Forum Rumors: Pagefile on an SSD… Is it a bad thing?

Posted: January 13, 2011 in Forum Rumors
Tags: , , , , , ,

Today I ran into a post on a forum I frequent regarding whether or not it was damaging to put your Windows pagefile (or Linux swapfile) on an SSD. The argument goes something along the lines of “Lots of quick reads and writes could damage the SSD over time and decrease performance”.

So I began researching. What I found is that many, many forum users believe this to be true. However, I could only find one example of actual research done on the subject.

The Microsoft Developer Network Blog has done research into the topic by looking at pagefile activity.

Should the pagefile be placed on SSDs?
Yes. Most pagefile operations are small random reads or larger sequential writes, both of which are types of operations that SSDs handle well. In looking at telemetry data from thousands of traces and focusing on pagefile reads and writes, we find that

  • Pagefile.sys reads outnumber pagefile.sys writes by about 40 to 1,
  • Pagefile.sys read sizes are typically quite small, with 67% less than or equal to 4 KB, and 88% less than 16 KB.
  • Pagefile.sys writes are relatively large, with 62% greater than or equal to 128 KB and 45% being exactly 1 MB in size.
  • In fact, given typical pagefile reference patterns and the favorable performance characteristics SSDs have on those patterns, there are few files better than the pagefile to place on an SSD.

    After half a dozen pages of Google search results I was not able to find another reputable website that covers the topic. About the closest I could find was Tom’s Hardware stating:

    Windows is smart enough to store data from minimized applications or idle applications in the swap file so that active processes can access more RAM. The disadvantage of this solution is obvious. The performance of a mechanical hard drive is only a fraction of the throughput realized with memory modules. Solid state drives (SSDs) are an improvement, but the general problem remains. Frequent swap operations are poison to an SSD, possibly decreasing its speed.

    Possibly decreasing its speed. Tom’s Hardware is careful not to say one way or the other as they haven’t seen any research either.

    Essentially what it comes down to is who to believe. Would you believe MSDN with their research into pagefile activity and potentially risk damaging your SSD? Or would you believe the hundreds of forum posters who claim that it will indeed damage your SSD but have no research to prove it? I think this one comes down to a situation of “Can you afford to replace your SSD?”. If the answer if “yes” then relish in the performance benefits. If your answer is “no” then I would play it safe and put your pagefile on a separate hard drive.

    1. LUKEOUT says:

      My question is what is the lifespan of an SSD compared to a normal hard drive?

    2. astrogiblet says:

      It seems to vary greatly. Some manufacturers are claiming a very unrealistic figure of “2,000,000 hours”. Intel says you can write 100GB/day to their MLC SSD’s for 5 years straight and they won’t fail (they only have a 3 year warranty though).

      Unfortunately SSD’s are so new and their technology is constantly changing/updating that its impossible to say a set amount of time your SSD will last. I think most people seem to agree that SSD’s should last longer than HDD’s, though.

    3. Evan Allan says:

      Electron mitigation effects all circuit components. I can say that overtime heavy or light io will result in decreased performance of any ssd as ecc on the drive will become more prominent due to the mitigative effect of electron flow.

    Leave a Reply

    Fill in your details below or click an icon to log in: Logo

    You are commenting using your account. Log Out / Change )

    Twitter picture

    You are commenting using your Twitter account. Log Out / Change )

    Facebook photo

    You are commenting using your Facebook account. Log Out / Change )

    Google+ photo

    You are commenting using your Google+ account. Log Out / Change )

    Connecting to %s