Timeline for Should I disable swap file if I have lots of RAM or should I move it to a virtual RAM drive?
Current License: CC BY-SA 3.0
52 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Feb 23, 2022 at 16:09 | answer | added | Patrick Burwell | timeline score: 0 | |
Jun 29, 2021 at 4:36 | review | Close votes | |||
Jul 4, 2021 at 3:01 | |||||
S Dec 7, 2020 at 7:07 | history | bounty ended | CommunityBot | ||
S Dec 7, 2020 at 7:07 | history | notice removed | CommunityBot | ||
Nov 29, 2020 at 10:08 | answer | added | harrymc | timeline score: 3 | |
S Nov 29, 2020 at 5:31 | history | bounty started | user1306322 | ||
S Nov 29, 2020 at 5:31 | history | notice added | user1306322 | Current answers are outdated | |
Jun 19, 2017 at 2:17 | answer | added | Robert Fischer | timeline score: 0 | |
May 4, 2016 at 16:14 | comment | added | Ƭᴇcʜιᴇ007 | possible duplicate: Windows Swap (Page File): Enable or Disable? | |
Mar 4, 2016 at 11:32 | answer | added | user127379 | timeline score: 5 | |
Nov 30, 2015 at 0:08 | comment | added | Sam Watkins | Even for jobs that seem to require more memory than is available, swapping whole pages in and out is rarely the most efficient way to utilize the memory. Better if the programmer designs the job to use RAM and files efficiently as needed in the first place. | |
Nov 30, 2015 at 0:01 | comment | added | Sam Watkins | Paging virtual memory is like overconfidence, "Yeah, I can handle this job by myself, no worries!" whereas in fact the computer would take a month of swapping to finish the job and doesn't even know it. If a job works well with serial access it should be using files in the first place. If a job requires rapid random access on a large chunk of RAM (e.g. for an in-memory hash, or for executable code), in no way is it acceptable to be swapping that RAM in and out from disk during the operation. The computer pretends it has infinite RAM, but this is a lie and not a helpful one. | |
Nov 30, 2015 at 0:00 | comment | added | Sam Watkins | IDK, but virtual memory / swapping / delayed loading certainly massively decreases responsiveness compared to older systems (such as Acorn Archimedes, or even old 8-bit Micros) which did not do this. It's a sorry state of affairs when a modern wordprocessor is less responsive for many operations (when first loading, or swapped out) than an 8-bit 2Mhz machine. Personally I would rather have "out of memory" error any day of the week than swapping to disk and the resultant massively degraded performance. This would encourage programmers not to waste RAM. | |
S Feb 12, 2015 at 6:15 | history | bounty ended | user1306322 | ||
S Feb 12, 2015 at 6:15 | history | notice removed | user1306322 | ||
Feb 7, 2015 at 22:48 | answer | added | Michael | timeline score: 2 | |
Feb 7, 2015 at 20:55 | answer | added | harrymc | timeline score: 7 | |
Feb 7, 2015 at 19:18 | history | protected | Keltari | ||
S Feb 7, 2015 at 18:54 | history | bounty started | user1306322 | ||
S Feb 7, 2015 at 18:54 | history | notice added | user1306322 | Reward existing answer | |
Feb 7, 2015 at 15:33 | comment | added | usr | @DanW I have never heard about that. That would rquire the app to explicitly check whether the paging file is on and stop working if it is not. Few reasons an app would care about the paging file. Assuming, though, that such apps exist: set the file size to 1MB. | |
Feb 7, 2015 at 15:30 | comment | added | Twinbee | @usr: Again, whilst you're right in theory (and I agree with the spirit of that), there are a few reasons to not disable it completely. The most notable being that some applications (particularly older ones) require the pagefile to run. | |
Feb 7, 2015 at 15:27 | comment | added | usr | @DanW those users should have disabled the file and would have gotten even better results. Putting the paging file on a RAM disk is always inferior to having no paging file at all. | |
