Analyzing Windows Azure Storage Logs

A few weeks ago, I wrote a post that described how to maximize throughput between HDInsight clusters and Windows Azure Storage. One of the suggestions I made was to adjust your HDInsight cluster’s self-throttling mechanism – i.e. tune the fs.azure.selfthrottling.read/write.factor parameters. I also suggested that the best way to find the optimal parameter values was…

0

Accessing Hadoop Logs in HDInsight

One of the questions the HDInsight team sees a lot is a variation of the question “How do I figure out what went wrong when something does go wrong?” If you are familiar with Hadoop, you are probably also familiar with rolling up your sleeves and digging into Hadoop logs to answer this question. However,…

4