SAP HANA Vora runs on Apache Spark and was announced by SAP four years ago. Technology validation has begun around the world, partly because the role of the product has become more relevant as it strengthens collaboration with Spark, which is also the future standard as a big data base. HANA Vora is software that enables in-memory interactive analysis of data from the platform “Hadoop” that processes large-scale data in a distributed manner. Extend and use the execution framework of the in-memory distributed processing platform “Apache Spark“.
In conjunction with the company’s in-memory database “SAP HANA” and the HANA platform, it enables analysis combining unstructured data such as data obtained by IoT (Internet of Things) and transaction data of business applications possessed by HANA. . In addition to being able to stand-alone in Hadoop environments that do not uses SAP HANA, the company’s cloud service “SAP HANA Cloud Platform” seems to offer it as a SaaS (Software as a Service).
Vora is derived from ‘voracious’ meaning “appreciate” and has a strong desire for data. This technology can be said to embody the notion that big data and business intelligence (BI) are mutually complementary. Vora is a big data technology such as “SAP HANA” that is an existing product, “SAP Business Warehouse (BW)”, “SAP BusinessObjects”, “SAP ERP”, “Apache Hadoop” and “Apache Spark”, which are positioned downstream.
HANA Vora will be able to perform drill-down analysis (multi-faceted analysis) with interactive operations such as OLAP (OnLine Analytical Processing) on traditional transaction data even for data in the IoT and Big Data fields. One of the points. In addition, data scientists and developers will use Spark R and Spark ML to mash up corporate data and Hadoop / Spark data, and explain that it is an innovation that lowers the hurdles in big data analysis of the enterprise.
Existing Technology and New Technology
HANA Vora is based on the combination of Hadoop 2.0 / YARN and Apache Spark. Then, by delegating push-down queries, the connection function with the existing product SAP HANA is provided. In addition, Spark SQL is also available, providing the same functionality as hierarchical queries and compiled queries that relational databases and data warehouses have been using for years.
In short, Vora can be said to combine data lakes with structured data owned by companies without causing the cost of data movement (in order to execute unique queries in the case of “conventional” HANA). Furthermore, while realizing multi-dimensional hierarchy definition, it is possible to use the definition in analysis queries for all data handled by Vora.
No Need for Dedicated Hardware
Vora does not require a dedicated hardware infrastructure to coexist with cluster nodes on which Hadoop and Spark are deployed. If you want to integrate Vora with your existing HANA, of course, you need HANA’s own infrastructure. However, Vora does not require any additional hardware and can be used standalone.
Vora is a very reasonable way for SAP customers who want to use technologies such as Hadoop, Spark and Big Data. Since Vora is commercial software provided by SAP and not open source software, it can be adapted to the existing business model without making any strategic changes.
Vora realizes an exquisite combination in various ways. These include big data and BI, emerging technologies and proven enterprise software, data lakes and carefully organized data, in-memory and disk based storage and processing.
Does HANA Vora work with HANA server?
No, the software HANA Vora itself is not usually installed on a HANA server called an in-memory database. This is a point that is easy to misunderstand because the word HANA is at the top, and as an easy-to-understand example, HANA Vora can be operated without having a HANA server or DBMS. So, where it will run, it will be run in coexistence with “Apache Spark execution framework” as the first explanation.