Strata Hadoop World NY 2016 - Hadoop internals & development sessions
Strata Hadoop World NY 2016 had following interestinig talks in its Hadoop internals & development sessions.
Tuning Impala: The top five performance optimizations for the best BI and SQL analytics on Hadoop by Marcel Kornacker and Mostafa Mokhtar
Performance tuning your SQL-on-Hadoop deployment may seem overwhelming at times, especially for BI workloads that need interactive response times with high concurrency. Marcel Kornacker and Mostafa Mokhtar simplify the process and cover top performance optimizations for Apache Impala (incubating), from schema design and memory optimization to query tuning.
Elastic data services on Mesos via Mesosphere’s DC/OS by Adam Bordelon and Mohit Soni
Adam Bordelon and Mohit Soni demonstrate how projects like Apache Myriad (incubating) can install Hadoop on Mesosphere DC/OS alongside other data center-scale applications, enabling efficient resource sharing and isolation across a variety of distributed applications while sharing the same cluster resources and hence breaking silos.
Debunking HDFS erasure coding performance myths by Zhe Zhang and Uma Maheswara Rao G
The new erasure coding feature in Apache Hadoop (HDFS-EC) reduces the storage cost by ~50% compared with 3x replication. Zhe Zhang and Uma Maheswara Rao G present the first-ever performance study of HDFS-EC and share insights on when and how to use the feature.
Apache Kudu: 1.0 and beyond by Todd Lipcon
Apache Kudu was first announced as a public beta release at Strata NYC 2015 and recently reached 1.0. This conference marks its one year anniversary as a public open source project. Todd Lipcon offers a very brief refresher on the goals and feature set of the Kudu storage engine, covering the development that has taken place over the last year.
Rethinking operational data stores on Hadoop by Vinayak Borkar
Starting from first principles, Vinayak Borkar defines the requirements for a modern operational data store and explores some possible architectures to support those requirements.