0

I've ready that the pagefile.sys writing frequently to an SSD drive reduces its lifespan and since and SSD drive doesn't much benefit from putting the data in one location (since it's truly random access) there is little benefit.

Is that generally true?

  • 1
    “Is that generally true?” - Which part of your question? The fact a page file isn’t useful if your using a SSD (which is false) or putting all data in one location brings no benefit (which is false) – Ramhound Feb 23 '18 at 0:04
  • 1
    This question and your other one today about SSDs would have made (some) sense a decade ago, not today. – user772515 Feb 23 '18 at 1:04
  • I don't think so. Check this out: seagate.com/tech-insights/… – iTechieGamer Feb 27 '18 at 15:30
3

No, pretty much all of that is false.

First, if you are writing to the pagefile frequently, then it's really going to affect system performance. In that case, having the pagefile on the fastest device you possibly can will provide a huge benefit.

Second, pagefile read access is largely random. SSDs truly shine for random access because they don't have heads to move or platters to wait to reach the right spot.

Third, most of the time under most conditions, there will be almost no use of the pagefile at all. So it's very unlikely to have any significant affect on the life of a modern SSD drive.

The one good reason not to put a pagefile on an SSD drive is if SSD space is precious (because you have a relatively small SSD) and you don't expect to use the pagefile very much (because you have lots of RAM). Otherwise, under almost all realistic conditions on modern hardware, it will either make almost no difference or be better to put the pagefile on the SSD.

Some of this was true some time ago with much older SSDs that were very small and didn't support wear leveling. But it's persisted as a myth, similar to battery "memory" myths that applied to ancient NiCd batteries that we still hear even though laptops use Lithium batteries today.

| improve this answer | |
0

The operational and performance characteristics of an SSD are an almost perfect match for the typical usage patterns of the pagefile. On the other hand, the match between a conventional drive and the pagefile could hardly be worse. In years past conventional drives were used for this because there was nothing else available. The development of the SSD changed all that.

Typical usage patterns of the pagefile consist of small reads mixed with other disk access. This means the disk heads will be moving around a lot and further time will be spent waiting for the disk platter to rotate to the required data. A perfectly unfragmented drive won't help much. This means the drive will be spending far more time in overhead than actually reading and writing data. A conventional drive does quite well with long sequential reads (but still not as good as an SSD) but that hardly ever happens with the pagefile.

OS designers are well aware of the performance of hard drives (both conventional and SSD) and try very hard to minimize pagefile reads and writes. Microsoft estimates that under typical conditions pagefile reads outnumber writes by a factor of about 40 to 1.

As an SSD has no moving parts it will do well with small random reads and writes, which is typical for pagefile usage. There are few files that will be benefit more from being on an SSD than the pagefile.

It is true that SSDs do have limited writes. But with modern drives that number is so high that you really have to work hard to reach it. It isn't an issue under real world conditions. SSD manufacturers have published some very impressive numbers that show this. And this appears to be an unusual situation where manufacturers claims are not exaggerated but actually very conservative. SSD endurance tests have shown that they survive well beyond what published numbers would indicate.

The bottom line is that modern SSDs are not delicate little things that must be treated with great care but are actually very durable. They can certainly endure normal computer usage without difficulty. Most drives fail for reasons not related to excessive writes. Or more likely they are replaced because they have proved to be too small for the intended purpose.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.