Increasing The Capacity Of Harddrive

[chewy0914]

Program a hard drives way of writing data?
Or a software that can do it.


Raid hard drives:
Can this already be done?
Half+ data on different parts of disc?
4x+/- with say 2TB drive with data placed along the disc repeated 4 times?
making 4x512gb with 4 spots to read from? 3 backups?


--Duplicate data on opposite sides of the disk to
have the disk to only have to spin half/quarter way to read any data.



2. Set the harddrive to write in sets of data instead of 1/0.
0=00,1=01 top of platter certain partition?
0=01,1=11 bottom of platter different partition?
Or a have software to interpret the data that way.
Possibly to be extended to multiple platters with greater variations.
Using 000-111(9),,,,0000-1111(16)(+) depending on the number of hard drives in raid or platter count.

example- 512gb hd use 00-11 using 4 partitions/platter/other sets. Making a 512gb Hdd into a 1tb hard drive.
Also half/quarter the data location on the disk to increase read speed?
Creating with a 00-11 set a 256gb or 512gb hdd with 4 duplicates for reads instead of 128gb or 256gb duplicated.

Maybe 6tb 6 platter hd, made into a
12/ 24tb+/-? hard drive?

Also I would like to partition or
just use a software or whatever to try to do say 100+ partitions/or read areas/other set.

10 digit+ 0000000000-1111111111
Increasing the storage amount by up
to 10X depending on how data is stored?
If that is to much for a direct hardware allocation possible a software that assigns values like that to increase data storage amount ?100+fold or more?


Other- Could the 00-11 be stacked?
I'm not sure if the hard drives do this already or not, but if a harddrive
was programed to place top neg=00, top pos=01
bottom pos=11,bot neg=11. Could you add another layer in software
to make code at a higher level splitting into 4 way coding?
Doubling or many more times the capacity of a drive?



It would be nice if harddrives were made like Cpus'
Cpu/Ram? access type with material
from hardives deposited in the cpu die.

Layered cpu style.
Resulting Hdd size?
-
-
Replace ram and hd?
Cheap intel CPU?$40?
1 for Hdd,
1for Cpu/Gpu or one of each?
1 for Ram if needed or wanted?
1 for all controller,wifi,other?


Other Possibilities?

Use a cypher/other for the data? compression?

0000 0001 0011 0111
=0 =1 = 01 =11

1111 1000 1100 1110
=111 = 00 =101 =110


0101 1010 1001 0110 1101 1011 0100 0010
=10 =001 =010 =011 =110 =1011= 0100 =0010

Code or other in program to select and assign values depending on most present code combos in data?

01 10 0011 1101
0 1 11 10


Specify different types of Hdd sector cluster types?
example:0000-1111+/- to specify type of cluster any use?
0000=a 0001=b 0011=c 0111=d.........


Third variable in HDD non negative or positive--
-
-
-
Use a bit based way of assigning data a value. Even and odd numbers?

1 2 3 4 5 6 7 8 9 10 11 12 13
0=00 0=01 0=00 0=01.............................
1=10 1=11 1=10 1=11..............................

1 =10 if bit 3072 is 0 = 10 if 1= 1110
4096/4=1024, 4096-1024=3072(last quarter)
if 3926 =0 1110 if 1 =1110 0110(or any combo 1001 0110, 1110 1001)
As many more times as may work.

Cluster labeling?

if in an A cluster doubles to 1110 0001 1100 0110
If in a B cluster changes to 1100 0011 0011 0111
if in a C cluster changes to .......
if in a D cluster check specific cluster (cluster x-541)(or0000000001000) if 01 in (cluster x-541)-(4056)
code or double or change or other =..............
Or any other combo?


2bit base sets instead of 1bit.

1,2 3,4 5,6 7,8
11=1000 11=0001 11=0011 11=1100
00=1111 00=0000 00=1100 00=1011
10...............................
01...............................

