Created
August 10, 2012 17:35
-
-
Save jmikola/3315992 to your computer and use it in GitHub Desktop.
Test batch insertion of documents and referenced GridFS files
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #!/usr/bin/env php | |
| <?php | |
| require_once 'config.php'; // mongodb-odm/tools/sandbox/config.php | |
| use Doctrine\ODM\MongoDB\Mapping\Annotations as ODM; | |
| /** | |
| * Test batch insertion of documents and referenced GridFS files. | |
| * | |
| * For each test iteration, a number (i.e. batch size) of documents and files | |
| * will be persisted and flushed. Time and peak memory usage will be recorded | |
| * between tests, with the collections dropped and DocumentManager cleared | |
| * between tests. | |
| * | |
| * CLI arguments determine the size of the initial and final batches, and the | |
| * step to increase the batch size between iterations. | |
| */ | |
| if (4 !== $argc) { | |
| printf("Usage: %s [start] [end] [step]\n", basename(__FILE__)); | |
| exit; | |
| } | |
| $test = function($iterations) use ($dm) { | |
| for ($i = 0; $i < $iterations; ++$i) { | |
| $image = new Image(); | |
| $image->name = 'foo'; | |
| $image->file = __FILE__; | |
| $profile = new Profile(); | |
| $profile->name = 'bar'; | |
| $profile->image = $image; | |
| $dm->persist($profile); | |
| } | |
| $dm->flush(); | |
| }; | |
| $units = array('b', 'k', 'm', 'g', 't', 'p'); | |
| for ($i = $argv[1]; $i <= $argv[2]; $i += $argv[3]) { | |
| $dm->getDocumentCollection('Profile')->drop(); | |
| $dm->getDocumentCollection('Image')->drop(); | |
| $start = microtime(true); | |
| $test($i); | |
| $elapsedTime = microtime(true) - $start; | |
| $dm->clear(); | |
| $size = memory_get_peak_usage(true); | |
| $peakMem = round($size / pow(1024, ($_ = floor(log($size, 1024)))), 2).$units[$_]; | |
| printf("size: %d, time: %f, peak: %s\n", $i, $elapsedTime, $peakMem); | |
| } | |
| /** @ODM\Document(collection="profiles") */ | |
| class Profile | |
| { | |
| /** @ODM\Id */ | |
| public $id; | |
| /** @ODM\String */ | |
| public $name; | |
| /** @ODM\ReferenceOne(targetDocument="Image", cascade={"all"}) */ | |
| public $image; | |
| } | |
| /** @ODM\Document(collection="images") */ | |
| class Image | |
| { | |
| /** @ODM\Id */ | |
| public $id; | |
| /** @ODM\String */ | |
| public $name; | |
| /** @ODM\File */ | |
| public $file; | |
| } |
To add some context to this gist, it's related to: https://jira.mongodb.org/browse/PHP-453
Thanks to a one-line optimization in UnitOfWork::computeAssociationChanges() (doctrine/mongodb-odm@6653a60), execution time is more than halved:
$ ./gridfs_batch.php 50000 50000 10000
size: 50000, time: 128.727638, peak: 667m
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Results for batches in the early thousand range:
And for a large 50k batch size, with mongostat output (starting with actual Mongo insertions three minutes after the test began):