Databases
StackOverflow2013
Postsdbo
POWhy does this python multiprocessing script slow down after a while?
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POWhy does this python multiprocessing script slow down after a while?
    primarykey
    data
    text
    <p>Building on the <a href="https://stackoverflow.com/questions/18204782/runtimeerror-on-windows-trying-python-multiprocessing">script from this answer</a>, I have the following scenario: A folder containing 2500 large text files (~ 55Mb each), all tab delimited. Web logs, basically.</p> <p>I need to md5 hash the second 'column' in each row of each file, saving the modified files elsewhere. The source files are on a mechanical disk and the destination files are on an SSD.</p> <p>The script processes the first 25 (or so) files really quickly. It then slows WAY down. Based on the first 25 files, it should complete all of them in 2 minutes or so. However, based on the performance after that, it will take 15 minutes (or so) to complete them all.</p> <p>It's running on a server with 32 Gb of RAM and task manager rarely shows over 6 Gb in use. I have it set to launch 6 processes, but the CPU usage on the cores is low, rarely going over 15%.</p> <p>Why is this slowing down? Read/write issues to the disk? Garbage collector? Bad code? Any ideas about how to speed it up?</p> <p>Here's the script</p> <pre><code>import os import multiprocessing from multiprocessing import Process import threading import hashlib class ThreadRunner(threading.Thread): """ This class represents a single instance of a running thread""" def __init__(self, fileset, filedirectory): threading.Thread.__init__(self) self.files_to_process = fileset self.filedir = filedirectory def run(self): for current_file in self.files_to_process: # Open the current file as read only active_file_name = self.filedir + "/" + current_file output_file_name = "D:/hashed_data/" + "hashed_" + current_file active_file = open(active_file_name, "r") output_file = open(output_file_name, "ab+") for line in active_file: # Load the line, hash the username, save the line lineList = line.split("\t") if not lineList[1] == "-": lineList[1] = hashlib.md5(lineList[1]).hexdigest() lineOut = '\t'.join(lineList) output_file.write(lineOut) # Always close files after you open them active_file.close() output_file.close() print "\nCompleted " + current_file class ProcessRunner: """ This class represents a single instance of a running process """ def runp(self, pid, numThreads, fileset, filedirectory): mythreads = [] for tid in range(numThreads): th = ThreadRunner(fileset, filedirectory) mythreads.append(th) for i in mythreads: i.start() for i in mythreads: i.join() class ParallelExtractor: def runInParallel(self, numProcesses, numThreads, filedirectory): myprocs = [] prunner = ProcessRunner() # Store the file names from that directory in a list that we can iterate file_names = os.listdir(filedirectory) file_sets = [] for i in range(numProcesses): file_sets.append([]) for index, name in enumerate(file_names): num = index % numProcesses file_sets[num].append(name) for pid in range(numProcesses): pr = Process(target=prunner.runp, args=(pid, numThreads, file_sets[pid], filedirectory)) myprocs.append(pr) for i in myprocs: i.start() for i in myprocs: i.join() if __name__ == '__main__': file_directory = "E:/original_data" processes = 6 threads = 1 extractor = ParallelExtractor() extractor.runInParallel(numProcesses=processes, numThreads=threads, filedirectory=file_directory) </code></pre>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. USCommunity
    1. USClay
    plurals
    1. PL
      singulars
      1. LTLinked
    1. PL
      singulars
      1. LTLinked
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    2. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POWhy does this python multiprocessing script slow down after a while?
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. POWhy does this python multiprocessing script slow down after a while?
      1. This table or related slice is empty.
      1. VTUpMod
    1. COIt's possible that you're getting a performance boost because the first files are cached in memory by the operating system, so no disk I/O happens. You can easily check this by rebooting the server and see if the processing slows down. If you can't reboot, you should fill the cache with something else, by reading enough files from the disk to fill the physical memory. If you have local access, you can simply listen closely for disk seeks. It's worth mentioning that performing hashing on files is most certainly bound by the disk, not CPU, so performing it in parallel is useless in the best case
      singulars
      1. POWhy does this python multiprocessing script slow down after a while?
      1. USgoncalopp
    2. COIn addition, if your source files are on a mechanical disk, it's probable that 6 processes reading from them at once will slow you down tremendously, especially on Windows where I/O scheduling is notoriously bad. What happens when you move your source files to a SSD?
      singulars
      1. POWhy does this python multiprocessing script slow down after a while?
      1. USMax Noel
    3. COActually, if I/O scheduling is the issue (which is likely if your CPU usage stays low), you should get an increase in performance by lowering ``numProcesses`` to 1.
      singulars
      1. POWhy does this python multiprocessing script slow down after a while?
      1. USMax Noel
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload