorite Windoze flavor goes south, for example, and a "proper" shutdown is impossible, the power switch is a harsh but effective alternative. What would happen in this scenario when you CAN'T clear memory to get rid of the infinite loop inside a rogue ring 0 driver?
Are we now going to have a "reset" button for our memory devices? Barring that, I suppose one could always keep a small screwdriver and a pinout of the device handy... hehehe
"This is great news, but what am I going to do with the extra hour or so a day?"
Extra hour a day?! So... Err... You're a windows user, right?
Back at the dawn of time, I was programming a (Data General) Nova II mini-computer which had "core" memory (which is where the term "core dump" comes from). Core consisted of tiny doughnut (ummm doughnuts) shaped magnets with (read/write) wires through it. It was incredibly slow by today's standards, but it did retain memory even when powered down. I'd shut the machine down at the end of the day. The next morning, I'd turn it on, and immediately pick up where I left off.
[Insert pithy quote here]
This has got sod all to do with rebooting in a PC context. This would be more usefull as storage. If it going to be cheaper than existing memory (although it doesn't specify what type of memory), this could be a step towards the end of hard disk drives as the primary storage device.
Jason
This is all about CHEAP, NON-VOLATILE MEMORY.
Jason.
Hmm. Rebooting nowadays with 'traditional' OSes is to flush inappropriate state information out of the memory - an unusual sequence of events resulting in the system getting into a state it should never be in during regular operation....this might be either accidental (a crash) or semi-deliberate (an upgrade of a software component which needs a reboot to get it co-ordinated with the rest of the system). Having memory which maintains this state information will make the problem worse, not better!
What's needed here to achieve systems that don't need rebooting is operating systems which deal with all of these unusual events and states correctly..this means they'll catch errors and will be specifically designed to allow things like dynamic update to system compoents. I'm probably a bit biased but the best example a no-more-reboots kind of environment I see today is the OSGi.
C'mon, I know that I have to reboot windows every couple of days to get rid of libraries that errant programs didn't unload and windows doesn't seem to let go of.
;)
Also, what if computers weren't *allowed* to reboot. You couldn't run a dual boot system. Which is something I suspect Microsoft would like. (I had to throw in a groundless msoft conspiracy...
*everything* is Orwellian to cats.
The article is not even clear on whether this development is supposed to be replacing RAM or hard disks. But either way, it cannot eliminate the need for rebooting. The primary reasons for rebooting are either to reset the operating system to a known state, or to upgrade low-level software (such as the kernel in Linux, or your web browser in Windows). Neither of these necessities go away with non-volatile RAM, regardless of how fast, cheap, or capacious it may be. These are software issues, and they need software solutions.
use constant PERL_IS_BROKEN => $] >= 5.006;
Hmm I see that some ppl have lost their hopes in Microsoft making a stable OS ....
Migx
I cannot see how this will kill rebooting. I thought rebooting was due to buggy operating systems, and little to do with the memory technology.
:-)
Or will there be something in the chips which will disallow Windows from running?
i'm hard pressed to guess who is responsible for filling that article with nonsense: clueless journalist, clueless editor, clueless "researcher", clueless marketeers...
some or all of them should know that rebooting isn't only done because the machine was deliberatly powered down.
some or all of them should know that simply having all memory intact isn't good enough if external environment has changed (eg network connections).
some or all of them should know that disk-based versions of 'instant reboot' have been around for years, and don't always work.
and what's this about faster multimedia downloads? what's memory got to do with it?
this whole thing reads like marketing bumpf. wishfull thinking on the marketeers' part, as usual.
Everything should be made as simple as possible, but not simpler. -- A.E.
9:12AM up 13 days, 17:56, 2 users, load averages: 0.64, 0.54, 0.42
The last time I rebooted was because I thought it had died. Turns out the battery drained when I left it on overnight plugged into a deactivated outlet.
And by the way, I run buggy Microsoft programs, betas, shareware, DivXes--so it's not like I'm not using it. What's the point of speedy boot times when I'm not rebooting? Viva la OS X!
On many systems, mandatory, periodic rebooting is part of the ressource management of the operating system (think of memory leaks, descriptor leaks, and so on). Even if you implement RAM using ferrite-core memory (or something else, like this new approach), these maintainance reboots won't go away, and they won't become faster.
In any case, such memory devices would be great for storing the journal of certain file systems, or even as replacement for traditional mass storage.
How would this have anything to do with faster downloads of multimedia content? I didn't even see that in the article, but this has nothing to do with bandwidth.
Flash is the speediest memory technology? Surely they mean speedier than eeprom.
How does this prevent reboots? I say without any doubt whatsoever, that the majority of reboots has to do with M$'s ~90% marketshare and numerous system level flaws. Does this memory plug its own leaks? Or do third rate OS programmers and ugly billionaire monopolists actually become smarter when exposed to this, sorta like Superman and kryptonite?
