Memory optimizing - bad for SSD ?

Discussion in 'other software & services' started by pb1, Sep 29, 2015.

  1. pb1

    pb1 Registered Member

    Joined:
    Apr 4, 2014
    Posts:
    343
    Location:
    sweden
    Hi
    Does it ad a lot of writings to the disc if it works like this - http://www.rizonesoft.com/download/memory-booster/ - optimizer does, is this not safer then defraging it ?

    And the 2:nd question; making this API call makes the processes drop significantly for me and therefore saves writings to the disc, so the net result - despite the use of above - would be that it saves writings ?!

    Anyone want to elaborate on the above and help me out here because i am not that good at this, yet.
    Thank you.
     
  2. J_L

    J_L Registered Member

    Joined:
    Nov 6, 2009
    Posts:
    8,516
    The only memory optimizer I use is CleanMem. No writing to the pagefile at all.
     
  3. Rolo42

    Rolo42 Registered Member

    Joined:
    Jan 22, 2012
    Posts:
    569
    Location:
    USA
    I'm gonna go ahead and say, "snake oil" to both (they claim to do the same thing the same way). And, yes, the CleanMem guy mentions this but doesn't provide any reason why it isn't.

    The only time I know of a RAM "cleaner" being useful was early, early, early Android (like, pre-Gingerbread/2.3, which didn't manage RAM well at all) and, unfortunately, they are still sold/used while actually slowing down Android, which now optimises memory very well on its own.

    But to answer your question, this has nothing to do with your drive.
     
  4. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    2,271
    Location:
    Nebraska, USA
    :argh: Why is it so hard to understand that the CPU does not push stuff out to the page file unless it does not need it right away. High priority data will remain in memory. Only low priority data gets pushed to the PF. If you keep more low priority data in system RAM, you force the system to actually save off data (that is, close files) which then take even longer to retrieve from the HD than if the data is in the PF. PFs are good things, not evil. Especially if you don't have a lot of RAM installed.

    I am glad the author of CleanMem makes it clear this program does not replace buying more RAM. I note too he cites a Tom's Hardware article but the article really has nothing to do with his program, neither does standby memory. Plus it is from 2008. Newer versions of Windows are much better at managing memory.

    Yeah, I agree. Snake oil.
     
  5. RJK3

    RJK3 Registered Member

    Joined:
    Apr 4, 2011
    Posts:
    855
    Hi pb1, just trying to understand why you think a memory optimiser worth using would increase writes to the SSD? The majority of 98/XP era memory boosters used to cause instability by forcing useful information out of RAM and into the pagefile - thus reducing overall performance. I tried more than a few and rejected them as pointless.

    From reading the brief synopsis, this one seems more benign like CleanMem. CleanMem does have the potential to reduce pagefile usage in low RAM scenarios, or when using programs with memory leaks. It won't cause you any harm, but on a well managed machine doesn't appear to make much of a difference either. Personally I use it on Windows XP machines with 512mb memory or less - and only after rationalising autoruns, services, etc first.

    I found a post from the author of CleanMem which explains his rationale for creating the program, and the scenarios for which it can be useful:
    https://www.wilderssecurity.com/threads/cleanmem-experimentation.371695/page-2#post-2442098
     
  6. roger_m

    roger_m Registered Member

    Joined:
    Jan 25, 2009
    Posts:
    5,242
    The post @RJK3 linked to, explains exactly why CleanMem (Memory Booster works the same way) is not snake oil, and may be useful for some people.
     
  7. Rolo42

    Rolo42 Registered Member

    Joined:
    Jan 22, 2012
    Posts:
    569
    Location:
    USA
    Paging was created for low RAM scenarios!
     
  8. RJK3

    RJK3 Registered Member

    Joined:
    Apr 4, 2011
    Posts:
    855
    Of course, but better to minimise the need for it - especially on old machines with slow HDDs where disk thrashing is more noticeable.
     
  9. Victek

    Victek Registered Member

    Joined:
    Nov 30, 2007
    Posts:
    5,121
    Location:
    USA
    Yes, younger users might not know that ram used to be ridiculously expensive and virtual memory, aka the page file, was created to allow more code to load then the actual ram could handle. Now ram is affordable and no longer a bottleneck.

