Mμgen 2 has evolved from a virtual data integration platform into a new 'SMART Data Blending' platform and can blend all of the siloed decentralized data within a business, from flat files like Excel and CSV to cloud data, such as Salesforce and SharePoint, to NoSQL DBs.
Mµgen is an information integration and analysis platform for the IoT / Big Data era. With the expansion of IoT datasets having grown exponentially, new-generation business intelligence systems need to support huge datasets across disparate systems with the smallest possible workload on IT divisions. The latest evolution of SMART InSight's platform, Mµgen 2, solves these challenges. Mµgen 2 has evolved from a virtual data integration platform into a new ‘SMART Data Blending' platform. By embedding CData JDBC Drivers, Mµgen can blend all of the siloed decentralized data within a business, from flat files like Excel and CSV to cloud data, such as Salesforce and SharePoint, to NoSQL DBs.
Challenge:
- Integrate siloed business data across RDBMS, CSV, SaaS, and NoSQL DBs
- Enable high performance, ad hoc analysis of big data at terabyte and petabyte scale
Solution:
- Embed CData JDBC Drivers to enable direct access to SaaS, NoSQL, & Big Data sources
- Store abstract table data in the Mµgen Column Data Store to yield high performance gains
CData recently interviewed SMART Insight CEO Kiyoshi Machida:
Q: What trends do you see in corporate data utilization? Machida：Mµgen is a data integration platform used by major manufacturers like Denso and Honda and companies in the entertainment sector such as Badai-Namco. Recently, there is a rapidly growing need for cloud data integration. There is a lot of useful business data stored in Groupware, such as SharePoint and SFA/CRM (Salesforce and Dynamics CRM), and a lot of sensor/log data stored in NoSQL DBs or in CSV on file servers. With Mµgen, siloed application and IoT data can be utilized for analytics and automation on a terabyte-to-petabyte scale.
Q: What is SMART Data Blending?
Machida: SMART Data Blending is a new concept for Mµgen 2. The term Data Blending is not a common term in Japan, but users need to blend data, meaning ad hoc data exploration with cross analysis of various data in order to get insight that can be used to drive business. The main users of Mµgen are the Strategic Planning Divisions and managers unique to Japan who are specially assigned to derive business plans from data analysis. In Japan business plans come from middle management. Research overseas in the USA and in India and the uniqueness of the Japanese market drove SMART Insight's development of the new concept.
Q: Why choose the CData JDBC Drivers?
Machida: There were 2 major reasons: the large variety of data sources and abstract table data modeling. With the increasing number of cloud services in business, it is almost impossible to support all the APIs with in-house development efforts. With JDBC drivers for 80+ data sources, CData provides a standardized way to access all of these APIs from our application. Also, Mµgen deals with terabyte-and-petabyte-size data. Performance matters a lot to the user experience. Because CData drivers model all data from APIs in abstract tables, even unstructured data, we could use the same repository as RDB and CSV to process SaaS and NoSQL data. We realized outstanding high performance: Over 100 million records processed in a fraction of a second.
Q: What is your next move in the cloud data field?
Machida: Using CData drivers, we have saved years of development effort for cloud data source support in Mµgen. The next step is to develop the Mµgen API to enable users to use blended data sets created in Mµgen in other applications. We look forward to working together with CData for API development, too.
