Programming Fully Disaggregated Systems
Open Access
- 22 June 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 188-195
Abstract
With full resource disaggregation on the horizon, it is unclear what the most suitable programming model is that enables dataflow developers to fully harvest the potential that recent hardware developments offer. In our vision, we propose to raise the abstraction level to allow developers to primarily reason about their dataflow and the requirements that need to be met by the underlying system in a declarative fashion. Underneath, the system works with typed memory regions and uses the notion of ownership that allows for more flexible memory management across the different compute devices and the tasks mapped onto them. This requires a holistic approach that crosses multiple layers of the system stack, opening exciting systems research questions.Keywords
Funding Information
- SPP2037 (Ke 401/22-2)
This publication has 17 references indexed in Scilit:
- Towards a fully disaggregated and programmable data center2022
- Clio: a hardware-software co-designed disaggregated memory system2022
- User-defined operatorsProceedings of the VLDB Endowment, 2022
- Profiling dataflow systems on multiple abstraction levels | Proceedings of the Sixteenth European Conference on Computer Systems2021
- Building blocks for persistent memoryThe VLDB Journal, 2020
- MosaicProceedings of the VLDB Endowment, 2020
- Building the Software-Defined Data CenterProgramming and Computer Software, 2019
- Evaluating end-to-end optimization for data analytics applications in weldProceedings of the VLDB Endowment, 2018
- P4ACM SIGCOMM Computer Communication Review, 2014
- Exascale Computing Trends: Adjusting to the "New Normal"' for Computer ArchitectureComputing in Science & Engineering, 2013