Verdict: Marketing fluff.
The banishment of the venerable reboot means that Windows XX (insert stupid letter code here) will be running 24/7. For argument's sake, I'll assume they somehow make it stable enough to accomplish that. What does this mean? Simple, with the rate of memory leaks in Windows apps, you will need to have hot-swap RAM slots, and you'll need to feed your computer more ram each month.
So, perhaps M$ will start bundling persistant (PRAM?) ram sticks along with your new license agreement of the month?
Or just get a laptop with the slight performance hit: 30 watts, plus a built in UPS supply. Other advantages include easy to transport to another location without removing dozens of cables or powering down.
Less heat, less power, why haven't these caught on?
Or do laptops have a habit of mysteriously walking away compared to their boat anchor bretheren?
I suppose this could be useful on systems that can do suspend-to-RAM, like laptops. Such systems still need a trickle of energy from the batteries to keep the data stored in memory from decaying.
Also, a system with persistent memory would be like the old mainframe and minicomputers that had core memory. In the event of a crash, the memory could be examined. I suppose this could be somewhat beneficial to operating system developers..
As I've already seen alluded to in other posts, surely there will be need for Win XX machines to get back to the initial start state after they've crashed. I am starting a company that will sell reset buttons to accomplish this. No *nix users in the target market -- sorry.
Laws affecting technology will always be bad until enough techies become lawyers.
I think where we all got off track because they used the wrong term. They mean that PCs using this technology will be more like "instant on"
They just mean that the nonvolitile ram will save you from having to go through the boot sequence to recreate the state of the machine before you turned it off
Of course, just like current machines returning from standby, it will have to reinitialize the devices, but that would be some time savings
If I dont have to reboot anymore, when am i supposed to go to the bathroom at LAN gatherings?
If you can't see the value in jet powered ants you should turn in your nerd card. - Dunbal (464142)
Another thing not mentioned in the article is the memory's capacity. Is this thing only going to be available in a 16MB module at first, or will I be able to get a 1GB module? Inquiring minds want to know.
Read my keyboard review.
This is great. High speed, solid state storage on the cheap.
The real killer app for this technology would be to replace hard disks and give your archived information store the same bandwidth as RAM. In the short term I don't think people will want to merge long term and short term storage because of the addressing issues (except for 64 bit computing) and also because of the major rethink in terms of OS architecture required.
I'm not sure how much confidence I'd have in this stuff to store my critical data. What if a power spike killed a stick of this stuff? I guess a lot of work would have to be done making it robust enought to use in a home enviroment.
You can't win Darth. If you mod me down, I shall become more powerful than you could possibly imagine
from the article:
"It can be used in PCs, cell phones, networks -- anything that needs massive memory."
i know the japanese are big on fancy cell phones, but do they really need massive memory yet?
Everything should be made as simple as possible, but not simpler. -- A.E.
It seems to me that we're not understanding how this could be set up.. Why not have this as a device in your machine, that has an interface to the BIOS, where the user can set/format the unit to fool an OS into treating this nv memory-space as a fat32 or ext2 disk? You go into the BIOS, flush out the nvramdisk and 'reformat' and you are ready to re-install your OS should that become necessary. The rest of the time it runs as a really fast 'C:'... There's no need to replace normal RAM as your actual main memory during operation. Windows need not be aware of what is actually happening, you just boot/use faster, that's all.
Recycling the power will no longer cure the BSOD.
This is an appalling summary - and the article is no better.
"The technology is highly suitable for broadband Internet connections, Hsu said, noting that it combines the features of low voltage, high speed and low power consumption."
Yes, fantastic. That's great for those broadband internet connections. Faster memory is always good, but choosing this as an application is just a moronic use of buzz words.
"Ignatiev said the new technology is about 1,000 times faster than flash, which is nonmechanical and currently the speediest memory on the market. "
Flash memory is the fastest type of memory on the market? No, it is a form of non-volatile memory, which is very slow by RAM standards.
"is based on resistor logic rather than the traditional transistor logic"
Actually, you'll find that DRAM in most modern computers are capacitative devices - the techniques to make them are the same as MOS transistors, but they do not use switching to store values, IIRC.
I wish people would not spout such rubbish.
I run a modern version of UNIX (Solaris 8) on my workstation and haven't ever rebooted it except when we apply kernel patches or when a harddrive failure occurred.
While persistent RAM might be nice, there still is no substitute for mature software that manages its memory properly. Corrupt persistent RAM is still corrupt.
Healthcare article at Kuro5hin
Basically, strictly speaking this isn't preventing 'rebooting', it is enabling a system to boot really fast and load the state of the OS from non-volitile memory and have the state preserved. This just allows the boot process to skip the OS initilization bit (which is significant, but excludes BIOS startup).
