Time to merge memory?
-
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
-
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
My first though is about an idiot user who didn't delete the right file and he is wondering why he cant start the application again. If RAM and hard drive merge its gonna be a pain because everyone has to clean the not needed data. But if there is a program cleaning this for you and the hard drive is fast enough (there is no drive with enough speed for "running windows and doing some other stuff at the same time" created yet) This reminds me for a really old joke. The son of Bill Gates comes to him and ask him -Dad show me the multitasking of windows -Just wait for my diskette to format and i will.
-
My first though is about an idiot user who didn't delete the right file and he is wondering why he cant start the application again. If RAM and hard drive merge its gonna be a pain because everyone has to clean the not needed data. But if there is a program cleaning this for you and the hard drive is fast enough (there is no drive with enough speed for "running windows and doing some other stuff at the same time" created yet) This reminds me for a really old joke. The son of Bill Gates comes to him and ask him -Dad show me the multitasking of windows -Just wait for my diskette to format and i will.
I think that's the case already - delete the wrong file and your application won't start. As you say speed is an issue but that's where you'd have caching. With 16GB of RAM being so cheap that would act as your cache. (As in no longer RAM as we know it, but cache memory). All which is happening, in a way, is we're making the caching automatic whereas we do manually at the moment through loading things into processes and putting them back on to disc manually when we're finished. Applications which didn't load - because they are already there - it has an appeal to it.
Regards, Rob Philpott.
-
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
Well there's still a place for old fashioned hard disks: you can get 1TB of storage a lot more cheaply in a good old magnetic spinny thing than you can in solid state memory (either RAM or persistent SSD type). I think there could be a place for semi-permanent memory, though, which is more persistent than RAM, but not long term storage. Temporary files, hibernation state, and large application resident objects could live in this intermediate-lifespan storage area, which would be physically implemented with a SSD and could be up to 128 or 256 GB in the near future as a standard part of a package. This would blur the boundary between files and memory objects, but that's already the case in an opaque way: temporary files, although nominally 'on disk', are usually in memory in a modern OS; on the flip side, memory mapped files give us large 'in memory objects' which are actually not in memory.
-
My first though is about an idiot user who didn't delete the right file and he is wondering why he cant start the application again. If RAM and hard drive merge its gonna be a pain because everyone has to clean the not needed data. But if there is a program cleaning this for you and the hard drive is fast enough (there is no drive with enough speed for "running windows and doing some other stuff at the same time" created yet) This reminds me for a really old joke. The son of Bill Gates comes to him and ask him -Dad show me the multitasking of windows -Just wait for my diskette to format and i will.
Argonia wrote:
(there is no drive with enough speed for "running windows and doing some other stuff at the same time" created yet)
We got'em... DRAM based SSD's See wikipedia[^]
-
Well there's still a place for old fashioned hard disks: you can get 1TB of storage a lot more cheaply in a good old magnetic spinny thing than you can in solid state memory (either RAM or persistent SSD type). I think there could be a place for semi-permanent memory, though, which is more persistent than RAM, but not long term storage. Temporary files, hibernation state, and large application resident objects could live in this intermediate-lifespan storage area, which would be physically implemented with a SSD and could be up to 128 or 256 GB in the near future as a standard part of a package. This would blur the boundary between files and memory objects, but that's already the case in an opaque way: temporary files, although nominally 'on disk', are usually in memory in a modern OS; on the flip side, memory mapped files give us large 'in memory objects' which are actually not in memory.
I prefer to keep my temp or swap (depending the OS in use) on my RAM... by creating a RAM disk... It's safe and recommended highly if you're on an SSD
-
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
Rob Philpott wrote:
I don't know how wide address buses are on 64 bit processors,
The I7 has a 192-bit memory bus width and the Phenom II x6 1100T has 128-bit memory bus width. From Tom's Hardware[^]
Rob Philpott wrote:
I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space.
That definitely does seem reasonable, except that SSD's have a lifetime of about 5 years (somewhere on ArsTechnica was an excellent article on how SSD's work and why they have limits to their life), about the same as a hard disk, and DRAM etc. has a much longer life expectancy. Also, it still has to be worked out how to designate some data as persistent and other data as not persistent, hence why we have a file system to begin with - the user decides that, not the computer. But regardless of the technical issues, I think you have a good point as to where things might be heading. Marc
Testers Wanted!
Latest Article: User Authentication on Ruby on Rails - the definitive how to
My Blog -
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
Maybe. But I don't really like the sound of "automatic persistence". That will lead to a ton of crap accumulating due to programs refusing to fix what used to be harmless memory leaks. A nicer memory-mapped-file API would be nice of course. Putting a file at a fixed address has problems though: how much space do you reserve for it? Some files get pretty big, and you're likely to have a TON of small files.. that would take up a ton of address space if everything gets a block the "maximum file size". Address space is about 48 bits now IIRC, for example with a maximum size of 128GB that gives you space for just 2048 files.
