Planned Results

CloudDBAppliance aims at producing a European Cloud Database Appliance for providing a Database as a Service able to match the predictable performance, robustness and trustworthiness of on premise architectures such as those based on mainframes. CloudDBAppliance will evolve cloud architectures to enable the increase of the uptake of cloud technology by providing the robustness, trustworthiness, and performance required for applications currently considered too critical to be deployed on existing clouds.

CloudDBAppliance will deliver a cloud database appliance featuring:

  • A vertically scalable in-memory operational database for time-critical applications able to process high update workloads such as the ones processed by banks or telcos, blended with a fast analytical engine delivering real-time data, able to answer analytical queries over operational data in an online manner
  • An operational Hadoop data lake that enables the use of Hadoop analytics frameworks over operational data
  • A cloud hardware appliance leveraging the next generation of hardware to be produced by Atos, the main European hardware provider. This hardware will be a many-core architecture with 800 cores and 140 TB of main memory. This hardware will go beyond the one of mainframes enabling in-memory processing on big data

Both the operational database and the in-memory analytics engine will be optimized to fully exploit this hardware and deliver predictable performance. CloudDBAppliance will provide high resilience by tolerating catastrophic cloud data centres failures (e.g. a fire or natural disaster) providing data redundancy across appliances located on different cloud data centres.

The main characteristics of the CloudDBAppliance platform are the following ones:

  • New Many-Core with Extra-Large Memory Hardware: The new Bullion that will be built in the context of the project will provide the infrastructure to develop the vision of CloudDBAppliance as a platform that goes beyond mainframes in the cloud with a radical in-memory architecture for real-time big data analytics
  • An ultra-scalable SQL operational database able to scale up linearly to 100s of cores with very low latencies exploiting the features of the new Bullion such as NVDRAM, 3D-point and NVMe
  • A many-core in-memory fast analytics engine scaling up to 100s of cores and answering analytical multidimensional queries (written in the MDX query language) in online times
  • An ultra-scalable in-memory streaming engine for parallelizing analytics algorithms and running over operational data
  • A set of parallel analytics algorithms for data mining and machine learning algorithms running on the streaming engine
  • A real-time big data platform enabling analytical queries over operational data
  • An operational Hadoop data lake supporting Hadoop analytics infrastructure over operational data