    Regarding reducing SSD writes, how about creating a ram disk and pointing browser cache and temp files at it?
     
  10. RJK3

    RJK3 Registered Member

    Joined:
    Apr 4, 2011
    Posts:
    855
    Not even necessary - most consumers will never exhaust the write endurance of their SSDs within the useful lifetime of the device. The famous SSD endurance tests showed that in spades.
     
  11. pb1

    pb1 Registered Member

    Joined:
    Apr 4, 2014
    Posts:
    343
    Location:
    sweden

    Hi
    Not so much optimizing as reducing use of it. My IE11 browser uses much RAM initially when started but stays with high consumption even though it is actually a lot lower, so by cleaning up after 5 min and also after c:a20 min it consumes less for the rest of the time. Without any problem. Windows is really bad at cleaning up redundant RAM usage.

    Without Cleanmem - for instance - the RAM consumption is steadily at 77% with it - 56% with above usage profile. I only use it for IE.
     
    Last edited: Sep 30, 2015
  12. pb1

    pb1 Registered Member

    Joined:
    Apr 4, 2014
    Posts:
    343
    Location:
    sweden
    I have a RAM disc. For some reason it works fine for temp.files but not for browser cache - then all kinds of problems happens.
     
  13. pb1

    pb1 Registered Member

    Joined:
    Apr 4, 2014
    Posts:
    343
    Location:
    sweden

    Where can i find that endurance test ?
     
  14. Victek

    Victek Registered Member

    Joined:
    Nov 30, 2007
    Posts:
    5,121
    Location:
    USA
  15. RJK3

    RJK3 Registered Member

    Joined:
    Apr 4, 2011
    Posts:
    855
    Yep TechReport.com did a long running series on it, and there was a website dedicated to an SSD endurance test which now appears to be down.
     
  16. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    2,271
    Location:
    Nebraska, USA
    SSD endurance is really a non-issue with today's generation SSDs for the vast majority of home and small office/business users. Note that more and more corporate servers and data centers are even going with SSDs and it is not just for speeds, but energy use and heat generation too. If today's SSDs suffered from poor endurance, it would not be cost effective for data centers to use SSDs.

    Note that many SSDs today are coming with 10 year warranties! They just could not do that if endurance was an issue. And the various SSD makers are in deep competition for your money. They really don't make products that then require 3rd party programs to make them run better. That makes no sense.

    If their SSDs need optimization, they build the routines in the drive controllers and have worked with all the major operating system developers (MS, Linux, Apple, etc.) to ensure the necessary optimization routines are supported by the latest versions of the OS too.
     
  17. Rolo42

    Rolo42 Registered Member

    Joined:
    Jan 22, 2012
    Posts:
    569
    Location:
    USA
    Browsers already have a memory cache and moving the disk cache to memory defeats the point of it: persistence.

    Which really means you're wasting RAM by not using it. Utilising RAM and running out of RAM are completely different things.

    It's cost effective because of what you stated: heat and power, which make up about half of a data centre's operating cost. I do, however, agree that SSD longevity isn't a concern due to improvements and misconceptions.
    My next build won't even have a platter drive installed.
     
  18. J_L

    J_L Registered Member

    Joined:
    Nov 6, 2009
    Posts:
    8,516
    Unless you can prove that IE 11 is taking a performance hit not "utilising" that RAM, it only proves that it was wasting that RAM in the first place.

    I remember you stating that some "moderate" scenarios take up 8 GB of RAM, wouldn't that be the perfect "snake oil" testing ground?
     
  19. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    2,271
    Location:
    Nebraska, USA
    RAM disks were great when hard disk space was so expensive and RAM was relatively cheap. All a RAM disk does is take RAM away from the OS and CPU.

    My point was if "today's" SSD endurance was a big issue (as it was just a handful of years ago), the cost of replacing SSDs all the time would greatly surpass the cost of heat and power - especially since SSDs draw so little power and generate much less heat than convention hard drives.

    So it is win win win all around with SSDs. Plus, SSDs are physically much smaller than HDs so even floor space is saved. This will be even more significant with the new 4Tb SSDs coming on the market with 8Tb SSDs coming soon. I think the largest HD for home use is only 6Tb (I know there are 10Tb and larger for commercial use, however).
     
