Microsoft OneDrive mapped as network drive for Burst mining



  • Ok here is my google drive set up. I have 13 google drives with each a 15gb plot file fully optimized. That is 195GB free...

    it reads in about 16 sec



  • Could you explain how you set this up and mapped them thanks



  • Its pretty simply just fill your cloud drive with the largest plot possible and fully optimized.

    Then I use a program called net drive. Net Drive allows you to map your google cloud drive as a network drive. From there you just specify your drives in your miners config.

    Happy Mining!



  • thanks for the reply I was just wondering if it is the same way on a mac. I have both mac and windows machines


  • admin

    We have some miners that may not be able to effort new/bigger drives.
    There are no big profits possible with cloud drives, as they only offer a few GB of storage. But it is enough to participate.
    If i think of thousands 'little' miners using cloud drives, this would be good for burst network i think.

    Sure additional bandwidth is needed, but some may have a flatrate anyway.
    With 1MB per sec. or ~10Mbit download and 60sec. time, it is possible to mine ~250GB ... some may even have ~100Mbit, and 120sec. would be quite ok, too.
    That would make 5TB possible. For every case, please make sure you do not pay in any way for traffic.

    Thanks @crutsy for showing your setup.



  • I finally managed to map OneDrive as a network drive, and plot a 7,5 GB file from the portable Burst client to a local disk. I then uploaded it to OneDrive, with speeds ranging from 16-20 Mbps, thus taking about an hour to transfer. Is it an option to plot directly to OneDrive and then optimize the file in OneDrive using a plot optimization tool?

    I tried downloading the file from OneDrive just to test the allocated bandwidth. I had speeds ranging from 30-70 Mbps, but it mostly stayed around 70 Mbps. I guess these figures may vary a bit from day to day, as I have only tested it today. Based on 70 Mbps and a round time of 240 seconds, I could have a maximum plot size of 8,4 TB.

    Obviously, my 100 Mbit connection is capped by OneDrive, but in theory I may be able to mine a few burst coins given that I am allowed to upload 8,4 TB of plots in my OneDrive business account, and given that I am not reported for abusing their service by using too much bandwidth or something like that.

    When I tried to run the miner I got an error message on the screen as shown here
    0_1467719384030_2016-07-05_13-46-56.png

    Is this error connected with opening the file through a network drive (OneDrive), and is it an error that I can ignore? My Burst ID is BURST-8P3Q-7AQT-L9VK-53LKT, so you can check it against pool.burst-team.us. Here is my log file
    0_1467719592598_2016-07-05_13-52-57.png


  • admin

    @Propagandalf From the log it certainly looks like you're NOT mining or submitting nonces. You'll need to get to the bottom of that error opening file.



  • @haitch I presume that is because the log contains additional info when a deadline is found and submitted? Because when I mine an identical file through my "regular" network drive at work, I get no error message in the miner, and the log looks the same as far as I can tell. What both mining locations have in common is that the plot is so small that I won't find a deadline anyway, and so that is perhaps why the logs look similar.

    When I plot a file using the wallet client, it creates not only the plots file, but lots of other stuff in the Burst folder. Is any of that needed in order to read a plots file? I can't seem to locate the read error problem that I am getting mining through OneDrive.

    @crutsy Did you have any read errors when you mined through Google Cloud Drive?


  • admin

    @Propagandalf The additional files around the plot folder are the plotter files. They can be deleted after plotting has finished...



  • I never had any problems with google drive. I do remember trying One Drive last year when they still had 15GB free. For some reason it didn't work or was very slow I can't remember. I will try to upload a plot to microsoft one drive to test it out. The program you use to map the drive may have something to do with it also. It seems like google gives you the best access to their cloud drives.



  • @crutsy Thank you! If you try, and get the same error as me, then maybe the problem lies in some restriction on the server side. I still haven't come any closer to a solution.



  • @Propagandalf Sorry it took me so long I have been very busy. I have uploaded a 5GB plot to the free microsoft one drive account and I can say that it works just as good as the google drive. Doesn't seem to be any slower. No errors so far. I'll let you know if anything changes.



  • @crutsy
    Thanks for the reply. I'm glad it worked for you!