-
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
Welcome to Back to the Future.... 1979, to be precise. That was when the System/38 was introduced by IBM with a 48-bit addressing scheme. The system did away with notions of main memory being different from disk storage. Disk was managed as an extension of main memory. But then, according to the "we-know-best" Unix crowd, if it is not in Unix, it is not worth having. And in computer system architecture courses in Brain Washing Facilities -- sorry, Universities -- the only architecture worth learning teaching is the Intel '86 series!
-
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
Rob Philpott wrote:
and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors
That's a heck of a "gloss" you'd need there! Processors already cache your HDD accesses, so I suspect that to get a significant "real" mapped space, you would need huge amounts of RAM - and the cost of that would be horrific. Think about it: how much is 16GB ram? I bought a 2TB USB3 HDD last week for £67...that's a similar cost!
The universe is composed of electrons, neutrons, protons and......morons. (ThePhantomUpvoter)
-
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
-
“Education is not the piling on of learning, information, data, facts, skills, or abilities - that's training or instruction - but is rather making visible what is hidden as a seed”
“One of the greatest problems of our time is that many are schooled but few are educated”Sir Thomas More (1478 – 1535)
-
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
Basic RAM runs at ~20GB/sec, 2 x SSD in RAID0 does 1GB/sec Still a lot slower, but sure a lot better for virtual memory. I cannot fathom the amount kittens killed daily by my laptop hard drive... :((
-
Rob Philpott wrote:
I don't know how wide address buses are on 64 bit processors,
The I7 has a 192-bit memory bus width and the Phenom II x6 1100T has 128-bit memory bus width. From Tom's Hardware[^]
Rob Philpott wrote:
I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space.
That definitely does seem reasonable, except that SSD's have a lifetime of about 5 years (somewhere on ArsTechnica was an excellent article on how SSD's work and why they have limits to their life), about the same as a hard disk, and DRAM etc. has a much longer life expectancy. Also, it still has to be worked out how to designate some data as persistent and other data as not persistent, hence why we have a file system to begin with - the user decides that, not the computer. But regardless of the technical issues, I think you have a good point as to where things might be heading. Marc
Testers Wanted!
Latest Article: User Authentication on Ruby on Rails - the definitive how to
My Blog192 bit?? I'm presuming that's the width of data and address bus combined but even then I'd expect 128 bits as a maximum. I need to read up, clearly...
Regards, Rob Philpott.
-
Basic RAM runs at ~20GB/sec, 2 x SSD in RAID0 does 1GB/sec Still a lot slower, but sure a lot better for virtual memory. I cannot fathom the amount kittens killed daily by my laptop hard drive... :((
leppie wrote:
I cannot fathom the amount kittens killed daily by my laptop hard drive
Ask Ninja to count them in the morning and then count them again in the Evening; should be a fairly accurate measure. OT You know she tweeted about JB without warning? A guy could lose his lunch with that kind of disgusting imagery!
Reality is an illusion caused by a lack of alcohol
-
leppie wrote:
I cannot fathom the amount kittens killed daily by my laptop hard drive
Ask Ninja to count them in the morning and then count them again in the Evening; should be a fairly accurate measure. OT You know she tweeted about JB without warning? A guy could lose his lunch with that kind of disgusting imagery!
Reality is an illusion caused by a lack of alcohol
Nagy Vilmos wrote:
You know she tweeted about JB without warning?
Thank goodness I missed that one ;p
-
Nagy Vilmos wrote:
You know she tweeted about JB without warning?
Thank goodness I missed that one ;p
Sort it out will you! :laugh:
Reality is an illusion caused by a lack of alcohol
-
leppie wrote:
I cannot fathom the amount kittens killed daily by my laptop hard drive
Ask Ninja to count them in the morning and then count them again in the Evening; should be a fairly accurate measure. OT You know she tweeted about JB without warning? A guy could lose his lunch with that kind of disgusting imagery!
Reality is an illusion caused by a lack of alcohol
Nagy Vilmos wrote:
Ask Ninja to count them in the morning and then count them again in the Evening; should be a fairly accurate measure.
I only got her to recite primes up to 51 so far. Even with O(1) factorization, it might not be enough!
-
192 bit?? I'm presuming that's the width of data and address bus combined but even then I'd expect 128 bits as a maximum. I need to read up, clearly...
Regards, Rob Philpott.
-
Now that SSDs are becoming more common and with the rise of 64 bit architectures I've been wondering whether we might see the day soon where we don't distinguish between RAM and 'disc' storage, instead just mapping everything into a vast address space. I don't know how wide address buses are on 64 bit processors, but I doubt they are 64 bits but if they are (and they could be), and we gloss over problems to do with speed of access it seems possible. You'd need some sort of caching - a bit like on processors. So ingrained is the idea of RAM and disc storage its hard to imagine what it would be like without it. Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system. Interesting thought - well if you're me anyway.
Regards, Rob Philpott.
Rob Philpott wrote:
Persistence would disappear, as once you've instantiated an object there it would stay, indefinitely. I don't know what that would mean for the traditional idea of a file system.
Presuming that they do become equal in replacement that would still not happen. The bus is still serial. Ip is still serial. Application bottlenecks still exist. Geographic redundancy needs still exist, etc.
