Benchmarking Pagefile on SSD drive versus an RST accelerated 6GB/s SATA3 drive

[birket]

We are using Z68XP-UD3-iSSD boards which have the on board 20GB SSD drive.
We have 16GB RAM.
Our main drive is a 120GB Corsair Force GT drive.
Our secondary, accelerated drive is a WD Green SATA3 6gb/s drive.

We are wondering if the pagefile (24GB) should sit on the Corsair drive or on the accelerated SATA3 drive. Ideally, we would like it to be on the SATA3 drive as it frees up more space on the C drive, but is this going to give a performance hit?

We use intensive 3d modelling, rendering and Photoshop type software.

How would I benchmark each setup?

[Gloup_Gloup]

Hi birket,

Have you setting each 3D and Photo Software.
Like a designed a device (Hard disk) for temporary files used by ex. filters, texturing, etc

Check a documentation about this, if need. By internet, manual paper or .pdf files.

I think if you move a pagefile on other hard drive than OS System, this is the best.

Maybe de pagefile is too big.
If is too big, windows use it, but the best is when Windows use a hard disk, not a pagefile.

So made a change, modifications and start again test with HDtune Pro or your software test.

Bench Soft:
http://www.hardwareheaven.com/photoshop.php

PhysX FluidMark 1.2.0 Available with Multi-Core CPU Support
http://www.geeks3d.com/20100521/gpu-tool-physx-fluidmark-1-2-0-available-with-multi-core-cpu-support/

http://dee.cz/lightsmark/index.html








Gloup_Gloup

[Dark Mantis]

Hi

To be honest your best bet would be to run without a pagefile/swapfile. You have plenty of memory so it shouldn't be a problem and will get you out of the sticky position you are now in.

You shouldn't run a pagefile on a SSD as it is not healthy for them and will destroy them much faster than normal. Your other drive is a "Green" drive though and they are far from ideal for pagefile usage because of their always trying to power down.

I have been running without a pagefile on all my machines with the smallest amount of RAM being 6GB.

[pat5zero]

I have a Corsair force 3 120GB SSD with a  WD Black Sata 3 1TB on the same Marvell 9128 controller. with 1gb of pagefile and all temps and local and roaming directories on the WD Black 1TB.
I have only 6GB of RAM, You should let everything load into your 16GB ram and have nil or a small pagefile.
If you use internet explorer you should move it's temp directories to the WD Green HDD also as it writes to much to the SSD causing more wear.
In fact it is possible with some editing of "paths" to move all your programs and systems temp directories to another drive.. IE the WD Green HDD you have.
BTW The read/writes are quite remarkable this way with bursts of up to 550MB/s, but averages of 192MB/s to 250MB/s between the two drives.

[Dark Mantis]

I have a Corsair force 3 120GB SSD with a  WD Black Sata 3 1TB on the same Marvell 9128 controller. with 1gb of pagefile and all temps and local and roaming directories on the WD Black 1TB.

BTW The read/writes are quite remarkable this way with bursts of up to 550MB/s, but averages of 192MB/s to 250MB/s between the two drives.

You don't mention what motherboard you have but if it is one of the older ones like a X58 chipset board then that could be one of the reasons why you are having weird data performance fluctuations because of the controller they are connected to. Please post back with details of your motherboard.   

[Gloup_Gloup]

Hi Dark Mantis,

Pat5zero have a GA-X58A UDR3 ver 2.0.

Source : Running 2 kits of 3x2GB memory?
http://forum.giga-byte.co.uk/index.php/topic,7313.msg57702.html#msg57702

Gloup_Gloup

[Dark Mantis]

Hi Gloup_Gloup and thanks for the info. That is probably why he is having not normal data transmission rates then because of his use of the Marvell 9128 controller ports. He would certainly be better off using the SATA2_0 port for the SSD and the SATA2_2 port for the other drive.