This has been a rewarding journey. Hundreds of customers are using MapR every month.
Today we are excited to announce the availability of the MapR M7 Edition on EMR. M7 is an enterprise-grade, high performance platform for Apache HBase. In addition to MapR's standard enterprise-grade Hadoop capabilities, the M7 architecture provides seamless splits, no compactions and instant recovery from any failure, delivering consistent low latency and high throughput for all HBase operations.
M7 is supported on multiple AWS instance types, including the latest high performance instances with SSD storage, and can scale to thousands of nodes (instances) in a single cluster. YCSB benchmark tests on hs1 instances have shown M7 delivering over 100,000 operations per second per node.
EMR users can dynamically provision M7 clusters to process and analyze data that already resides in the cloud, such as rich, high-volume sources of search and click-stream data, without costly and time-consuming data-integration and data-movement steps.
To launch an M7 cluster, select the MapR M7 Edition in the EMR New Job Flow Wizard or use the elasticmapreduce CLI. To launch the latest version of M7 on EMR, use the following command:
./elastic-mapreduce --create --alive --instance-type hi1.4xlarge --num-instances 5 --supported-product mapr --args "--edition,m7"