Do More with Amazon EMR and MapR

I’m writing to share some good news about the MapR Distribution on Amazon EMR. As of 4/6, version 4.0.2 of the MapR Distribution including Hadoop will be available for customers to deploy using Amazon EMR. Customers can quickly and easily use Amazon EMR to deploy and manage MapR, which features updated Apache projects - including YARN and Hive 0.13 - and takes advantage of new features, including promotable mirrors, that MapR has added to our differentiated MapR data platform.

With the introduction of YARN, the already strong multi-tenancy features from MapR get even better. YARN enables enterprises to deploy multiple data processing engines, including batch-oriented systems like MapReduce and Hive, along with more real-time components like Storm. Speaking of Hive, this new release of MapR on Amazon EMR includes Hive 0.13, which our joint Amazon EMR – MapR customers have requested for its dramatic performance increase and expanded support for SQL-like operations such as subqueries.

Beyond improvements delivered by the open source community, MapR continues to invest in our differentiated MapR data platform. We’re excited to offer customers of the MapR Community Edition on Amazon EMR access to our top-ranked NoSQL key-value store, MapR-DB, which supports real-time data ingestion and higher read and write speeds than HBase while requiring zero administration. So if you are struggling to impact your business “as it happens” due to latency spikes and region server failures in HBase, I definitely recommend that you try out MapR-DB by quickly launching an Amazon EMR cluster with MapR Community Edition today – it’s easy, and there is no additional cost!  

Finally, MapR continues to improve your options for disaster recovery and intra-cluster synchronization with our new promotable mirror feature. This capability allows you to mirror MapR storage volumes between clusters and dynamically promote the read-only mirror target to an active read-write data volume. The result is a dramatically lower recovery time objective (RTO) for your critical applications. Additionally, the promotable mirror feature can be used to efficiently transfer operational data from your production cluster to development or QA clusters with minimum overhead. 

To learn more about MapR on Amazon EMR, including deployment options and pricing, visit this page.


Streaming Data Architecture:

New Designs Using Apache Kafka and MapR Streams




Download for free