Page 2
As another web access mechanism, OAT 3.11 also includes a free smart phone application that provides database status and alert information. Available via the app stores, this app gives mobile DBAs critical information in order to help them make the right decisions to keep the application running smoothly.
Large amounts of Sensor Data? Big Data, No problem…
Based on the application requirements concerning the number of devices, the number of sensors to be employed, and the number of locations where readings will be collected, it is clear that a large amount of sensor data will be collected. Further requirements of making all of this data available, either from a single location or via sharding, for analytic queries indicate that some form of data warehousing and BI software will need to be utilized.
Informix 12.1 raises the performance bar even further by adding new compression functionality for sensor and time series data which will pack more sensor readings into the same physical storage page and in the allocated memory. More data in less space results in better performance, which means fewer servers – and less space, power, and cooling costs.
In addition to the enhancements to the time series feature set, the TimeSeries data type is now better integrated with other Informix features. As an example, Informix 12.1 now supports the processing of time series data within IWA. Sensor data residing within TimeSeries objects on disk can be processed in memory by the accelerator. While the Informix TimeSeries technology is the perfect solution for real-time and near real-time time series processing, the accelerator is able to run extremely complex analytics at memory-speed thereby giving sensor data queries a huge performance boost.
While Informix has supported data replication for many releases, replication of time series data was not fully supported. Informix 12.1 changes this so the replication of our factory application’s sensor data is well supported. Within the application, Informix can be configured to replicate the sensor data to the centralized data mart. Additional replication can be configured for additional secondary data access or for disaster recovery.
Analyzing Big Data – in Real Time
Informix introduced significant data warehousing functionality over the past few years, allowing applications to generate mixed-load, both OLTP and warehousing, utilizing the same Informix environment. These enhancements allow users to design, deploy and maintain a solid data warehousing infrastructure making time-sensitive data quickly available to the various decision makers in the enterprise. Support for traditional data warehousing schemas, joins, analytics, as well as the inclusion of IWA are significant additions that make Informix a powerful option for many more types of applications other than its core OLTP area of strength. IWA maximizes parallelism and scan speed, thus dramatically improving query responses usually more than 100 times, with some queries accelerated over 700 times.
Informix 12.1 includes new techniques for better synchronization of the accelerator’s in-memory snapshot with the underlying data on disk. Keeping the snapshot current greatly improves data availability, making it available for more types of applications. Rather than refreshing all the data on a regular basis, Informix 12.1 supports refresh at the partition level, which could easily correspond to refreshes of individual fact or dimension tables. Partition refreshcan be configured to allowInformix to automatically determine, based on the changes in the various partitions, which partitions to refresh. Informix 12.1 also adds support for a trickle feed update where selected data is buffered and passed to the accelerator in a continuous stream. The ability to continually refresh the in-memory snapshot keeps the snapshot always current and always available, making real-time data warehousing a reality.
The initial version of the accelerator supported a large number of the SQL constructs commonly used for data warehousing. Informix 12.1 extends this support by adding an entire class of ANSI SQL online analytical processing (OLAP) functions which are also accelerated within IWA. These functions will greatly simplify the writing of common business analytical queries such as generating running totals and ranking.
The Informix administration utility, OAT 3.11, is now enhanced to support IWA and includes a simple dashboard for users to administer and monitor warehouses, thus allowing data warehouse administration and general Informix server administration within a single graphical environment.
After some discussion and review of Informix 12.1 functionality, it is clear that Informix meets or exceeds all of the requirements for the factory automation application. It easily handles typical OLTP requirements, as well as the key embedded, SCADA sensor data and the warehousing aspects of this application.
About the author:
Shawn Moe is a software architect at IBM, focusing on UI modernization, SQL compatibility, application development, and tooling support for the Informix database servers. For more information, go to www-01.ibm.com/software/data/informix.