Presentation on available technologies and review of their features/limitation for querying any data storage on top of the current infrastructure or on future upgrades of this architecture.
Project Title: Evaluation of an abstraction layer for next-generation analytic engines and scalable databases
The interest in using Big Data solutions based on the Hadoop ecosystem is constantly growing in HEP community in particular for use cases related to data analytics and data warehousing. At the same time, OLTP databases continue to be critical parts of many systems, notably online acquisition systems that need to store and retrieve metadata and configurations in real-time (eg: Accelerator control). Many of these users currently have a traditional based database (eg: Oracle) for most their goal is to move these parts out to next-generation systems that allow them to scale-out so that they can profit more of their data. The objective is the evaluation of existing abstraction engines that would allow us to provide a long term service while reducing the risk of the very fast changing pace of these technologies.
Justification: Long term support for analytic services and scalable databases: many of the open source projects in the "big data" space change rapidly, we wish to provide an intermediate layer for stability of the developments`