Skip to main content

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

5
  • A RAM disk is a terrible idea in almost all situations, and in EVERY "limited RAM" situation. And add-on "memory managers" almost always do more harm than good. They operate on the principle that free RAM is good. In fact, free RAM is wasted RAM. Commented Mar 18, 2015 at 22:56
  • @JamieHanrahan:I wrote RamDisk in the sense of Memory-Swap. Additionally it can be compressed:vflare.org/2009/03/ram-is-not-enough-memory-compression.html. Commented Mar 19, 2015 at 9:05
  • I'm not sure what "in the sense of memory-swap" means, but if you mean putting a page file on a RAM disk? That isn't even wrong. It will increase the page fault rates. Not all the additional faults will be resolved to the RAM disk (most, statistically, go to mapped files). Of those that are, yes, the PF on RAM disk will be faster than a hard drive - but NOT committing that RAM to a RAM disk, hence not incurring those additional faults at all, would be even faster. Commented Mar 19, 2015 at 9:16
  • @JamieHanrahan:I'm not sure what is your problem. You can read the article about compcache. It uses a compressed ram-disk for the swap. In some cases it helps. Commented Mar 19, 2015 at 9:20
  • I did read the article. It is easy to devise specific workloads in which such things appear to help. Even a simple "memory optimizer" (which simply evicts as much as it can from RAM) will appear to help with the startup time of the next program you start. But to do so it kills the performance of everything else you had running, including much of the OS. Showing across-the-board performance gains with no downside is MUCH more difficult. Commented Mar 19, 2015 at 20:39