SPP 2037:  Scalable Data Management on Future Hardware

Over the last years, the social and commercial relevance of efficient data management has led to the development of database systems as ubiquitous and complex software systems. Hence there is a wide acceptance of architectural patterns for database systems which are based on assumptions on classic hardware setups.However, the currently used database concepts and systems are not well prepared to support emerging application domains such as eSciences, Industry 4.0, Internet of Things or Digital Humanities: From a user¿s perspective flexible domain-specific query languages or at least access interfaces are requires; novel data models for these application domains have to be integrated; consistency guarantees which reduce flexibility and performance should be adaptable according to the requirements; and the volume and velocity of data caused by ubiquitous sensors have to be mastered by massive scalability and online processing. At the same time current and future hardware trends such as many-core CPUs, co-processors like GPU and FPGA, novel storage technologies like NVRAM and SSD as well as high-speed networks provide new opportunities.In order to open up the exemplarily mentioned application domains together with exploiting the potential of future hardware generations it becomes necessary now, to fundamentally rethink current database architectures. Thus, the objective of the priority program is to answer the scientific questions related to these issues. As a result, we expect the development and evaluation of architectures and abstractions for flexible and scalable data management techniques which provide extensibility regarding new data models including processing and access mechanisms for emerging applications, and exploit the features of modern and heterogeneous hardware as well as system-level services.

DFG Programme Priority Programmes

• ADAMANT: Adaptive Data Management in Evolving Heterogeneous Hardware/Software Systems (Applicants Pionteck, Thilo ;  Saake, Gunter )
• Adaptive Query Compilation for Stream Processing (Applicants Markl, Volker ;  Zeuch, Ph.D., Steffen )
• Coordination Funds (Applicant Sattler, Kai-Uwe )
• Distributed, fault-tolerant in-place consensus sequence on innovative hardware as building block for data management. (Applicant Reinefeld, Alexander )
• Energy-Efficient Event Processing on Modern Hardware (Applicant Seeger, Bernhard )
• Hybrid Transactional/Analytical Graph Processing in Modern Memory Hierarchies (#TAG) (Applicant Sattler, Kai-Uwe )
• Managing Very Large Data Sets on Directly-Attached NVMe Arrays (Applicant Leis, Viktor )
• MxKernel: A Bare-Metal Runtime System for Database Operations on Heterogeneous Many-Core Hardware (Applicants Spinczyk, Olaf ;  Teubner, Jens )
• Query Compilation for the Heterogeneous Many Core Age (Applicant Markl, Volker )
• ReProVide: Query Optimisation and Near-Data Processing on Reconfigurable SoCs for Big Data Analysis (Applicants Lenz, Richard ;  Teich, Jürgen ;  Wildermann, Stefan )
• Scalable Data Management on Next-Generation Networks beyond RDMA (Applicant Binnig, Carsten )
• Scalable hardware-aided trusted data management (Applicants Kapitza, Rüdiger ;  Leich, Thomas )
• Scaling Beyond DRAM with PMem without Compromising Performance (Applicant Kemper, Ph.D., Alfons )

Spokesperson Professor Dr.-Ing. Kai-Uwe Sattler