This has been around (save-to-disk hibernation), though using non-volitile memory would increase the speed of the process could increase dramatically. It seems that they are proposing a non-volitile ram technology that claims comparable performance to the volitile memory we use today, so it would be always ready to restore from that state, even if the shutdown is unexpected (power outage, for example).
However, the annoying part of the boot process to me is the PC Bios. After it's part is done, I can tweak things to start fast, but BIOS, even after tweaking is unbearably slow. I presume on restart a computer may still go through BIOS before restoring state, and even then I presume it needs to offer the option of starting over (don't want a BSOD to be permament). I'm more interested in a BIOS that doesn't take forever to come up...
XML is like violence. If it doesn't solve the problem, use more.
I see the reboot issue as minor, compared to the other potential advantages of this technology. I will expect to be rebooting, for one reason or another, for years to come and am not too bothered.
The article glosses over what I consider the important advantages:
- [assumedly] great power savings. Great for portables and remote embedded systems.
- No moving parts! If this tech can really replace and even surpass in speed, Hard Disk Drives, reliability and performance should make a gain of at least an order of magnitude.
I've been waiting for years for computers to become eletronic-only devices. I've harped before that CRT's (vaccum tubes, for God's sake!) and HDD's need to join the Dodo in oblivion. This new tech, in the common mass storage area (HDD's, CD'c, floppies), along with flat panel technology, would put us right on the verge of that ideal. The last hurdle would be cooling without moving parts.
I understand that's booting and not rebooting that technology promise to get rid of. But how did you do hardware reset, IRQ/DMA peripherical association without a boot sequence? How the CPU state is stored to go back where you have been before shut down (as CPU registers are not stored in main memory)? Did you need special OS to detect this kind of memory and work with it?
Check out Squeak, the free, portable Smalltalk machine. Like all Smalltalks, Squeak runs in an "image". The image is your entire language, programming environment, and execution environment, all at once.
The interpreter, programming tools, and even the GUI all exist as long-lived objects in this large (sometimes very, VERY large) memory space. When you aren't using Squeak, the image gets stored as a file on disk.
There are also projects to run Squeak on bare metal--no intermediate operating system like Windows or Linux. Squeak itself becomes the operating system.
This memory technology would be ideal for a Squeak machine. The image would always live in NVRAM. In such a case, there isn't a distinction between the operating system as it exists in static form (files on disk) and executing form (code in memory). There are always just objects in memory. Very elegant.
"Genius may have its limitations, but stupidity is not thus handicapped." --Elbert Hubbard (1856-1915)
Your first paragraph showed you at least read the article, the second paragraph is a nonsequitor. It's not clear to me why you think Windows would somehow be negatively impacted by this and no other OS would. Look at the number of times changes have had to be made to the Linux kernel in order to get it to boot on new hardware such as the Pentium 4.
Isn't it likely that if this technology came to pass, the people responsible for various OSen would test their OS in that environment, and make changes as appropriate to support it?
Interesting highlights:
The trasentric paper quoted Electronic Buyer's News:
The interesting elements of this:- Much of this research is funded by a DARPA contract which means it is the money of US Taxpayers at work.
- Samsung is part of the same contract.
Methinks that perhaps Samsung and IBM are using the same (or very similar) technology.The Wired article is fairly lengthy and also details the biography of Stuart Parkin. Parkin is the IBM fellow that has been driving most of the MRAM research.
Ciao.
Now I do my thinking during the downtime that I'm on Slashdot.
Miko O'Sullivan
This computer's sooo fast I can reboot TWICE as often in half the time!
(Yeah, yeah, there'll be a glib Windows sux reply to this one, I'm sure.)
"Draco dormiens nunquam titillandus."
I assumed they were talking about Windows PCs, and my first thought was "How does memory keep the machine from crashing?"
Then I realized that they meant "turn on the computer for the first time today" booting, not RE-booting. Doesn't affect me, the only machine I ever turn off anyway is my laptop.
This is great news, but what am I going to do with the extra hour or so a day?
Find a better operating system.
Oh, the In Memory Virus writers are gonna love this stuff
reduces time required to transfer and download multimedia content and other massive files.
Now I'm really nosy how in freak'n hell any memory technology can reduce multimedia download times? That's just non-sense, it seems the word "multimedia" must be in everything you want to sell.
Download times dependand on things like your internet connection, compression used, your providers connetion, etc. but not my memory.
I still remember an intel guy claiming thi Pentium 3 will make the internet faster... how can somebody even dare to claim nonsense like this? And the really sad thing is: nobody started laughing as he said that...
--
Karma 50, and all I got was this lousy T-Shirt.
I can't stand these Instant-On PCs that don't actually turn off when shut down. There are many times when you just want a clean start, without having the (possibly flakey) state of the system retained across a shutdown/startup cycle. With these "convenient" systems you have to pull the power cord to actually turn off your machine.