4bit base sets instead of 2bit.
1243 5678 9101112 13141516
0000 0000 0000 0000
= = = =
01010101 10101010 11110000 00001111
......................................
and on...It would seem that since data I'm pretty certain is in even numbered(8bit?) amounts that any amount of even sets should work.

2+ materials in Hdd , Option to create weaker and stronger fields?
4 types+ of bits= what?
1 neg=1/ 2 neg=01/ 3neg=??
1 pos=0/ 2 pos=10/ 3pos=??


Solid State Quantum Computing?
Use a hdd type cpu(cpu with hdd material in it) and use solid state bit holding in cpu for quantum computing?
quantum computer qubit--- pos and neg at once?
1=neg 2neg 3
1=pos 2pos 3



I looked at hard drives in the light of what parts of the pc slows it down.

What are the limits of disc based hard drives?

I found spin speed? Arm movement? maybe circuitry?
Size of the disk?

I read a few post on how it has been tried to make them faster.
Is it just cost or what is limiting hard drive manufacturers?

Thoughts--

2 arms in a hard drive? both able to read and write at the same time?

Multiple arms so any part of the hard drive can be read from in the same amount of time?

Disc- Rotated from the outside keeping the center of the disc?

Magnetic rotation instead of motor based? mag lev stabilization?
Spinning and holding from outside/middle/inside of disc?
Or even roller/other based?

different kind of arms?
or no arm at all?
Arm that allows each layer of arm to be in whatever position from tension or other?
1 arm motor moves layer of head each head in set ways?
*

Strip/other going the whole length of hard drive able to read or write from entire disc surface at once?
*

*multiple arms/strips that do do don't move to cover entire area.

Better aerodynamics inside of hard drive reduce friction?
Not just helium-- internal structure.
Increase speed from air flow produced?

Better heat disbursement?
Air flow in around hard drive?

Heat Sink?

Sealed area around disc..
Air flow around area in harddrive..
Reverse flow blow out dust?


Single surface that covers entire disc?
Be able to read or write from any part of disc at once?
Rotating or not?
Example:
Cones multiple read heads per cone/ mechanical/other redirect
for multiple locations?

head head
\ / \ _ /
\ / \_!_/switch
disc disc

Would the optical group be able to make a disc that can all be read at once with light?
reflection?
Pass through for whole disc reading?
0=spot
1=hole
or whatever?


What is the limit in write speeds?
Multiple discs and heads on 1 arm?
Write each bit of data one after another on different disc?
Then put them together or however in down time?

*
What capacity of hard drive could a chip manufacturer make in a cpu or smaller die size using current fab 14nm or larger? (Ram?)


Update:

Multi materials in HDD- Each different output value-
one material in HDD= neg-10, Pos 10
other material = neg-20, pos-20
+ more types/other


Track value assigning.
odd tracks even tracks (other sets)
Even track neg=00(0000+/-) pos=01
odd track neg=11 pos=10
Other-0000-1111,00000000-11111111+/1


Applying similar over multi tracks

1-2,00-11............
3-10,0000-1111...........
1-2,00-11 + 3-10 0000-1111.......+(other)
1,0 over tracks 1-31,32-63 0000000-11111111
outputting 00000000-11111111 depending on track


Track Assigning+Disk Top Bot/Pos/Neg+ Other Material+Other
X multiplicative data capacity




Update 2

Router Encryption
1111111111111111111, 11111111111111, 11111111111,

origin---router--destination
Other\ \
--------\-\-router-\---/destination/
---------\ \-router-\-/--destination/
----------\-router--\-/--destination/
1111111111 sent to router
Tor like network?

Split data or part of with set amounts physically-
11111111111=010101010101010101+ split into 111111111, 1111, 111111
send to 3+ servers back and forward or not-------- arrives 1111111, then 1111, 111111
or split more into 1111 1111111 1111111 1111 11111.

