Windows 10: Performance: 2 HDD VS 2 Partitions in 1 HDD

I have
C: for OS,
D: for my documents.

I use for game dev, mostly using Phoshop, Unity

Which one of the follow options have better performance?
2 HDD (1 for OS, 1 for documents)
1 HDD 2 Partitions (1 for OS, 1 for documents)

:)

 

Max disk size for Windows 10

Bios/MBR based drives with NTFS are limited to 2 TB partitions:

BIOS/MBR-based hard drive partitions

While deploying windows to HDD with GUID Partition Table (GPT) ,instead of MBR, each partition can have a total of 18 exabytes:

UEFI/GPT-based hard drive partitions

It is the partition table used on the HDD that will limit that size of the supported disk.

 

partitioning 2 TB Hdd

It has three replies, waiting on input from you.

Participate.

partition hard drive

 

2 HDD will generally be better - in addition you can backup images of your system configuration on the second HDD.

You should consider a SSD or SSD + HDD for vastly improved speed.

 

It rather depends what you are doing. You will always get marginally better performance with two hdds as they have two data seek heads rather one (the head has to move around more if only one drive).

In most cases, the difference is barely noticeable, but where you get a real bottleneck is when physically copying a lot of data from one partition to another on one drive, as the sata interface has to read and write simultaneously. With two drives, one reads and one writes.

 

Even if you use 2 HDDs, it is better to use only a small partition for Windows. Using the whole disk is just wasteful, about 100 GB would do, in case of problems, you would remove the partition. That said, 60-120 GB SSD would suffice OS only.

 

Absolutely..... I'm convinced . *Biggrin

OP , I dual boot 14393 on metal with 14905 being the daily driver on an SSD you don't want to go back to metal once you use an SSD windows boot drive it's torture on a metal HDD once you been on a SSD even on my Core i7 gaming desktop in the PC game /music room or the metal drive in here *really.

I use the metal 14393 Windows healthy partition , library's and storage and run some exe off there that will work like that .

I have most exe I use and *some files ,folders and data I use frequently on the 14905 SSD ,my daily use OS, but the majority of storage I manage to keep the SSD from getting full is on the 14393 metal drive or another Seagate USB 3.0 expansion desk .

I haven't made a separate NTFS storage volume on the 14393 spindle drive but it may happen .

This is a brand new PC in my specs and I have plenty of storage on the Seagate USB 3.0 external HDD I migrated out of the old PC and had previously on the external drive also .

This Haswell Core i5 PC in my specs is wicked fast on an SSD Windows drive for everyday computing ,the web and power multi tasking .

Here is my storage arrangement :

 

up vote here *chuckle

 

I don't think you will find a real world situation where one drive would perform better than two. Two drives gives you two sets of read write heads and two drive caches which function independently. Even in an all SSD system the advantage would go to the two drives, although the advantages would be minimal. The only significant advantage a single drive has is lower cost.

The primary factor in the performance of conventional drives is head seek time. It will usually take more time to seek to the data required than to read the data. To maximize performance you want to minimize both the number and extent of head seeks.

From a strictly performance standpoint it would be best to install the OS on a small partition of a much larger fast drive. This confines the heads to a small portion of the drive and minimizes the extent of seeks. But this is very wasteful because you can't use the rest of the drive for anything, except for something rarely used, without compromising performance. An even bigger problem is that the large drive will cost you as much as a smaller SSD which would provide much better performance. I never claimed this was a practical thing to do.

Typical usage patterns for the OS drive is frequent but small reads scattered over many portions of the drive. That means frequent and lengthy head seeks. With their almost non existent seek times an SSD would excel at this but it is a real problem for a conventional drive.

Typical data drive usage tends toward much larger reads which minimize head movement. Conventional drives handle this quite well.

Using two drives the usual configuration is to have a small SSD for the OS and a larger conventional drive for your data. Using an SSD for data would provide better performance but a drive of adequate size is still quite expensive. But that is changing. This configuration provides most of the advantages of an SSD but at a significantly lower cost. If you are going to have two drives this is the most logical configuration.

 

Is your D: for installed programs? Is it beteer have seprate for OS and Programs as well?

 

There is little to be gained in having a separate drive for programs. If you have to reinstall the OS you would have to reinstall all applications anyway so there is no advantage there. It is often done to conserve space on an SSD that is too small to accommodate all applications. It tends to make things more complicated as most applications will install to c: by default and some won't give you a choice. And if D: is a separate partition you would lose performance.

Applications are installed on a separate drive for a variety of reasons but it isn't really an optimum configuration.

 

That's true if all drives are same model. Otherwise two slow drives vs one fast drive would not always result in 2 drive win situation.
There is also thing with HDD platter size/data density on platter. More densely packed data on platter would require less movement from reading heads thus it would result to better sequential read performance at lower rpm.

 

Thank you all you guys help *Biggrin
It makes me clear now!