The real innovation would be for mainstream OSs (read: Windows) to be stable enough that they don't need to be restarted often. The time it takes to reboot your machine is not an irritant if you hardly ever have to do it.
you could still recover from a bsod or other crash. you would just have to reinitialize your os. rebooting itself is not what fixes your crash problems, it is the fact that your os is reset to a (hopefully) stable state, by rebooting.
The reason for the original poster picking on Windows is that Windows, in most folk's experience, has to be rebooted frequently because of error accumulation. In other words, I could leave my Win98 box running (with no additional applications up) for 12 hours and when I came back it would be locked up. This isn't about shutting down during upgrades or installs, it's about shutting down because of frequent OS corruption during everyday use. In this case, you *need* memory to clear itself out.
"Can I say you're my lovepuppy?" Founding member of SODAMNHOTT
Recall that non-volatile memory gives sleep mode a new meaning, so probably sleep modes can be done to have a very small power consumption -- smaller than nowadays.
because he's never had a girlfriend.
That's what I wanna know... when are we gonna get our isolinear chips?
As always, hardware is ahead of software
Karma stuck at 50? Add 2-5 inches.. err.. 2-5x Karmas Count to your pen1es.. err.. Karma all naturally and private
What's so exciting exactly? They invented faster FLASH memory.
This is not the end of rebooting computers and, unfortunately, not the end of mechanical hard drives.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
i wonder how much heat this kind of memory produces, if it's purely resistive then isn't all the power it consumes turned into heat?
not so much worried about not being able to reboot anymore all you'll need is some kind of reliable bios flushing routine to flush the OS out of active memory if it does crash.
[root@nerdhero /root]# uptime
4:53pm up 307 days, 23:05, 18 users, load average: 0.21, 0.26, 0.27
--- Sigmentation Fault - Comments Dumped
Most likely, we will still partition disks; but instead of a swap file, you'd probably reserve a coule of gig for "memory space" where programs make a copy from the "disk space" for running.
Please consider making an automatic monthly recurring donation to the EFF
If resistor logic is used, then i wonder how much heat will this new device generate? (Since resistors are currently used to actually produce heat!)
--Manuel
"I hate quotations, tell me what you think"
In the last two years I have not had to reboot hardly any of my machines. I run, Linux, Solaris, MacOSX, Novel, and yes even Windows 2000/XP all of which rarely make you reboot, even with a system update.
The Amiga had a very similar feature.
The amiga had soft reboots and hard reboots.
The problem with soft reboots were that some memory was purposely left unblanked (usefull sometimes), the problem with this was that viruses like 'Lamer Exterminator' and 'Saddam' took advantage of the fact and a soft reboot didn't remove them from memory.
So scale the memory and CPU power acording to Moores law and you have something similar to that the Amiga did in 1985. Not to mention Kickstart, think PC BIOS but with further abstraction and what about suspended PCs?
e4 e5
... is not the boot time improvement, although that is important. The primary benefit of tech like this is the (long overdue) elimination of the artificial barrier between volatile and permanent storage. With this tech, OS memory management can be optimized to a much greater degree. Available RAM is synonymous with available "disk space". Swapfiles cease to exist in their current form, since if you fill RAM you fill the disk. Disk caching becomes unneccessary.
However, these benefits are only attained if the RAM is at least as fast as current tech. If DDR/RAMBUS is faster than this flavor of NVRAM, there's no reason to unify system RAM and storage.
I assert ownership of all trademarks and copyrights on this page.
As long as we are swinging around our uptimes. True, this box doesn't do a whole lot, but still... Can anybody top this? ;-)>
ls-1010>sh vers
Cisco Internetwork Operating System Software
IOS (tm) LS1010 WA3-7 Software (LS1010-WP-M), Version 11.2(15)WA3(7), RELEASE SOFTWARE (fc1)
Copyright (c) 1986-1998 by cisco Systems, Inc.
Compiled Mon 14-Dec-98 16:54 by integ
Image text-base: 0x600108D0, data-base: 0x60448000
ROM: System Bootstrap, Version 201(1025), SOFTWARE
ROM: PNNI Software (LS1010-WP-M), Version 11.2(5)WA3(2b), RELEASE SOFTWARE
ls-1010 uptime is 3 years, 6 weeks, 5 days, 23 hours, 56 minutes
System restarted by reload at 07:54:52 MNT Sat Feb 27 1999
System image file is "slot0:ls1010-wp-mz.112-15.WA3.7", booted via slot0
cisco LS1010 (R4600) processor with 32768K bytes of memory.
R4600 processor, Implementation 32, Revision 2.0
Last reset from power-on
1 Ethernet/IEEE 802.3 interface(s)
13 ATM network interface(s)
125K bytes of non-volatile configuration memory.
16384K bytes of Flash PCMCIA card at slot 0 (Sector size 128K).