Without knowing hardware patterns timing other data would only be sets
of 111111111111111111
111111111111,1111111,1111,11111111,11111111,111111
changed to 1111111111,11,1,111111111,111,11111111,11111111 or other
has to arrive in any set order to make.
010101110011011010101+


Use similar in hardware or software in harddrive.
Hardware or software in ram together or seperate/other.

HDD=
111111111111,11111111111,111111111,111111111,111111
Hardware when reading =111111111111,11111111111,111111111,
Turns into01010101010100010101010101+
runnng 0101010 into ram or 111111111 another level 10101010, or
11111111111 to 010101010101010+

Update:
Energy Savings:
0=no pulse/light/electricity/other

Split data or part of with set amounts physically-
0000000000=010101010101010101+ split into 000000 (1), 0000 (1), 000000 (1)
send to 3+ servers back and forward or not-------- arrives 000000, then 0000, 000000
or split more into 0000 000000 0000000 0000 00000.

Without knowing hardware patterns timing other data would only be sets
of 00000000000000
0000000000,0000000,0000,000000000,00000000,000000
changed to 0000000000,00,0,000000000,000,00000000,00000000 or other
has to arrive in any set order to make.
010101110011011010101+


Infinite Harddrive-
Electrons sets or not, just pure electrons,

Gate System- channels that change.

Start system 111111111111111111111111111111111111111111111
11g1111111g11111111g1111111g111111111g1111111g11111111g111
As data is added
1111111111g111111111g111111111g111111g1g1g1g11111111g1111
Change patterns of gates/other many cycle gives differing kinds of data?


Update:
Energy Savings

Gate System- channels that change.

1 electron/photon/other-------------------------------sent--------------------
Start system 1-00000000000000000000000000000000000000000
00g000001g0000000g0000g00000000g000000g000000g000
As data is added, changing timing
1-00000g0000000g000000g000000g0g0g0g0000000g000000
Change patterns of gates/other many cycle gives differing kinds of data?

Timed and set up to send one electron photon other per gate change and running in parallel multi cycle

 

[dave_the_nerd]

Program a hard drives way of writing data?
Or a software that can do it.


Raid hard drives:
Can this already be done?

Yes.

Half+ data on different parts of disc?
4x+/- with say 2TB drive with data placed along the disc repeated 4 times?
making 4x512gb with 4 spots to read from? 3 backups?

All those things are options.

--Duplicate data on opposite sides of the disk to
have the disk to only have to spin half/quarter way to read any data.

Bad idea - duplicate data on opposite sides of the disk means when that data is modified, it has to be rewritten twice.

2. Set the harddrive to write in sets of data instead of 1/0.

They already write a sector at a time, minimum.

0=00,1=01 top of platter certain partition?
0=01,1=11 bottom of platter different partition?
Or a have software to interpret the data that way.
Possibly to be extended to multiple platters with greater variations.
Using 000-111(9),,,,0000-1111(16)(+) depending on the number of hard drives in raid or platter count.

Data is already stored on all available surfaces within a drive. There's no "free" space on the opposite side of the platter. (That space is already in use.)

example- 512gb hd use 00-11 using 4 partitions/platter/other sets. Making a 512gb Hdd into a 1tb hard drive.

See above.

Also half/quarter the data location on the disk to increase read speed?

This is called "short stroking."

Creating with a 00-11 set a 256gb or 512gb hdd with 4 duplicates for reads instead of 128gb or 256gb duplicated.

Yeah, you can do that. But it only speeds up reads, not writes.

Maybe 6tb 6 platter hd, made into a
12/ 24tb+/-? hard drive?

You mean a RAID array.

Also I would like to partition or
just use a software or whatever to try to do say 100+ partitions/or read areas/other set.

You can make a lot of partitions.

10 digit+ 0000000000-1111111111
Increasing the storage amount by up
to 10X depending on how data is stored?
If that is to much for a direct hardware allocation possible a software that assigns values like that to increase data storage amount ?100+fold or more?

Math does not work that way.

