It was an unusual week in the analytics ecosystem as Hadoop monetarily stepped aside to let Google occupy the spotlight when it opened its cloud-based alternative to developers with the release of an open-source SDK designed to simplify the creation of Big Data processing services. The move raises the stakes in the race to bring about the era of pre-packaged analytic apps that pundits have been predicting for the last few years.

Cloud Dataflow, as the service is known, makes it possible to dissect historic information and real-time streams through the same programmatic interface with comparable functionality. That sharply contrasts with Hadoop, which requires the use of multiple components tuned to the specific workload. Complexity is often cited as a barrier to Hadoop adoption in the enterprise, and Google sees that as an opportunity.

While the search giant is taking its chances challenging Hadoop, the rest of the industry has mostly accepted its dominance, even traditional vendors such as Teradata Corp. that long prospered selling proprietary solutions. Teradata continued its refocusing campaign last week with the acquisition of RainStor Inc., a top provider of archiving software for the platform.

Born out of a project at the UK Ministry of Defence, the startup’s namesake database enables organizations in regulated industries such as financial services and telecommunications to harness Hadoop’s economies of scale for storing massive archives in regulator-friendly fashion. The acquisition was Teradata’s fourth in the Hadoop market in recent months, reaffirming its commitment to become a major force there.

NoSQL database maker Aerospike Inc. also made waves last week with the addition of greatly expanded support for Hadoop in its high-performance platform. Customers can now move data back and forth from their batch processing clusters without having to reformat it, speed up the time it takes to make information accessible for historical analysis and access the results as soon as they are available.

The latest version of the system offers several other notable improvements. One is the ability to customize how data requests are handled on an individual basis, a feature that can be useful for applications that handle different types of operations. Instead of having to choose between executing operations quickly or reliably, developers can now choose the approach that best suits each scenario.