8192K bytes of Flash internal SIMM (Sector size 256K).
Configuration register is 0x102
ls-1010>
The technology to prevent you from having to reboot your computer daily can be found here and here. It's been around since the early 90's, and was invented by a grad student...
--
I Hit the Karma Cap, and All I Got Was This Lousy
Much of this research is funded by a DARPA contract which means it is the money of US Taxpayers at work.
So are we really talking about desktop PC's or something more like a missile fire control system on a warship which needs to work straight after being forcably power cycled, before the next bomb or antiship missile is launched at it?
resetting the contents of ram is not the only thing [which is why hibernate mode doesn't alwayswork].
:-)
Not only do you have to make sure the RAM is intact [which is what this design solves???] but you have to make sure every single volatile register in every single piece of hardware is in the same state.
The latter part is something I don't see this addressing.
Here's a trick. Open up WinTV in winxp. Go full screen then hibernate. Try to resume...
Tom
Someday, I'll have a real sig.
Well, that is as cute as I can be this morning, but I hope the point is clear. I was willing to reboot an XT running MS-DOS 2.0 from time to time - it was a crude system and we didn't expect too much of it. But the "reboot" virus has spread FROM Microsoft systems all the way INTO the world of distributed controls. I actually have control system techs say to me "reboot and see what happens". Hello! It isn't supposed to be this way! Systems (particularly embedded sysetms) are supposed to work, not not work!
Faster rebooting would be a crime, not an improvement, since it would help take everyone's attention off the problem, which is that the system failed.
sPh
Correct me if I'm wrong, but isn't the _reason_ that Windows gets rebooted, is because of something going wrong? IE a memory location getting corrupted, or just a program with a memory leak?
If you'r machine has just GPFed and hung, reloading the memory state is going to do nothing useful...
Reboots may be irritating, but the advantage is they 'reset' the state of the machine to a know situation. Getting rid of all the entropy that's crept in due to bad memory handling and just general system untidyness...
So persistent memory would lose this advantage, taking you back to the 'reboot and reload everything'. If it's faster, then your machine will start a bit faster, but it's not going to be much...
Probably both, possibly neither. Your options are neither exclusive nor exhaustive.
Ciao
Where do they come up with this stuff? Unless your memory is Extremely slow, and I mean slower than anything used in personal computers since the days of the Altair, AND you are using a faster than 10M net connection (not realistically possible on an altair) on that slow memory, this will have ZERO effect on download speeds. Any computer faster than a 386 can handle damn near GigE where the limiting factor will be the PCI bus, disks, etc. - not the memory. (well, maybe non GigE on a 386, but 100M easily.) Sheesh.
Considering I can't even get DSL or Cablemodem service in SILICON VALLEY, I don't think we will be seeing memory speeds being the limiting factor in downloads anytime soon - like not in the next 10 years even if computers stopped getting faster.
This depends on the meaning of frequent and reboot. Consider:
I'll gladly concede that using a reboot as a trouble-shooting device was an area heavily pioneered by help desks supporting various flavors of Windows. However, many devices need periodic resets due to defects in software and/or hardware.
Ciaio.
Because it would end with my argument with my boss
phone rings:
boss: "Where the heck is file youre suposed to send me?"
me (turning off Quake 3 and opening eudora): "im sending it right now, I had to reboot the computer because of a Windows crash and I lost the file... I had to do it again! Dont worry, I just finished it and im sending it right now!!"
Fabio - Sumare/Sao Paulo/Brazil/South America/Earth/Solar System/Milky Way/Universe
http://www.morroida.com.br
Honestly, at the risk of sounding cliche, real unix systems, that are bound to their hardware have fantastic uptime. The RS6000 we've had for a year has only been taken down for failover testing. If the resolution is hardware based then this Wintel duopoly isn't of much use. But the biggest question is, what will help desk people do if they can't tell people "Reboot the system." to take care of the problem. The vast majority of them may actually have to be trained in technical problem resolution. Millions will be thrown out of work (because it obviously exceeds their capabilities). Its the fragile nature of Windows that keeps the economy moving! Hopefully they will rethink this before its too late.
We already know how to save out all RAM to disk. The problem is in saving the CPU state(s). Hardware needs to be initialized and configured before it can be used efficiently. Someone has to tell the hard drive to switch to ATA/100 bus speeds, and someone has to tell the video card to init the display in 1600x1200. Loading the OS from a memory dump is the easy part.
-Billco, Fnarg.com
Man you should get WinXP then, I got my reboot time down to 32 seconds using the VisBoot tool microsoft has for XP. And even before that my boot time was about 42 seconds.