Feb 7, 2015 at 15:20 | comment | added | Twinbee | @usr: I was just responding to the point of "never accomplishes anything", as all those users experienced a performance increase when putting the pagefile into the Ramdisk. It shouldn't happen, but it does. | |
Feb 7, 2015 at 15:07 | answer | added | Twinbee | timeline score: 5 | |
Feb 7, 2015 at 9:08 | comment | added | user1306322 | I guess it only takes to try it out to find out if it works for you or not. | |
Feb 7, 2015 at 9:02 | comment | added | usr | @begtodiffer it is true that Windows misjudges paging decisions all the time. But with out page file at all, instead of one on a RAM disk, there is nothing to misjudge. | |
Feb 7, 2015 at 1:05 | comment | added | Twinbee | @usr: Whilst in theory you're correct, in practice, you're not taking into account design misjudgments from the Windows kernel programmers. See all the users who would beg to differ. | |
Oct 22, 2014 at 8:35 | vote | accept | user1306322 | ||
Oct 10, 2014 at 1:01 | audit | First posts | |||
Oct 10, 2014 at 1:02 | |||||
Sep 26, 2014 at 20:40 | audit | First posts | |||
Sep 26, 2014 at 22:20 | |||||
Sep 20, 2014 at 20:01 | audit | First posts | |||
Sep 20, 2014 at 20:01 | |||||
Sep 18, 2014 at 5:19 | comment | added | Sun | Why aren't you installing the game on the ramdrive and running it from there? | |
Sep 16, 2014 at 16:25 | audit | First posts | |||
Sep 16, 2014 at 16:26 | |||||
Sep 15, 2014 at 10:47 | answer | added | nsn | timeline score: 3 | |
Sep 14, 2014 at 11:34 | comment | added | Victor Zakharov | I am running with swap disabled on my home PC, never had an issue. 16GB of RAM. This can accommodate host + VMWare guest running development environment (VS2010 + SQL server 2014). OS is Windows 7. | |
Sep 13, 2014 at 17:52 | audit | Suggested edits | |||
Sep 13, 2014 at 17:53 | |||||
Sep 13, 2014 at 9:53 | comment | added | Lesmana | memory management and pagefile/swapfile management have changed radically over kernel versions and will continue to change radically. please state in your answers clearly for what kernel version your answer is valid. | |
Sep 13, 2014 at 2:24 | comment | added | Ry- | I would (and do) only turn it off for an SSD, since you only get so many writes (even if it’s a lot now), with 6GB. It works well. | |
Sep 12, 2014 at 23:35 | answer | added | peter | timeline score: 2 | |
Sep 12, 2014 at 22:07 | comment | added | Rob Moir | @user367257 creating a ram disk to store your page file on is like lending your friend £10 so that he has enough money to allow you to borrow £10 from him. It might be technically possible but all you've accomplished is to needlessly complicate a journey to nowhere. | |
Sep 12, 2014 at 6:18 | comment | added | user541686 | The answer is yes but there are plenty of nonbelievers. | |
Sep 11, 2014 at 21:10 | comment | added | hookenz | In makes some sense to do this on Linux in some cases where the ram disk hosting the swap file is actually compressed. See: en.wikipedia.org/wiki/Zram. However I don't think Windows has such a feature available to it. | |
Sep 11, 2014 at 15:43 | answer | added | David Schwartz | timeline score: 148 | |
Sep 11, 2014 at 15:42 | answer | added | demonkoryu | timeline score: 9 | |
Sep 11, 2014 at 15:28 | answer | added | JamesRyan | timeline score: 5 | |
Sep 11, 2014 at 14:47 | comment | added | usr | Having a paging file on a RAM disk never accomplishes anything. You take away a certain amount of available memory and add a certain amount of virtual memory. Null-sum. Just have no paging file. | |
Sep 11, 2014 at 8:55 | history | tweeted | twitter.com/#!/super_user/status/509988636796780544 | ||
Sep 11, 2014 at 8:55 | answer | added | Art Gertner | timeline score: 41 | |
Sep 11, 2014 at 8:52 | answer | added | Peter | timeline score: 14 | |
Sep 11, 2014 at 8:38 | review | First posts | |||
Sep 11, 2014 at 8:39 | |||||
Sep 11, 2014 at 8:34 | history | asked | user1306322 | CC BY-SA 3.0 |