Most Hadoop deployments to date have met the needs of multiple user groups (aka tenants) through a “shared nothing” model, with siloed physical infrastructure that often results in low utilization and high management overhead.
However, as enterprises build out and expand their Hadoop clusters, there is an increasing imperative to provide secure multi-tenant environments with shared infrastructure – to improve business agility, reduce overhead, and improve utilization.
There is a Hadoop reference architecture for multi-tenancy on a single physical cluster based on YARN. But in real-world scenarios, where multiple versions and/or different distributions of Hadoop need to co-exist with strict isolation, this model is a non-starter. Another challenge with this approach is the fact that complex, time-consuming configurations are required to maintain compute and data isolation between tenants.
To meet today’s enterprise requirements, a secure multi-tenant Hadoop architecture must accommodate multiple user groups, multiple concurrent Hadoop jobs, multiple applications, multiple versions and/or distributions of Hadoop, security isolation, and service level guarantees while utilizing shared infrastructure.
This session will discuss these requirements and provide recommendations on how to deploy a secure multi-tenant, multi-cluster Hadoop environment.
This session is sponsored by BlueData
Anant Chintamaneni is vice president of products at BlueData, where he is responsible for product management and focuses on helping enterprises deploy big data technologies such as Hadoop and Spark. Anant has more than 15 years’ experience in business intelligence, advanced analytics, and big data infrastructure. Previously, Anant led the product management team for Pivotal’s big data suite.
©2015, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • email@example.com
Apache Hadoop, Hadoop, Apache Spark, Spark, and Apache are either registered trademarks or trademarks of the Apache Software Foundation in the United States and/or other countries, and are used with permission. The Apache Software Foundation has no affiliation with and does not endorse, or review the materials provided at this event, which is managed by O'Reilly Media and/or Cloudera.