Alongside a number of leading European partners, TUD Dresden University of Technology is launching the research project “CHORYS: Open and Programmable Accelerators for Data-Intensive Applications in the Cloud.” In close collaboration with the University of Copenhagen, Politecnico di Milano and TU Darmstadt, and with the active participation of the companies Menta SAS and Codasip GmbH, the project strives to significantly increase the performance of data-intensive applications in the cloud.

At the core of CHORYS is the development of a new hardware and software architecture that is specifically tailored to the needs of data-intensive applications in the cloud. The goal is to effectively integrate near-data processing technologies and asynchronous data services, which should enhance efficiency while optimizing resource usage.

An innovative aspect of CHORYS is the confluence of RISC-V-based processor extensions and specialized accelerators such as OpenSSD-V and OpenNIC-V. These technologies enable database and analysis processes to run directly at the source of data generation — minimizing the need for time-consuming data transfers across traditional storage and network infrastructures.

Particularly noteworthy is the combination of programmable SSDs, SmartNICs (network interface cards) and data-parallel processing technologies that are integrated within an open architecture. In addition, a powerful library is being developed that allows for efficient execution of data-intensive calculations on various hardware platforms.

The results garnered through the CHORYS project have the potential to make a significant contribution to the continued development of the cloud computing market and to exponentially advance data-intensive business applications, scientific data processing and open-source projects.

© CHORYS

As part of the project, the Database Research Group at TUD is leading the "Data-Intensive Patterns Library and APIs" work package. The goal here is to create a standardized programming interface that seamlessly distributes parts of data-intensive algorithms across multiple computing units, maximizing efficiency and optimizing hardware resource utilization.

Central to the research work is the further development of the SIMD Hardware Abstraction Layer Library TSL, which already enables the cross-platform execution of data-parallel codes on various architectures such as x86, AArch64 and oneAPI FPGAs.

The project is being funded by the European Union with over EUR 4M over a period of four years.

Contact:
TUD Dresden University of Technology
Database Research Group
Prof. Wolfgang Lehner

Tel. +49 351 463 38383