Jump to content

Saphir

Members
  • Posts

    22
  • Joined

  • Last visited

  • Days Won

    2

Posts posted by Saphir

  1. Hallo,

    is it normal that there are Empty folders everywhere in my Pool? The folders are correct but It looks like the balancer were moving the files to another disc and leaving empty folders behind.

  2. If you have set up the file placement to put a specific folder on a specific drive it should work. I wouldn't use the Ordered File Placement Plugin for this because it seems that this Plugin is for a general order priority for drives (will fill on drive after another with all the files are coming in). The file placement I marked should do what you want.

     

    20210315_001.JPG

  3. Hi Glauco

    I just changed my Server from WHS 2011 to Win 10 with deleting from my OS drive the old partitions and installing Win10 frech from an USB Stick. The drives with my data which were in a Pool I disconnected in this time. Afterwards I installed DrivePool new on the new OS, switched the PC of and connected the data drives. after starting Win 10 I looked in DrivePool and the Pool was there. It worked with my Backup Pool too. Only the drive letter was changed.

    You only have to install DrivePoll and connect all your Pool drives. DrivePool scannes the drives and will find the information it needs.

  4. On the main setting side of the balancer you can set if the balancer is starting at a specific time or if a certain amount of data has to be moved. I have set my SSD to 600 GB as example. If there is not the need to move 600 GB the data will stay on the SSD.

  5. Thanks for your answers. I saw that my sync software watches after files and folders. Checking a folder I saw that it has ecerywhere the same timestamp and on one place not but there was a timestamp from today and the problem was there before too.

    My setup is PoolDrive Z:Data and Y:Backup. Allwaysync makes the sync oneway from Z to Y. The folder structure is synced too. Y: lookes the same as Z: + a hidden folder for the file history in every folder.

    Empty folders I found but have to check if this happens continuously. For the folder I checked because the files shouldn't go to the history in my backup there where no empty folders.

    I think I have to do a lot of testing and see what happens.

    FreeFileSync I didn't use for a long time maybe I will have a try on this now since it looks much better now. Bu I think booth programs should do the same. I'm thinking too to use a backup software like snapraid, do you have any experience with this?

  6. Hi,

    For Backups I use Allwaysync withn Versioning. The Backup goes to Pool which HDDs are in case that is connected with USB to the Server. The Server has the Pool with the original Data.

    Since a long time I have enabled the duplication of the most important folders. Every thing seemed to be OK with the Backup. This days I switched on the duplication for all folders and I have changed one HDD. The balancing was moving a lot of files arround and now the duplication is doing his job. What I see in the Backup is, that in the versioning there are being files copied in which I havend changed. This makes my Backup Pool extremly full. Normally Allwaysync will only do a copy of files which are new or have been changed. Can it be that the duplicated files will goe to my backup Pool too? I don't think so because the duplications are actually new files and schouldn't go to the versioning.´ but I'm not sure. maybe it's the balancing which lets Allwaysync think the files have been changed but if it is so why? The files have an old time stamp for the last change.

    Did anybody have this issue too? How are you doing Backups?

    Thanks for your help.

  7. @Shane Can you change the title of this thread in "DrivePool and SSD combinations" I think that would fix better to the subject, thanks.

    Not needed anymore. With holfing the right mouse button on the title I was able to change it by myself. Didn't see that feature before and found it now by chance.

  8. Thanks for your answers.

    This are the answers from t he costumer service. I post them here because others may have the same questions an will find this answers helpfull too:

    question 1:
    Is the duplication working with having one SSD-drive set for cache. Is DrivePool writing the files to the cache SSD-drive and a duplication to the HDDs in the Archive?
    answer:
    Yes, but if you have duplication enabled, and don't have two SSDs, then yeah, it's falling back to using one of the HDDs.

    question 2:
    How is the reading of this files, is DrivePool reading the files on the cache SSD-drive first if they are still there or will the duplication on the HDD slow it down?
    answer:
    if the files are duplicated, then it has a number of checks to optimize the reads, if read striping is enabled. 
    You can read about this here: https://stablebit.com/Support/DrivePool/2.X/Manual?Section=Performance Options#Read Striping

    question 3:
    Would it be better to have two SSD enabled in the SSD Optimizer as SSDs to get for the files which are actually in the cach the duplication in the cache too, would this work?
    answer:
    Yes.  Ideally. 

    question 4:
    If I use two SSDs in the cache how do I set them? Together in a subpool? Or adding them simple to the main pool and set them both as SSD?
    answer:
    Just add both to the pool, and enable them as "ssd"s in the SSD Optimizer balancers.

    question 5:
    Would it maybe a better solution to get one SSD in the pool and not enable it as SSD-drive in the "SSD Optimizer" to use it as archive with file placement rules for my most used files?
    answer:
    Honestly, either works, but it depends on what you want in your setup.  

    question 6:
    If I try it like in 5. how is it with the duplication would DrivePool read the files from the fast SSD or the slower HDD? Maybe the slow HDD makes the SSD slow too?
    answer:
    It should generally use the SSD, unless it's busy, basically. 

