amazon web services - Reading many small files from S3 very slow -


loading many small files (>200000, 4kbyte) s3 bucket hdfs via hive or pig on aws emr extremely slow. seems 1 mapper used data, though cannot figure out bottleneck is.

pig code sample

data = load 's3://data-bucket/'  using pigstorage(',') (line:chararray)

hive code sample

create external table data (value string) location  's3://data-bucket/';

are there known settings speed process or increase number of mappers used fetch data?

i tried following without noticeable effects:

  • increase #task nodes
  • set hive.optimize.s3.query=true
  • manually set #mappers
  • increase instance type medium xlarge

i know s3distcp speed process, better performance doing lot of tweaking including setting #workerthreads , prefer changing parameters directly in pig/hive scripts.

you can either :

  1. use distcp merge file before job starts : http://snowplowanalytics.com/blog/2013/05/30/dealing-with-hadoops-small-files-problem/

  2. have pig script you, once.

if want through pig, need know how many mappers spawned. can play following parameters :

//  set mapper = nb block size. set true 1 per file. set pig.nosplitcombination false; // set size have sum(size) / x = wanted number of mappers set pig.maxcombinedsplitsize 250000000;

please provide metrics thoses cases


Comments

Post a Comment

Popular posts from this blog

Magento/PHP - Get phones on all members in a customer group -

i working on custom magento extension. working around adminhtml part , i've created custom form there. here form code: <?php class vivasindustries_smsnotification_block_adminhtml_sms_sendmass_edit_form extends mage_adminhtml_block_widget_form { public function _preparelayout() { $extensionpath = mage::getmoduledir('js', 'vivasindustries_smsnotification'); $head = $this->getlayout()->getblock('head'); $head->addjs('jquery.js'); $head->addjs('vivas.js'); return parent::_preparelayout(); } protected function _prepareform() { $form = new varien_data_form(array( 'id' => 'edit_form', 'action' => $this->geturl('*/*/save', array('id' => $this->getrequest()->get...
Read more

php - Bypass Geo Redirect for specific directories -

i'm trying bypass redirect specific url within site. wondering if might able help. the code below code use, original urls replaces example.com i'd able access http://example.com/admin without redirect occuring. is there way this, , there way using own ip? function country_geo_redirect() { $country = getenv('http_geoip_country_code'); if ($country == "us" ) { wp_redirect("http://usa.example.com/"); exit; } else if ($country == "gb" ) { wp_redirect("http://uk.example.com/"); exit; } } add_action('init', 'country_geo_redirect'); you can bypass redirection code if server found ip address. if ip address == <my ip address> ' nothing here else country_georedirect() end if
Read more

php - .htaccess mod_rewrite for dynamic url which has domain names -

i want convert url ex:( example.com/link.php?url=facebook.com ) ( example.com/facebook.com.php ) how can using .htaccess. or there other process make job easier ? i tried adding script .htaccess file options +followsymlinks rewriteengine on rewriterule url-(.*)\.htm$ link.php?url=$1 but converting url adding "url-" ( example.com/url-facebook.com.php ) i want remove "url-" , want make ( example.com/facebook.com.php ) rewriteengine on rewritecond %{the_request} ^get.*link\.php\?url\= [nc] rewritecond %{request_uri} !/system/.* [nc] rewriterule (.*?)link\.php\?url\=/*(.*) /$1$2 [r=301,ne,l] not sure how above work edit of used use remove index.php url
Read more