:)
I'll probably get flamed for this, so that's why I'm posting AC, I love my karma too much
So it is non-volatile RAM. That makes four distinct NV-RAM technologies that I know of: battery-backed SRAM (fast, expensive, and low capacity), Flash and other electrically eraseable PROM's (slow writes, wears out), magnetic RAM, and resistive memory. The first two have been on the market for years, and capacity/price are nowhere near competitive with hard drives, although they are used where capacity can be much less than a PC needs and the environment is hostile to hard drives. MRAM is now being sold in small quantities, I think, but it's too young to tell how price and performance will work out.
What I did not see was any reason at all for thinking that resistive RAM would work out to a low enough price to be a hard drive replacement. I'll believe that, with enough work on the production process, it can beat SRAM on price and Flash on write speed (these aren't hard targets), but it has a very long way to go to compete with DRAM on price or speed, and then the price has to go down another 100 times to compete with hard drives. OTOH, start selling boxes with 256M of NVRAM and good non-bloated instant-on software, and maybe people will prefer them to MS's bloatware offerings on a 30G HD...
Finally, there have been much ballyhooed nonvolatile memories before that died once they hit the market. Bubble memory was supposed to replace hard drives about 20 years ago, but most slashdotters are too young to even remember it... I do like to see another technology out there, because if MRAM stumbles, now there's another chance of getting NVRAM that doesn't require major compromises.
Comment removed based on user account deletion
This means that they are not ubiquitous.
I probably know a few dozen people with hand held computing devices. None of them own a PocketPC. One owns a Sharp Zaurus.
They need to be periodically reset. Also consider that twenty to thirty hours of use translates to several months for most (not all) users of hand held computers.
Comment removed based on user account deletion
Think about it, I will need to reboot because my OS sucks. With this new memory, however, at least when I reboot, I would not have to refill my cache.
Just a thought.
As always, multiply the marketing department's wishes with pi. In this case, 3*3.14, it's something like 9.5 years. I'd say end of 2010 at best.
1) Basicaly a Transistor is a variable resistor.
2) Before Cmos and TTL ther was RTL logic. (Resistor Transistor Logic).
Yeah, It's gonna take FOREVER before we can have uptimes of +INFINITY.
And someone's got to take care of that exploding-sun problem and the entropy problem thingy.
NO MORE REBOOTING?!?!?!?!? Now how in the hell do I get rid of this blue screen with all of this garbage written on it?
the new version of windows will still find a way to make you HAVE TO restart wheather its a crash or a just BECAUSE it WILL make you reboot.
Ave Molech Setting
What about this black screen I have??? How is this going to prevent CVS from eating up every single bit of ram in the machine and hard crashing it?
That's a real knee slapper! You can only make a statement like that to people who aren't using UNIX. Ooops, my mistake, Lunix isn't UNIX. Thus the fact that it crashes constantly also doesn't apply to your RS6000 story. I guess you really shouldn't have even mentioned Windows. You might have actually made a point then. As it is it's pure FUD.
However, systems such as you talk of, don't NEED to be forcibly power-cycled. However, they are generally shut off, to avoid unnecessary power drain. But when entering a threat area, they are turned on.
General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
You can add on to the non-stop sitting and snacking! thats what being a computer programmer is all about
--JonnyBlog
Ok, so what happens when you *want* to reboot? Like, say, for installing certain Free OSs? Would it then be a case of 'you bought it, you live with it hahaha'?
I like either Linux or FreeBSD, and I don't see a reason to change... how much ya wanna bet devices with this will have CE on them?
You are all fartheads.
Have you not upgraded to Ext3 or are you running Windows 9x?
The article wasn't any better written than the summary. It seems like this is suitable as a replacement for flash memory, not for either disk (which is huge), or for RAM (which is really fast). Of course, having a flash-like technology be cost-effective would change things; you could keep a copy of system memory as it is when it has just been booted (but before it initializes devices) there. Then you "reboot" by copying the virtual memory table from the nvram to main memory, and the system is immediately ready to initialize devices and run.
It would also be useful if programs could put some of their data in the nvram region, so (for instance), your emacs buffers don't go away when the power goes out. It would also be a good place to put write buffers, such that, as soon as the data is written to nvram, it will definitely make it into the filesystem, whether or not you lose power. This means that you can accumulate more dirty buffers safely and write them out in larger chunks, which is more efficient.
Keeping everything in nvram (if that were fast enough) may or may not be a good idea. You'd still want to reboot on occasion to refresh the system (load a new kernel, e.g.), but there's no particular reason you'd want to reboot at exactly those times when you power down and back up. Of course, you'd need everything to be hotswappable (replace the processor with programs running?) and restartable (disks have to be told to spin up, e.g.).
You mean streaming formats, online flash, and general Internet media will be slightly less crappy? Wow!
Digital is just another way of saying;"more expensive, less relievant, less quality, more waiting.
The only situation when i reboot is when i compile a new kernel or when i change my hostname. Now how often does THAT happen?
cheers =)
Linux is what you need
Rebooting is sometimes used to refresh the state of the RAN, rather than just to power down. So there will be needed two new facilities.
