SAP HANA Gets an IoT Upgrade
The latest version of the SAP HANA in-memory database and application platform rolled out this week predictably emphasizes big data analytics while adding greater connectivity to the emerging Internet of Things (IoT).
SAP said Tuesday (June 16) its latest HANA release (SPS 10, or service pack 10) attempts to shift more functionality to the network edge where, the company asserts, business transactions and analytic workloads increasingly occur. At the same time, the new capabilities beef up datacenter "readiness" as more data must be synchronized with remote systems.
The HANA upgrades also reflect the growing shift toward moving processing power closer to the data that increasingly resides at the network edge. "We are readying our customers for the inevitable digitization of our entire economy," asserted Quentin Clark, SAP's chief technology officer.
The IoT is emerging as the latest manifestation of the relentless hyper-scale push sweeping enterprises. Much of this is being driven by the steady connection of devices and sensors to corporate networks. Forecasts on the number of connected "things" continue to grow exponentially. For example, 451 Research calculates that machine-to-machine connections alone could grow four-fold through 2019 to an estimated 908 million.
"We continue to be bullish that ultimately the hype of IoT will be proven to be warranted," 451 Research noted in a report released this week.
Along with its IoT emphasis, the latest version of SAP HANA released at a company event in Nice, France, includes an upgraded version of SAP's predictive analytics suite that is said to crunch "ultra-wide datasets." In a bid to streamline data access and improve big data management, HANA said its new analytics package includes the latest Hadoop distributions from Hortonworks and enterprise data hub specialist Cloudera.
The analytics upgrades also include a data transfer boost via Spark SQL along with a unified interface for Hadoop cluster management with SAP HANA using Apache Ambari. Taken together, the tools are making Apache Hadoop ubiquitous across enterprise IT, proponents note.
SAP also said it would offer multiple storage options as a way to speed access to data. In one example, data could be stored for a fixed amount of time in memory as older data is moved to Hadoop or disk storage. "Smart" data management features also include the ability to "cleanse" data via de-duplication.
What is needed to make all this work, of course, is scaling. With that in mind, SAP said the latest version of HANA targets large-scale infrastructure with more than 12 TB of memory required as big data workloads begin to dominate datacenters.
As SAP HANA's in-memory database scales out to the cloud, the software giant also has been fine-tuning its platform with workload management enhancements designed to optimize in-memory resources while boosting the ability to handle mixed workloads.
With its latest release, SAP also is betting that the torrent of data generated by connected devices will fuel the need within enterprises for greater big data capabilities. At the same time, it is attempting to buttress datacenters for the big data onslaught through features like remote data synchronization between corporate headquarters and remote offices at the edge of the network.