    Could you please give me a summary of each step you performed in order to get it working? I'm curious about for instance the miner and plotter type, how you mounted the drive, and where you plotted to and if you plotted directly or uploaded the file afterwards. If you uploaded, how did you do it?



  • @Propagandalf I'm on windows 7 64bit 6 core AVX cpu with 8GB memory.

    I try to keep everything very simple in the setup.

    I manually plot the file to the exact size of the cloud space. In this case Google's free 15GB's.
    I use the standard windows plotter wplotgenerator.exe, for me it is the most controllable way to plot, and 15Gb takes very little time.
    I then optimize the plot with plot optimizer GUI 1.0.3. It works great and is very easy to use. It is java tho.

    Once the 15GB plot is perfect size and optimized I will log into the cloud via the webpage. I find that this is an important step! Upload the plot... The web page seems to always be a faster upload and rarely fails. The upload is the part that takes the longest by far. Usually 6-8 Hours on my 25Mb internet. So again make sure your plot is perfect and not a mb over the cloud limit. Wait for the upload.

    When the plot is finished uploading log out of the web page.
    Open NetDrive, A program used to map cloud drives as network drives. I've tried a few others and obviously not all of them, but this one seems to work the best. The demo version only last 30 days at that point you can only connect one drive and it seems to cut the bandwidth in half. So try to get your hands on a full copy right out of the gate.
    Net drive is very easy to use just log in to the cloud service and connect. It will give a drive number which you can also see in My Computer. I left all the default settings in netdrive.

    From there I add the new drive letter to my miner config. I personally use Blago's miner, for many reasons, It is best on windows 7 and it compliments my setup well with the proxy feature.

    Your internet will become the bottle neck.

    So far I've only used Google and OneDrive. I can say most of the others don't have more than 5GB free. Hubic looked good at 50GB free but it fails to upload at 2-3GB and is horribly unusable slow.

    If you have any questions let me know



  • @crutsy Thank you for this detailed answer! I really appreciate it, and will give the plotting/mining on OneDrive another go as soon as I find time for it.



  • @crutsy @haitch @dawallet

    OK so looks like I found time right away, lol. I set up the networked drive using NetDrive and tried mining the same plot files that I originally had problems reading from work (but now I am at home using a different PC). This time I had no read errors at all.

    This is what I think has caused the problems:
    For some reason, I managed to map a folder called 'plots' and not the root directory within OneDrive as a network drive on my work PC, because I was first messing about with manually creating folders and uploading plots through the browser interface before mapping it as a network drive. So, when I later accessed the drive through my computer it would look like I was in the root directory, except I wasn't. So, I assume the CPU miner 'found' the file in G: - Burst - plots and tried to read it, but failed to open it, because the real path (visible if I logged in to OneDrive through the browser) was G: - Burst - plots - Burst - plots (yes, I know that looks retarded, and it is).

    Solution:
    I need to remove the mapped network drive at work, and map it again, this time mapping it to the root directory.



  • @blago
    Since reading through plot files on a network drive takes longer (because of internet speed bottleneck), I would like my miner to read those plots after reading plots on my internal and external drives, so that I can submit as many deadlines as possible before the next round starts. How can I configure my CPU miner to prioritize reading specific drives? What is the logic?

    This is from my miner.conf: "G:\Burst\plots","F:\Burst\plots","E:\Burst\plots","C:\Burst\plots", but it reads in this order: C, G, F, E. My network drive is G.


  • Mod

    @Propagandalf no, miner reads ALL paths parallelly. In you case, C < G < F < E, that's why miner final to read paths in that order.

    if you want read paths in series, you may use "+", like
    "G:\Burst\plots+F:\Burst\plots+E:\Burst\plots+C:\Burst\plots"
    logically, this will 1 Path (I'm not recommend)



  • @Blago
    I was not aware that it reads all drives in parallell. Does that mean that every time I see a white line in the miner (i.e. Thread "C:\Burst\plots" @0.06) it is simply because that particular drive has been finished earlier than the rest?

    If I understand correctly, configuring the miner to read 1 path will be slower then, because it will first read through one drive before moving on to the next (series, not in parallell)?


  • Mod

    @Propagandalf "Yes, that's right" to both questions


Log in to reply
 

Looks like your connection to Burst - Efficient HDD Mining was lost, please wait while we try to reconnect.