1) a clear function, which will set the ram to a known good value.
2) an initialize function to recover from connections that may have been left active, and timed out while the power was off. (And probably to do other recovery that I haven't thought of just yet.)
In the old core memory machines, the core was frequently cleared without turning off the computer, and then another IPL (initial program load) was done to start a program running. These features will need to be re-invented for more modern environments if ram becomes non-volitile. Of course, design could cause only some ram to be non-volitile, in which case it could be treated as an extermely fast disk (faster than the volitile ram? That sounds unlikely, but if so it could cause interesting design changes.)
I think we've pushed this "anyone can grow up to be president" thing too far.
And if they can hack the density and cost problem, replacing disks with this technology could actually work.
To sum up: Typically "we solve a major problem" false-promise headline. Maybe they are looking for funding.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted and ignored otherwise.
But... and software?
you could imagine your apps/desktop/prompt whithout a harddisk? is supossed that things evolve this way because they needed a disk (and still need a disk nowadays) and this mark the behavior of software but what will happen when those things are solved?
are you prepared to emulate hard disks using ram?
You know, it switches all the ones(1) to Zeros(0).
Then your 'puter is back to a base state with memory unused. You'll lose any un-saved info, so you better not have your HDD & RAM as the same device or you'll be as screwed as your everyday Palm Pilot user.
Oh, and it isn't just windoze that crashes. I've written plenty of bad software that will fux up a unix machine.
FOr those of us still learning the craft of software development (and aren't we all, always learning - pushing the limits..crossing the limits and crashing hard core) we need a machine with a reset button!
mlylecarlin
End of computer rebooting? What folly.
There are a couple of features that were elaborated on in the 'Instantly Available PC' initiative years ago that work around the volativity of DRAM to provide a rapid resume type function (i.e. your computer resumes to its previous state, with all your apps open as they were, without having to go through the Windows boot process). Suspend-to-Disk (S4) dumps the RAM content to disk and resumes from there. Suspend-to-RAM (S3) actually maintains the RAM content using the ATX power supply's +5VSB current, and even though the computer appears completely off, the RAM is powered as long as the PC is connected to the mains. As one might expect, S3 is much faster than S4, but is vulnerable if a power outage occurs. Both of these features have been plagued hardware and driver issues (particularly sound card-related), but things are improving.
But none of this has much to do with rebooting. What are the most common reasons for rebooting? Because some program, newly installed update, or configuration change demands it, or because the OS or application stuffs up the system. In both cases, the volativity of RAM has no effect on whether or not you need to reboot.
Aha, so this is why the memory systems are funded by DARPA. Having discovered that their Windows NT controlled ships are dead in the water after a system crash, they're trying to make computers that reboot faster!
-Paul Komarek
holds the promise of reducing the time required to transfer and download multimedia content and other massive files
Yes, I'm sure that most of us are limited by the bandwidth to the memory when downloading large files.
{I can see how it will help with moving large files around in memory though...)
-- Mike
"Ignatiev said the new technology is about 1,000 times faster than flash, which is nonmechanical and currently the speediest memory on the market. "
since when has Flash memory been the speediest memory on the market?
That seems to be an obscene comparison!
Is not life a hundred times too short for us to bore ourselves? -Friedrich Wilhelm Nietzsche
Win98 is referred to as "Legacy Windows". The current version of Windows XP uses the NT kernel and doesn't exhibit the types of problems you have.
It might be helpful in the future if you are talking about problems in Win98 to use phrases like "I remember when...", or "Years ago..." to point out the fact that you were talking about the way things used to be, not the way things are.
It's called Linux and BSD. No, they're not perfect, but the only times I've ever had to reboot a BSD (be it Free, Open or Net) or Linux box was a) because of hardware failure or b) because I fscked something up (like when I was working on kernel modules).
this looks like too much responsibility for crappy software writers to handle. while rebooting is annoying, a freshly rebooted system has completely clean memory. if things just stay in memory permanently, programs with memory leaks will be a huge problem. obviously [i hope!] there would be a way to clear out the memory, but even so, the problem of having to reboot computers seems more of a software problem than a hardware problem.
Back in 1988, I had an 80286 with 640 KB of RAM (Back when it was enough for everybody.
refute that line pointedly. The machine took 30 seconds to POST, for one thing. (It just sat there forever reading "Checking stuck NMI.")
After it actually did start to "load DOS" it took a good 30-60 more seconds. By contrast, the 700 MHz PIII I have today loads win98 in *well* under a minute, from POST to actual keyboard/mouse responsiveness (That is, after its finished loading all 8 or so things that appear in the systray, etc, and is ready to launch more programs).
So, sorry, but with the increase in speed over the last 14 years as I've experienced it, boot times have lessened as well.
