Google has replaced its MapReduce analytics service in favour of a cloud analytics system it has built called Cloud Dataflow.
MapReduce is based on the popular Apache Hadoop-based analytics platform and analyses data in batch mode, which allows companies to build their own big data application using low-cost hardware.
Cloud Dataflow, announced at Google’s I/O keynote in San Francisco, is based on Google’s own technologies, Flume and MillWheel, and does not have the same scaling restrictions of MapReduce.
The service, which is not yet available to developers, claims to allow companies to more easily build complex ‘data pipelines’, view trends and be alerted to insights in real-time.
Cloud Dataflow is expected to take on Amazon Web Services’ Redshift, a datawarehouse service, and Elastic MapReduce, which uses Hadoop to crunch large datasets.
Other services announced on Google’s cloud platform include Cloud Save, Cloud Debugging, Cloud Tracing, which provides latency statistics across different groups, and Cloud Monitoring.
