Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>These differences can be attributed to 1) communication overhead (especially if you run across nodes) and 2) performance overhead (if your job is not that intensive compared to initiating a parallelisation, for example). Usually, if the task you are parallelising is not that time-consuming, then you will mostly find that parallelisation does NOT have much of an effect (which is much highly visible on huge datasets.</p> <p>Even though this may not directly answer your benchmarking, I hope this should be rather straightforward and can be related to. As an example, here, I construct a <code>data.frame</code> with <code>1e6</code> rows with <code>1e4</code> unique column <code>group</code> entries and some values in column <code>val</code>. And then I run using <code>plyr</code> in <code>parallel</code> using <code>doMC</code> and without parallelisation.</p> <pre><code>df &lt;- data.frame(group = as.factor(sample(1:1e4, 1e6, replace = T)), val = sample(1:10, 1e6, replace = T)) &gt; head(df) group val # 1 8498 8 # 2 5253 6 # 3 1495 1 # 4 7362 9 # 5 2344 6 # 6 5602 9 &gt; dim(df) # [1] 1000000 2 require(plyr) require(doMC) registerDoMC(20) # 20 processors # parallelisation using doMC + plyr P.PLYR &lt;- function() { o1 &lt;- ddply(df, .(group), function(x) sum(x$val), .parallel = TRUE) } # no parallelisation PLYR &lt;- function() { o2 &lt;- ddply(df, .(group), function(x) sum(x$val), .parallel = FALSE) } require(rbenchmark) benchmark(P.PLYR(), PLYR(), replications = 2, order = "elapsed") test replications elapsed relative user.self sys.self user.child sys.child 2 PLYR() 2 8.925 1.000 8.865 0.068 0.000 0.000 1 P.PLYR() 2 30.637 3.433 15.841 13.945 8.944 38.858 </code></pre> <p>As you can see, the <strong>parallel</strong> version of <code>plyr</code> runs <strong>3.5 times slower</strong></p> <p>Now, let me use the same <code>data.frame</code>, but instead of computing <code>sum</code>, let me construct a bit more demanding function, say, <code>median(.) * median(rnorm(1e4)</code> ((meaningless, yes):</p> <p>You'll see that the tides are beginning to shift:</p> <pre><code># parallelisation using doMC + plyr P.PLYR &lt;- function() { o1 &lt;- ddply(df, .(group), function(x) median(x$val) * median(rnorm(1e4)), .parallel = TRUE) } # no parallelisation PLYR &lt;- function() { o2 &lt;- ddply(df, .(group), function(x) median(x$val) * median(rnorm(1e4)), .parallel = FALSE) } &gt; benchmark(P.PLYR(), PLYR(), replications = 2, order = "elapsed") test replications elapsed relative user.self sys.self user.child sys.child 1 P.PLYR() 2 41.911 1.000 15.265 15.369 141.585 34.254 2 PLYR() 2 73.417 1.752 73.372 0.052 0.000 0.000 </code></pre> <p>Here, the <strong>parallel</strong> version is <code>1.752 times</code> <strong>faster</strong> than the non-parallel version.</p> <p><strong>Edit:</strong> Following @Paul's comment, I just implemented a small delay using <code>Sys.sleep()</code>. Of course the results are obvious. But just for the sake of completeness, here's the result on a 20*2 data.frame:</p> <pre><code>df &lt;- data.frame(group=sample(letters[1:5], 20, replace=T), val=sample(20)) # parallelisation using doMC + plyr P.PLYR &lt;- function() { o1 &lt;- ddply(df, .(group), function(x) { Sys.sleep(2) median(x$val) }, .parallel = TRUE) } # no parallelisation PLYR &lt;- function() { o2 &lt;- ddply(df, .(group), function(x) { Sys.sleep(2) median(x$val) }, .parallel = FALSE) } &gt; benchmark(P.PLYR(), PLYR(), replications = 2, order = "elapsed") # test replications elapsed relative user.self sys.self user.child sys.child # 1 P.PLYR() 2 4.116 1.000 0.056 0.056 0.024 0.04 # 2 PLYR() 2 20.050 4.871 0.028 0.000 0.000 0.00 </code></pre> <p>The difference here is not surprising.</p>
    singulars
    1. This table or related slice is empty.
    1. POWhy is the parallel package slower than just using apply?
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USArun
    1. USArun
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. CO+1 for the example code. In the end it all boils down to effectively cutting up your problem in sizeable chunks.
      singulars
      1. PO
      1. USPaul Hiemstra
    2. COA very trivial example is to use a function that just sleeps for ten seconds, there running in parallel really works well. If you cut up a 20 hour job into 60 pieces, parallization is going to present a major improvement. For example processing a satellite image per ten rows of pixels is much efficient than feeding just one pixel per time to the workers, so slice wisely. And in case of plyr, switching to data.table will probably present a much greater improvement than running in parallel. So it is also a matter of choosing the right tool.
      singulars
      1. PO
      1. USPaul Hiemstra
    3. COIn addition, when running models in parallel from within R, using six workers presented a 5.6 times increase in performance, but those runs took hours.
      singulars
      1. PO
      1. USPaul Hiemstra
 

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