     

    The nice think is DrivePool is working how I thought it should be working :-)

    I ordered now another 1 TB SSD and will set both of my 1TB SSD in the optimizer as SSD. For my enabled duplication that should work best like the answer in question 3 says this is ideally. I will double my RAM to 16 GB too and keep my old CPU ans Mainboard.

  9. Yes I was only setting one drive as SSD cache. In the moment I have changed my setting to find out what is working how with DrivePool. Maybe I write my questions here too and you can have a look.

     

    My questions:

    1. Is the duplication working with having one SSD-drive set for cache. Is DrivePool writing the files to the cache SSD-drive and a duplication to the HDDs in the Archive?

    2. How is the reading of this files, is DrivePool reading the files on the cache SSD-drive first if they are still there or will the duplication on the HDD slow it down?

    3. Would it be better to have two SSD enabled in the SSD Optimizer as SSDs to get for the files which are actually in the cach the duplication in the cache too, would this work?

    4. If I use two SSDs in the cache how do I set them? Together in a subpool? Or adding them simple to the main pool and set them both as SSD?

    5. Would it maybe a better solution to get one SSD in the pool and not enable it as SSD-drive in the "SSD Optimizer" to use it as archive with file placement rules for my most used files?

    6. If I try it like in 5. how is it with the duplication would DrivePool read the files from the fast SSD or the slower HDD? Maybe the slow HDD makes the SSD slow too?

  10. Thanks Shane for your answer and help.

    My settings you can the in the screenshot. If I understood you right it should be the correct way, that the "file placement rules" are not override the the SSD Optimizer. What I don't undertand is why my SSD in the archive (with file placement to it and not set in the optimizer as SSD) and the Cache SSD (set as SSD in the optimizer) don't get new files which should be there. They remained empty.

    For A few minutes I was writing a support ticket because I can't find anythink in the documentation and in the forum what makes my questions clear. Can you help me with that ticket?

    20210305_001.JPG

  11. Thanks for your help.

    It seems we have realy a bit of an language problem. 

    I'm using drivepool now since a few yours but never used folder placement or SSD Optimizer before. Good do read that the SSD Cache is working. For me it wasn't working at all. I will test a bit and write to the support.

  12. Hmm I see  I mixed my questions about Archive SSD and Cache SSD :-(

    Maybe I wrote it not good enough. That the Cache is completely emptied, that's how I understand drivepool is working. Thge probelm now is that both SSD Archive and Cache are staying empty:

    For the Archive SSD I set the folder placement for only a few folders but they are not placed to the Archice SSD.

    Maybe it has a completly other reason because the SSDs are empty: Duplication. If I have enabled the duplication and have only 1 SSD as cache is that working? I thought DrivePool is moving the duplikated files to the archiv and let the main files in the cache till it is emptied. Maybe I'm wrong and I need two SSD in the cache. Maybe the Archive SSD has the same problem and can't make a dublication on an Archive HDD but this would be strange.

    In the screenshot *003 the both last drives are my SSDs. The Daten-6 is my Archive SSD.

    I read a lot now in the Forum and I have the feeling I should do this:

    - add another 1 TB SSD in my Server
    - remove the cache SSD from my pool
    - create a pool of both SSDs
    - add the new poot to the existing pool
    - mark the subpool as SSD in the SSD optimizer

    20210304_002.JPG

    20210304_003.JPG

  13. My oldest internal drive I use frequently is from 2010 (WD black) and  a few from 2011 (WD green, Barracuda green, Samsung). In the beginning of my Server I used "normal" HDDs because I had a some and they were cheaper. Since 2013 I buy WD red but only if I need them really. The old drive can run till they die. Only one Samsung/Seagate HDD has a few bad errors but is still working (I replaced it this days). Two Seagates have proplems with the LCC which is going very fast up. The WD black LCC goes a bit faster up to. Overall I'm very lucky with my HDDs.

    I have externel dirves too and they are older but working even if the LCC on some of them is very high.

  14. I have a question to your settings for the Cache SSD because I don't see that there is anything written on it. Did you allow the file placement to write on the cache SSD? I'm not sure how exactly that function works. If I enable it than DrivePool knows to use it as cache only because the SSD Optimizer Plugin says it or will it place Archive files than to it.

    The Idea is what I read from you. All written data should be go to the cache SSD and if a specific amount of date has to be balanced (600 GB in my case) it should be go from the cache SSD to the archive HDDs. I hope from this that my most used files stay for a long time on the fast cache.

    20210304_001.JPG

  15. 18 hours ago, gtaus said:

    Go to Manage Pool>Balancing>Settings>Automatic balancing - Triggers>check the box "Or if at least this much data needs to be moved:">set the trigger point to 100 GB.

