S3Guard

Mingliang Liu and Rajesh Balamohan explain why you shouldn’t use S3 as your primary Hadoop data store, as well as a tool which helps mitigate those problems:

Some of the real world use cases which can be impacted due to the S3 eventual consistency model are:

  1. Listing Files. Newly created files might not be visible for data processing. In Hive, Spark and MapReduce, this can lead to erroneous results from incomplete source data or failure to commit all intermediate results.

  2. ETL Workflow. Systems like Oozie rely on marker files to trigger the subsequent workflows. Any delay in the visibility of these files can lead to delays in the subsequent workflows.

  3. Existence-guarded path operations. Any action which fails if the destination path is present may see a deleted file in a listing, and so fail — even though the file has already been deleted.

Read on to see how S3Guard works and how to enable it in HDP 2.6.

Related Posts

Calculated Columns and Memory Usage in Analysis Services

Teo Lachev troubleshoots a customer issue: Scenario: A client reports a memory spike during processing. They have a Tabular semantic model deployed to Azure Analysis Services. They fully process the model daily. The model normally takes less than 50 GB RAM but during processing, it spikes five times and Azure Analysis Services terminates the processing task […]

Read More

Performance Tuning Neural Network Training

Sean Owen takes us through a few techniques for speeding up neural network model training: Step #2: Use Early StoppingKeras (and other frameworks) have built-in support for stopping when further training appears to be making the model worse. In Keras, it’s the EarlyStopping callback. Using it means passing the validation data to the training process for evaluation […]

Read More

Categories

June 2017
MTWTFSS
« May Jul »
 1234
567891011
12131415161718
19202122232425
2627282930