Storage of the future
CyrixInstead
Stoke-on-Trent, England Icrontian
I was having a conversation with some friends about storage formats of the future, the may one day replace current storage formats such as smartmedia, minidisc and compactflash.
I personally am of the idea that solid-state memory will become useful in the future for many things. It's robust and would be ideal for such things as music devices and digital cameras. I know it's taking off in the karaoke industry in a big way, replacing those massive laserdiscs currently used.
Also, I have seen of IBM developing a storage format using holograms. Each of many layers is able to be used for storage, but it is still just one of those 'maybe we'll have these in 50 years' things.
What are your opinions as to what will be used in the future, and for what purpose will your favoured format used?
I personally am of the idea that solid-state memory will become useful in the future for many things. It's robust and would be ideal for such things as music devices and digital cameras. I know it's taking off in the karaoke industry in a big way, replacing those massive laserdiscs currently used.
Also, I have seen of IBM developing a storage format using holograms. Each of many layers is able to be used for storage, but it is still just one of those 'maybe we'll have these in 50 years' things.
What are your opinions as to what will be used in the future, and for what purpose will your favoured format used?
0
Comments
I think solid state is the way to go. The only problem is the way that its stored.
I would assume it would be based on light and crystals in order for the transmit speeds to be reasonable (reminiscent of Superman).
IBM actually has a new generation of punch cards as well:
http://www.research.ibm.com/resources/news/20020611_millipede.shtml
Kinda crazy in my opinion, but it gives you the idea of how nanotech is changing. But is still is a solid state system.
However, the more important question is how are they going to deliver all this data? I'm assuming fibre optic (or a variation).
I'm thinking all this info will be used for large db applications and serving huge amounts of data over networks. Perhaps, most computers will be terminals and people will carry around "keys" that can hold terabytes of data on something the size of a zippo. The user "plugs" in and can use any computer.
Even with all the differences, there's going to be a time when things are pretty much the same or interoperable. Even now I'm seeing a convergence of Mac and Win and they seem to be drifting towards UNIX (or a something similar). Companies will have their flavours. It will be like going to a phone booth.
The major tech that needs to be put into place is an awesome wireless system (it will cut down on infrastructure costs) that can support large amounts of data everywhere.
The problem is that there is no real rush for the industry to find better storage since there's no killer app that requires it.
Who does full-frame video editing remotely? The systems in place can handle a lot and the "bus" systems are very fast compared to the storage systems and people are relatively happy. Video on demand is done through copper/fiber optic or satellite.
Can you think of anything that would require the kind of storage I'm dreaming about? Other than terminals with people carrying around their systems.
What I want to know is, in the future will we be able to get away from the spinning disc? What if one laser, just by changing it's color/intensity/etc. could access various parts of a sphere or disc or cube without any moving parts.
I think it's going to get more exciting. The only problem is economics. Who's going to upgrade to such a system other than large datacentres with adventurous managers.
If they can figure out a reliable way to transport a lot more data then, and only then, will there be a need for larger storage.
I don't know how many people remember the old days of the web. All files were so small and most sites were only text so that they could travel over 9600 baud modems. Nowadays, most people don't think twice about downloading a 100meg album of an artist that they might like (or a 1 gig download of a movie). Now, they want faster lines to put in hi-def streaming with hi-fi sound.
It's not the message, its the medium upon which it travels.
Not completely true at the moment. Take our current hard drives. SATA allows for 150MB/sec but the drives can only reach 50MB/sec at a push (unless you are RAIDing in which case you would be using 2 SATA channels anyway). So the media needs to vastly improve. I mean, solid state cards etc are only around 1-2MB second, which is dire to be honest.
NS
//EDIT//
I guess thrax is basically talking about what I just asked, as well as MarkTAW
Near future though... I don't see much change. I do see something similar to the Opteron though. I can see use using the extra RAM as a storage device using a new technology which I cannot remember what it was called, but it is next after DDR, non volitile though so you won't loose the information stored in it after a reboot.
However, I believe that a good system is a virtual string. I think some police started using it a few year back, I have no idea how it works though, but it seems like something you would only see in a movie or something.
Non-volatile random access memory.
Currently present in very miniscule amounts, and much slower than what memory developers are planning.
Once NVRAM becomes mainstream and cheap to produce, our HDDs will move into the nanosecond seek time range, as opposed to millisecond.
NS: you're still talking about transport of the data. It's not the platters' fault that the needle can read fast enough. I would consider everything from storage to processor and then processor to display as the "bus", in loose general terms.
No, I think the kind I am speaking of is called MRAM, becuase it is already faster than current RAM.