  20. Rolo42

    Rolo42 Registered Member

    Joined:
    Jan 22, 2012
    Posts:
    569
    Location:
    USA
    I don't recall that ever being the case and I go back to my IBM 4.77MHz PC with a 10 megabyte HDD.

    RAM disks are for speed.

    The burden of proof rests on the one making the claim (not me in this case) and the only anecdotal arguments were old systems running software under minimum requirements.
    It isn't just IE11; it's the box entirely as Windows manages hardware resources intelligently to minimise latency and even anticipate what the user will do next. So-called RAM cleaners/optimisers completely undermine that.
    "Free" RAM is RAM that isn't doing anything useful.

    No. I'm actually using that RAM. Somewhat related: I switched to Opera and it is a lot lighter on RAM and more responsive than Chrome. My point being that having a "cleaner" and inefficient software continuously duking it out isn't the best approach...especially if it pulls that stunt during real-time activities (streaming, gaming).

    I got your point. My point was that home users don't have massive power and cooling costs that would be offset by this and does nothing for the FUD surrounding SSDs. I remember someone getting on my case about using a server scenario in an end-user situation. ;)

    250GB Samsung SSD, Amazon, ~$90...must-have if you have the coin.
     
  21. Bill_Bright

    Bill_Bright Registered Member

    Joined:
    Jun 29, 2007
    Posts:
    2,271
    Location:
    Nebraska, USA
    Then I won't hold your youth and lack of experience against you! ;) I go back to before the IBM PC!

    Right. https://en.wikipedia.org/wiki/RAM_drive
    Those are the facts. But some "treehuggers" don't look at it that way. So instead, they buy a Prius disregarding the fact the manufacturing of the car pollutes the environment way more than the pollution the gas model would create during its life.
     
  22. roger_m

    roger_m Registered Member

    Joined:
    Jan 25, 2009
    Posts:
    5,242
    My expereince with Windows is that you don't want Windows to ever be using more than 74% of RAM if you can help it. Of course in such cases, adding more RAM is the best option. But, in the some cases that is just not possible.
     
  23. Rolo42

    Rolo42 Registered Member

    Joined:
    Jan 22, 2012
    Posts:
    569
    Location:
    USA
    Ah! That explains it. Guys like you had kids like me set the clock on your VCR for you...newfangled technology and all! :argh:
     
  24. RJK3

    RJK3 Registered Member

    Joined:
    Apr 4, 2011
    Posts:
    855
    Not really sure why CleanMem becomes an emotional topic for some. If something noticeably reduces disk thrashing and even non-techie people comment, then that's a win to me. If I think back, some of the above probably occurred in the time when Firefox was having memory leak issues. Either way, it was plain that CleanMem was doing something beneficial based on commit charges during idle and peak usage.

    I admit, it did take me awhile to come around to the idea of even giving it a chance - but even with the strongly emotional arguments made against it over the years, I've never seen anyone actually demonstrate any harm in using it.

    Didn't see this first time around, the alert took me to your second post. As Rolo stated, browsers like Firefox started using the memory for cache many years ago, so there's little benefit to using a RAM disk.


    Here I was impressed with the 20mb HDD added to the Tandy 1000 I still have sitting in the garage somewhere.
     
  25. TairikuOkami

    TairikuOkami Registered Member

    Joined:
    Oct 10, 2005
    Posts:
    2,509
    Location:
    Slovakia
    Windows has bad memory management and MS refuses to accept it still repeating, that it is supefetch and people just do not understand, how it works, so if people, who have low memory issues, are being told to buy more RAM instead of solving the flaws. I had the same problem and cleanmem fixed that and it is still needed in Windows 10, because of Microsoft's ignorance - http://www.tenforums.com/performance-maintenance/21560-high-ram-usage-2.html#post382833 Basically Windows preloads memory for needed software, which is good in theory and when another app needs RAM, it should release the unused RAM, but the problem is, that when app tries to run, it detects, that there is not enough RAM and it will crash, even though there is more than enough unused RAM. Cleanmem just tells Windows to release unused RAM, which solves the problem. Insanely simple. :isay:

    And his concern was not just high memory usage, but that it caused lagging during gameplay and such, which is a side effect of memory leaks.
     
    Last edited: Oct 1, 2015
Loading...