Other- Could the 00-11 be stacked?
I'm not sure if the hard drives do this already or not, but if a harddrive
was programed to place top neg=00, top pos=01
bottom pos=11,bot neg=11. Could you add another layer in software
to make code at a higher level splitting into 4 way coding?
Doubling or many more times the capacity of a drive?

They don't work that way either. Each piece of HD real estate (they're already using both sides of the platter) can be either positive or negatively charged. That's where the zeros and ones come from. You can't store a bit in both the positive and negative "space" of a single memory location.

It would be nice if harddrives were made like Cpus'
Cpu/Ram? access type with material
from hardives deposited in the cpu die.

That's called NAND. It's used in SSDs.

Layered cpu style.

Samsung makes stacked NAND.

Resulting Hdd size?

As big as you can duct tape NAND together.

Replace ram and hd?
Cheap intel CPU?$40?
1 for Hdd,
1for Cpu/Gpu or one of each?
1 for Ram if needed or wanted?
1 for all controller,wifi,other?


Other Possibilities?

Your HD has a CPU on it already.

Use a cypher/other for the data? compression?

0000 0001 0011 0111
=0 =1 = 01 =11

1111 1000 1100 1110
=111 = 00 =101 =110


0101 1010 1001 0110 1101 1011 0100 0010
=10 =001 =010 =011 =110 =1011= 0100 =0010

Cyphering it would make it bigger. Compression and dedup is done live by most file systems.

Code or other in program to select and assign values depending on most present code combos in data?

01 10 0011 1101
0 1 11 10

That is how compression works, yes.

Specify different types of Hdd sector cluster types?
example:0000-1111+/- to specify type of cluster any use?
0000=a 0001=b 0011=c 0111=d.........

No benefit - you still need the same physical space to store the same number of ones and zeros.

Third variable in HDD non negative or positive--
-
-
-
Use a bit based way of assigning data a value. Even and odd numbers?

1 2 3 4 5 6 7 8 9 10 11 12 13
0=00 0=01 0=00 0=01.............................
1=10 1=11 1=10 1=11..............................

1 =10 if bit 3072 is 0 = 10 if 1= 1110
4096/4=1024, 4096-1024=3072(last quarter)
if 3926 =0 1110 if 1 =1110 0110(or any combo 1001 0110, 1110 1001)
As many more times as may work.

Ternary computers work mathematically. They're just too complicated to build. And using base-3 data storage would make it much more likely to fail or error. (The reason it's easy to tell the difference between a 0 and a 1 is that they're opposite extremes. Introduce gradations into that and any interference results in immediate and undetectable corruption.)

Cluster labeling?

if in an A cluster doubles to 1110 0001 1100 0110
If in a B cluster changes to 1100 0011 0011 0111
if in a C cluster changes to .......
if in a D cluster check specific cluster (cluster x-541)(or0000000001000) if 01 in (cluster x-541)-(4056)
code or double or change or other =..............
Or any other combo?

That will also no increase capacity.

2bit base sets instead of 1bit.

1,2 3,4 5,6 7,8
11=1000 11=0001 11=0011 11=1100
00=1111 00=0000 00=1100 00=1011
10...............................
01...............................

4bit base sets instead of 2bit.
1243 5678 9101112 13141516
0000 0000 0000 0000
= = = =
01010101 10101010 11110000 00001111
......................................
and on...It would seem that since data I'm pretty certain is in even numbered(8bit?) amounts that any amount of even sets should work.

It's already arranged in 512 or 4096 bit sectors.

2+ materials in Hdd , Option to create weaker and stronger fields?
4 types+ of bits= what?
1 neg=1/ 2 neg=01/ 3neg=??
1 pos=0/ 2 pos=10/ 3pos=??

See above about ternary.

Solid State Quantum Computing?

Call me in 50 years.

Use a hdd type cpu(cpu with hdd material in it) and use solid state bit holding in cpu for quantum computing?
quantum computer qubit--- pos and neg at once?
1=neg 2neg 3
1=pos 2pos 3

I looked at hard drives in the light of what parts of the pc slows it down.

