Short freezes with "hard faults/sec" spikes

[Zhnigo]

My machine is randomly getting short freezes which can last from about 10 seconds up to a minute. This can be temporarily fixed by rebooting multiple times for a long period of running without hitches, but with every new boot the probability of the problem reappearing persists.
This current build has been running for more than a year with stability, and I haven't made any hardware or software changes prior to the problem appearing. I checked the resource monitor and noticed that the freezes corresponded with spikes on the "hard faults/sec" graph. Further googling suggested that it's related to the PC running out of RAM and using page files - even though the graph stated that only 8 out 16 gb ram was in use at the time. Does that mean that the problem is with my RAM or with my SSD where the page file is? Or is it something else?
One other thing I noticed: together with this problem there appeared a curious quirk. Before that, on boot all of my usb devices have been activated simultaneously. Now with every boot I hear several usb connection sounds in a row which correspond to my devices activating one after another, with short delay. Might not mean much, but I guess there's no harm in mentioning.

Specs:
Win7x64
4x4gb ddr3 ram, 2 sticks are about a year old and 2 others are about 3 years old
the SSD in question is a Crucial 120 gb, running for about two and a half years. CrystalDiskInfo shows no problems with it.

Thank you very much in advance!

 

[gamerk316]

"hard faults/sec" measures how many times the CPU requests data that is NOT currently in main memory, requiring data to be read off the HDD, which is a relatively time consuming process. This really shouldn't be happening much with 16GB of RAM though, certainly not enough to be noticeable.

How high are the spikes reaching? Are we talking 5 hard faults/sec, or are we talking 30-40?

Throwing a dart at a dartboard, it sounds like there's a background process that's requesting data periodically. Might be worth keeping an eye on the Disk tab of the resource monitor and see if any specific process suddenly jumps up in disk usage.

 

[Zhnigo]

How high are the spikes reaching? Are we talking 5 hard faults/sec, or are we talking 30-40?

Thanks for your reply!
During normal work they're zero or rarely single digits, but during a freeze it jumps up to the top of the graph, which caps out at 100.

 

[gamerk316]

Yeah, definitely sounds like something is requesting a ton of data off the HDD. Only thing to do is try and catch which process it is.