This webinar will describe the concept and internal mechanism using the stack of Spark+Alluxio in Kubernetes to enhance data locality even when the storage service is outside or remote.
We will introduce the key new features and enhancements such as: Support for hyper-scale data workloads, Machine learning and deep learning workloads, and Better storage abstraction.
This webinar reviews: The observation and analysis of trends of separation of Storage and Compute in Big Data ecosystem; Why and how to build a new data access layer between compute and storage in this data stack; Alluxio open source: history, overview, design, and architecture; Production Use case with Spark, Presto, Tensorflow and etc; A demo of running Presto on Alluxio on S3
Joint webinar – Mesosphere DC/OS is a production-proven platform that powers both modern app components – containers and data services – so businesses can accelerate time to market with confidence, and save. We have seen tremendous interest from users to be able to run Alluxio via DC/OS.