Well, I have a Sleep/Downtime replacement called "Coffee". It's relatively portable, lasts ages in it's pre-prepared form (may even be consumed in this form at a pinch), and only 5 minutes preparation time can replace up to 12 hours of Sleep. (Alternatives available)
Besides, as soon as I clocked the title I was worried. No more rebooting doesn't make much odds to *nix users. but while the average Joe Luser may not contribute jack shit to the Open Source community, money is money. Ask an average user (Box from high street chain, OEM windows install, clueless) "What ticks you off about Windows?" and none will say "The built in privacy violations. None will say "The bloated binaries." (hey, this thing is slow, better spunk away another grand on another one!) None will say "The attrocious security blunders". None will say "Contributing to a competitor-strangling monopoly."
It's crashes. Every time I've asked, anyway. Crashes are what tick off the ordinary Joe Luser. Losing work, watching scandisk (To quote Jack Dee, a mainstream comedian: "I didn't shutdown from the start menu because you crashed!), ruining marriage. Well maybe not the last one. But in a world without rebooting, how the hell are we going to sell *nix to the tech-clueless Windoze crowd? Face it, even "Free" software needs capital.
Ali
"Windows and Linux can co-exist on the same machine." - Microsoft Corporation.
If you read it closely, they said "manganite" (or something similar; my memory is starting to fade, too), not "magnatite." This suggests a manganese compound, that may have nothing to do with magnetism (and as noted, they said resistance, though it's yet to be said how the resistance is modulated to write bits)...
The past-trumpeted MRAM technology is closer to 'core' in design than this. However, one thing that nobody's touched on is that all three nonvolatile technologies allow programs to execute in place- there's no need to copy the program off storage and into RAM, when your storage *is* RAM. Much more efficient!
I haven't done the math to determine what "1,000 times faster than flash" works out to, but it's likely we'll still see maddeningly fast SDRAM buffers placed ahead of this new medium. That's the best of all worlds, though there's still the matter of keeping state in the event of a power loss..
Game consoles also don't have Operating Systems with the concept of multiple programs. If they even have a definable operating system at all. While they may need to be 'rebooted' they are also not really comparable to normal computers in that way. Since every program has its own OS contained on the cartridge/CDROM/DVD.
The only exception to this rule that I know of was the DreamCast, as all you ahd to do to change games was eject the cdrom, and then it went into a bios type screen where you could do memory card management and play audio cds.
After a reboot less software has run and left its mark on the system, causing less system turbulence.
This generally results in transient periods of stability immediately after the restart of the system as different components have not yet interacted in the complex ways you see on systems packed with traces of software that has already run.
Unless someone has yet another inspiration about how to ensure greater stability in the field of complex systems engineering we're doomed to at least a few reboots.
I thought that the two main memories in a computer were RAM and ROM.....since when did computers stop using Read Only Memory and start using this Magnetic Mass Memory stuff? by the way the article says, it leads me to believe that ROM is suddenly no more, that the two memories in a computer are RAM and MMM.....
Also, the article says
Nonvolitile doesn't mean nonmechanical. For example, a floppy disk is nonvolitile AND mechanical.Finally, this isn't really that new a technology. I read about this stuff in "Technology Review" a year ago.
I have a RH 7.2 box that has been up for almost 6 days. That is nothing special, either! It was up for somewhere around 40-50 days before I tried ironing a shirt with a tv, 3-75W incandescents, a 100W full-spectrum flourescent llght, and 2 computers (with monitors) all on in my bedroom.
Wow, when I write it out, I guess it's no surprise that I tripped the circuit breaker!
I know... what's the deal? Get on with it!
Here in the UK, in the editing department of the university where I work, mandatory breaks of 20 minutes away from the keyboard out of every two hours are enforced.
The idea is that this will prevent RSI complaints (whether or not you believe in RSI, the complaints are still real) and eye fatigue.
If a reboot is caused by a crash or hung application (as it often is) then although the break is only 5 minutes or so, the loss in productivity due to breaking one's train of thought is likely to be significantly longer...
-Jef.
-- What goes up must come down. Ask any SysAdmin.
Last MONTH I had to wipe and reinstall Win98 because the system was acting flaky, so no way can I say "Years ago..." or "I remember when..." because the problem is still with me here and now!
Before you tell me to replace that box, I work for a community college so if you are willing to forward a check to me I can replace that box, otherwise I need to keep using Win98. Also, the fact that many of the faculty and students I have to support are using Win98 means I need to deal with Win98 today, not yesterday.
Thinking about it, I only know of two people who even have XP and that is because they just bought brand new computers. I have also noticed that most of my non-techie friends are still running Win98. Again, if you are willing to pony up a check for buying them copies of XP I can get them to upgrade.
That, my friend, is how things are in the "Real World". Maybe you need to get out a little more and see what OS regular people are using these days (it ain't XP or even 2K)...
A man who wants nothing is invincible