The HDD is the thing slowing everything else down.

What are the limits of disc based hard drives?

I found spin speed?

That's one. Spin them any faster than 15k and they explode. Slower disks are easier to manufacture with looser tolerances, less expensive to produce, and easier to cram more data on them.

Arm movement?

Seek time is also a limitation.

maybe circuitry?

Not in practical terms, no.

Size of the disk?

High density platters go faster, yes.

I read a few post on how it has been tried to make them faster.
Is it just cost or what is limiting hard drive manufacturers?

Yes.

Thoughts--

2 arms in a hard drive? both able to read and write at the same time?

There's already one r/w head per platter. Adding a second set of "arms" created cost and reliability problems. I think it was Quantum that did it, or maybe Connor.

Multiple arms so any part of the hard drive can be read from in the same amount of time?

See above. Multiple arms = bad idea.

Disc- Rotated from the outside keeping the center of the disc?

Then where do you mount the arm?

Magnetic rotation instead of motor based? mag lev stabilization?

That would nuke the data on the magnetic disk.

Spinning and holding from outside/middle/inside of disc?
Or even roller/other based?

Physical contact is to be avoided when possible. It will make the platters explode.

different kind of arms?
or no arm at all?
Arm that allows each layer of arm to be in whatever position from tension or other?
1 arm motor moves layer of head each head in set ways?
*

Too much mechanical complexity to be economical.

Strip/other going the whole length of hard drive able to read or write from entire disc surface at once?
*

*multiple arms/strips that do do don't move to cover entire area.

Same.

Better aerodynamics inside of hard drive reduce friction?
Not just helium-- internal structure.
Increase speed from air flow produced?

Rotation speed is limited by the tensile strength of the platter material. (Exploding platters.) There's no aerodynamics to speak of at that scale.

Better heat disbursement?

Wouldn't matter.

Air flow in around hard drive?

That'll keep heat down.

Heat Sink?

HD coolers have been around forever.

Sealed area around disc..
Air flow around area in harddrive..
Reverse flow blow out dust?

All of the above.

Single surface that covers entire disc?
Be able to read or write from any part of disc at once?
Rotating or not?
Example:
Cones multiple read heads per cone/ mechanical/other redirect
for multiple locations?

head head
\ / \ _ /
\ / \_!_/switch
disc disc

You're repeating yourself.

Would the optical group be able to make a disc that can all be read at once with light?
reflection?
Pass through for whole disc reading?
0=spot
1=hole
or whatever?

They aren't optical. They're magnetic. Optical media (like CDs) tend to be slower.

What is the limit in write speeds?
Multiple discs and heads on 1 arm?
Write each bit of data one after another on different disc?
Then put them together or however in down time?

With a big enough RAID array, with enough disks operating in parallel, there isn't a real limit to speak of.

*
What capacity of hard drive could a chip manufacturer make in a cpu or smaller die size using current fab 14nm or larger? (Ram?)

For an individual chip? I think it's 128GB on a single chip. That's been going up, and of course the chips are smaller than a full size CPU. There's a size/cost calculation based on fab error rates and such.

 

[Railgun]

I think you took the bait.

 

[hojnikb]

*
What capacity of hard drive could a chip manufacturer make in a cpu or smaller die size using current fab 14nm or larger? (Ram?)

Actually, you can already put 256GB on a single package (840EVO msata does that, so does sandisk 512GB sd card).

 

[dave_the_nerd]

I think you took the bait.

'm bored.

 

[KingFatty]

Is this an example of trying to reinvent the wheel?

 

[dave_the_nerd]

Is this an example of trying to reinvent the wheel?

This is what happens when somebody reads this and then smokes weed.

You know... I need to start again.

 

[Batmeat]

I'll admit that I only read half if the OP. The only way to add space is to add physical platters. The only other way is to implement compression algorithms to read/write data in a compressed state.

That being said, I didn't read the entire OP