Is a small disk or partition better than a large one full of files?

[jhmac77]

I am trying to get several peoples opinion. I believe that a small disk that is bootable similar to a ssd is better that a large one full of data. There is less disk to deframent and it should be faster. The larger data disk will contain most of the files many of which will not be used a large part of the time. One can keep larger files such as pictures or videos in that data disk. I probably prefer a small ssd(less than 100GB) but am not sure that is the best way to go right now. Later on when the price and quality of ssd goes down I could substitute that drive for the small hd partition or drive. Anyone have a good article or information about this.

Jim

 

[kado897]

My opinion, for what it is worth.

A separate disk or partition for Windows and your programs is the way to go. You can store all your data on a second disk or partition. For simplicity don't try moving the User folders, instead use libraries to gather your files together. This means of course that you will still get some data on your system drive but not much, and it will be the most used data in the Appdata folder. I personally have only a few GB of data held that way and hundreds of GB on separate drives. The really big advantage for me is that backing up the Windows partition is relatively quick so is not such a pain to do frequently.

 

[gregrocker]

Kado +1. The way to go.

If Win7 becomes irreparable you simply reimage to C and your data is safe and current in it's own vault on D, easily relinked by rightclicking each User folder to add it to related
Library - Include a Folder - Windows 7 Forums

For this reason it makes it so much simpler to sort all of your data into the User folders, copy them to the new data partition, link to library, when it confirms they're present in that library on D, delete the User folder contents (only) on C. If fiiles land in C just drag them to D.

 

[jhmac77]

I saw the Wikipedia article on the disadvantages of partitioning and I am not sure I understand. See below:

Disadvantages of multiple partitions

Creating more than one partition has the following disadvantages, as compared to having a single partition spanning the same disk area:

• Reduces the total space available for user storage on the disk, as it forces the operating system to duplicate certain file system administration areas on the disk for each partition.
• Reduces overall disk performance on systems where data is accessed regularly and in parallel on multiple partitions, because it forces the disk's read/write head to move back and forth on the disk to access data on each partition[2][3] and to maintain and update file system administration areas on each partition. It also prevents disk optimizers from moving all frequently accessed files closer to each other on the disk, which could reduce the number and distance of required head movements. Files can still be moved closer to each other on each partition, but those areas themselves will still be far apart on the disk. (See "short stroking" considerations above.) This issue does not apply to Solid-state drives as access times on those are neither affected by nor dependent upon relative sector positions.
• Increases disk fragmentation because it lowers the average size of continuous free blocks on each partition - as compared to a single partition of the same overall size - after the same amount of data has been written to them.
• May prevent using the whole disk capacity, because it may break free capacities apart.[4] For example, if you have a disk with two partitions, each with 3 GBs free (hence 6GBs in total), you can't copy a 4GB DVD image file on that disk, because none of the partitions will actually provide enough space for that - even though you have more than enough free capacity in total on the disk. If the same files on those two partitions would have been stored on a single partition spanning the whole disk, then the 4GB file could be easily stored in the 6GB of free space.
• Slows down moving data between different parts of the same physical disk.[5] When moving data from one partition to other, the operating system actually has to copy the contents of the data file, even though it still remains on the same device in a single copy/instance. When using a single partition, moving data between directories will only require altering the file system administration areas, and the actual contents of the file will not be copied or moved inside the disk/device, thus resulting in a significantly faster completion of the operation.
• Hurts portability and might impose constraints on how entities might be linked together inside the file system. For example: the NTFS file system allows hard links to be created only as long as both the link and the referenced file reside inside the same volume/partition.[6] Also under Windows if you're referencing a file on another partition, you can do that only by specifying the partition's assigned drive letter - which, however, might change with time and depending on the drives installed. This renders references invalid and dependent on actual drive letter assignment, which is not an issue if you have to reference files/directories only on the same partition, as in this case you can use directory-relative or root-relative references, without including the drive/partition letter

 

[kado897]

Lets see.

1. This is true in that it will effectively create a second NTFS system with whatever overheads this has. It will not be significant with today's disk sizes.
2. True. If your data is being accessed at the same time as windows files then the disk head will have to reposition more often and probably over a larger amount. A separate disk is preferable.
3. I wouldn't think that was important given the size of today's disks.
4. Ditto.
5.If you are moving files from one partition to another this will be much slower. This should not be an issue as you have different types of files on each partition so you will not be moving them very often.
6. So what. How may symbolic links will you be creating. Drive letters tend to stay locked to the partition and even to external drives.

 

[DeaconFrost]

I've always preferred to have a small faster drive as the system volume and then a larger drive for user data. I don't ever redirect folders, though. No point in doing so. However, there's a point where it can become wasted thoughts and effort, especially if you are only considering HDDs. Ideally, right now, have a single SSD for the OS and programs, and then a large HDD for data.

 

[jhmac77]

I will be considering ssd's

 

[pparks1]

I've always preferred to have a small faster drive as the system volume and then a larger drive for user data. I don't ever redirect folders, though. No point in doing so. However, there's a point where it can become wasted thoughts and effort, especially if you are only considering HDDs. Ideally, right now, have a single SSD for the OS and programs, and then a large HDD for data.

I'm going to agree with this completely. I run an SSD for my OS, and a spinner for my data. I image C regularly. I don't use the My Documents stuff on my local computer as I have a file server at home accessible via a mapped drive...so that is a moot point for me.

My large disk, is a 1TB WD Caviar Black and is indeed 1 partition.