    Mind you, I have a 228 GB SSD, so 100 GB is just under half usage. From what I have read, you want to keep your usage under 80% of the SSD or you risk wearing out the SSD prematurely as it will continue to use the same sectors over and over at the end of its available space as the drive fills up.

    Thanks! I had this allready set to 100 GB but because of your answer I set it a bit lower and changed the setting above so that ist works. If I use the SSD like this maybe I should use my 1 TB SSD für the Cache and not the small 128 GB one.

  16. I think I will have soon a looke to the speeds in my network too.

    11 hours ago, gtaus said:

    In DrivePool, I set my SSD cache not to flush until it reaches 100 GB. All files written to DrivePool will first go to the SSD cache, and then when it hits that 100 GB limit, it will start to flush the data to the archive HDDs in DrivePool. For my temp folders directed to DrivePool, it will use that 100 GB to both read and write as the data is still on the SSD and has not flushed yet to the archive HDDs. Since my server's main drive is a standard 7200 rpm HDD (about 100 MB/s), the SSD I set up for use in DrivePool is much faster at about 480 MB/s.

    How did you do that? I use the "SSD Optimizer" Plugin but I can only set how much the SSD can be filled, not at which point it will be emptied to the archiv. What do you mean with temp folders directed to the Pool?

  17. Hi gtaus,

    thanks for sharing your expirience.

    1 hour ago, gtaus said:

    DrivePool has a feature called Read Striping, which you can turn on. If the folder is duplicated, DrivePool will read from both drives and that can speed up the read task. If you expand the performance tab, you can see both drives that DrivePool is reading from and the speed at which it is reading. You can almost double the read speed in some transfers.

    Read Striping is turned on thanks for the hint I checked it.

    1 hour ago, gtaus said:

    What I see on my system is that DrivePool will read from both my SSD and an archive HDD at the same time, but since the SSD is just so much faster, most of the read data comes from the SSD. I don't know if DrivePool is specifically programmed to recognize the faster drive and use that pipeline first, but in effect that is what I see happening on my system. My read speed, with Read Striping turned on, in those cases is almost exactly the same as the read speed on my SSD.

    That sounds great. I was assuming it but was not shure about it. If it works in your case it will work for me too and I leave the Archive SSD in the Pool.

    1 hour ago, gtaus said:

    FYI, I only have 1 SSD, so I do not know if DrivePool would Read Strip both of your SSDs and almost double your read speed from the SSDs. Also, in my case, the bottleneck on my transfers is not the DrivePool read speeds, it is the speed of which I can transfer the data over my home network ethernet, the wifi, or to maybe a destination USB HDD. In most of my cases, transfers on my system are slowed down by lots of things but DrivePool is not one of them.

    I have only 1 SSD for Archive files, the othere is the writing cache. Striping would not work in this case. Would be interesting to have the douple speed of an SSD but I really thing there are to many bottlenecks to use this speed how you told. I only put this Archive SSD in the Pool because it's from my old Desktop which I don't use anymore and for writing cache I thought 1 TB is to big. If it speeds up a little bit it's a nice to have. Actually the OS is on an third SSD too.
    1. SSD = Archive
    2. SSD = writing cache
    3. SSD = OS

    1 hour ago, gtaus said:

    The SSD write cache did increase speed on my system, but it depends on my file transfer scenario. If I am transferring files directly from my server's drive into DrivePool on that same server, I see my maximum speed of the SSD (about 480 MB/s). If I am transferring files over my home network ethernet (about 30 MB/s), or wifi (? MB/s), or from an attached USB HDD (about 80 MB/s), then even my DrivePool archive HDDs are fast enough to keep up with the transfer.

    Do you have Gigabit ethernet? You're completely right, it depends on how I use the data to get the possible speed.

    1 hour ago, gtaus said:

    I have set aside 100 GB of my 228 GB SSD as my DrivePool write cache and also use that as a cache for my working temp folders. In that way, the 100 GB SSD cache not only works for DrivePool writes, but also as my system cache for both reads/writes for my temp working folders. In those cases, the SSD cache read/writes at about 480 MB/s whereas just my DrivePool archive HDDs would max out around 80-100 MB/s. So, yes, the SSD can significantly increase your speeds, but it all depends on your specific setup.

    Is this the writing cache I enabled for my writing cache SSD? It said everything is writen first to that SSD and than sorted to the Archiv Disks or did you set something different too?

  18. Hello together,

    I just optimized my Drivepool a bit and set a 128 GB SSD as write cache. I had another SSD with 1GB I had left I put into the pool and set drivepool to write only folders on it which I use most and should be fast available. This folders have also a duplication (the other Archive drives are HDDs). I'm not sure if this is working or the SSD is useless for more speed.

    How is drivepool working:

    - does drivepool read both the original und the duplicated folders if I use them from the Client?

    - will drive pool recognize which drive is faster and read from this first?

    best regards and thanks for your answers

×
×
  • Create New...