Future-Proof Data Systems in the Post-Moore Era
FDS aims to revolutionize software infrastructure for data-intensive jobs by creating efficient system abstractions and cost models to optimize performance on modern specialized hardware.
Projectdetails
Introduction
Three major trends put immense pressure on existing infrastructure. First, there is an increased demand for compute given the exponential growth of data volume we are collecting. Second, as most data- and compute-intensive systems move to the cloud, it is vital to make data centers more efficient. Third, pressured by the waning laws of computing infrastructure (end of Dennard scaling and its impact on Moore's law), the hardware landscape has embarked on a major push towards specialization. Unsurprisingly, all of these have major implications on software development.
Proposed Solution
FDS addresses these challenges in a holistic fashion by establishing the foundations of a new software infrastructure. More specifically, I propose:
- New system software abstractions that enable more efficient execution of data-intensive jobs on modern hardware.
- New cost models to support reasoning about the performance/resource efficiency trade-offs when scheduling the said workloads.
Future-proof Data Systems
To enable the development of future-proof data systems, I propose replacing the traditional monoliths of a logical-graph optimizer and execution engine with:
- A new flexible framework for optimization and compilation.
- A runtime system that can exploit the features of modern hardware but remains amenable to its evolving nature.
Doing so requires innovation on many levels and combines work from many layers of the system stack, including:
- HW/SW co-design
- Compilers
- Operating and runtime systems
- Databases
- Distributed systems
Conclusion
It is a non-trivial undertaking, but it is a necessary one. And it will have a big and lasting impact on academia, industry, and society. In fact, I strongly believe that the only way forward towards reducing the increasing gap between data- and compute-demand in resource-hungry data centers is a software solution that boosts efficiency, embraces specialization, removes redundancy, and optimizes for data movement. FDS will be the pioneering system leading the way.
Financiële details & Tijdlijn
Financiële details
Subsidiebedrag | € 1.496.305 |
Totale projectbegroting | € 1.496.305 |
Tijdlijn
Startdatum | 1-1-2025 |
Einddatum | 31-12-2029 |
Subsidiejaar | 2025 |
Partners & Locaties
Projectpartners
- TECHNISCHE UNIVERSITAET MUENCHENpenvoerder
Land(en)
Vergelijkbare projecten binnen European Research Council
Project | Regeling | Bedrag | Jaar | Actie |
---|---|---|---|---|
Commoditizing Data Analytics in the CloudDevelop an open, cost-efficient cloud analytics database, CODAC, to reduce query costs and enable seamless data migration across platforms, enhancing large-scale data analytics accessibility. | ERC Starting... | € 1.498.125 | 2023 | Details |
Federated and distributed inference leveraging sensing and communication in the computing continuumThis project aims to develop a framework for federated and distributed inference systems that optimizes sensing data processing across edge and cloud environments, enhancing efficiency, security, and performance. | ERC Starting... | € 1.019.000 | 2023 | Details |
Computational Discovery of Numerical Algorithms for Animation and Simulation of Natural PhenomenaThe project aims to revolutionize numerical simulation and animation by integrating analytical tools, data-driven insights, and optimization techniques to efficiently model complex physical systems. | ERC Consolid... | € 1.936.503 | 2022 | Details |
Decentralized Cryptographic SystemsThis project aims to develop robust cryptographic systems that align theoretical models with real-world challenges, enhancing security and efficiency for decentralized infrastructures. | ERC Consolid... | € 1.998.351 | 2024 | Details |
Proof of Concept Prototype for Matching Applications on BELFORT HardwareThe project aims to enhance the efficiency of fully homomorphic encryption (FHE) for real-time operations on encrypted data in cloud settings, enabling seamless privacy-preserving queries. | ERC Proof of... | € 150.000 | 2024 | Details |
Commoditizing Data Analytics in the Cloud
Develop an open, cost-efficient cloud analytics database, CODAC, to reduce query costs and enable seamless data migration across platforms, enhancing large-scale data analytics accessibility.
Federated and distributed inference leveraging sensing and communication in the computing continuum
This project aims to develop a framework for federated and distributed inference systems that optimizes sensing data processing across edge and cloud environments, enhancing efficiency, security, and performance.
Computational Discovery of Numerical Algorithms for Animation and Simulation of Natural Phenomena
The project aims to revolutionize numerical simulation and animation by integrating analytical tools, data-driven insights, and optimization techniques to efficiently model complex physical systems.
Decentralized Cryptographic Systems
This project aims to develop robust cryptographic systems that align theoretical models with real-world challenges, enhancing security and efficiency for decentralized infrastructures.
Proof of Concept Prototype for Matching Applications on BELFORT Hardware
The project aims to enhance the efficiency of fully homomorphic encryption (FHE) for real-time operations on encrypted data in cloud settings, enabling seamless privacy-preserving queries.
Vergelijkbare projecten uit andere regelingen
Project | Regeling | Bedrag | Jaar | Actie |
---|---|---|---|---|
Simulatie optimalisatie door middel van een data-driven preconditionerDit project onderzoekt de haalbaarheid van een data-driven preconditioner om complexe simulaties te optimaliseren en kosten te verlagen. | Mkb-innovati... | € 20.000 | 2021 | Details |
Digital optical computing platform for neural networksDOLORES aims to develop a digital optical neural network processor to overcome current optical computing limitations, revolutionizing AI and deep learning applications across various sectors. | EIC Pathfinder | € 3.015.883 | 2024 | Details |
method for smart and affordaBle Evaluation of simultaneous faults in heating and cooling sYstems based ON compresseD vapor technologyThis project aims to develop a cost-effective, knowledge-based method for comprehensive fault detection and diagnosis in small vapor compression systems, enhancing efficiency and environmental performance. | EIC Pathfinder | € 1.181.750 | 2024 | Details |
Scalable Hardware for Large-Scale Quantum ComputingDeveloping a scalable, fault-tolerant quantum computer using advanced cryo-CMOS technology to enhance precision and efficiency in processing complex data across various fields. | EIC Transition | € 2.499.998 | 2023 | Details |
Smart Development & Maintenance Vertical Integrated PlatformDit project ontwikkelt een innovatief softwareplatform voor virtuele machine- en productieprocesontwikkeling, waardoor kosten en tijd voor fysieke prototypes aanzienlijk worden verminderd. | Mkb-innovati... | € 350.000 | 2017 | Details |
Simulatie optimalisatie door middel van een data-driven preconditioner
Dit project onderzoekt de haalbaarheid van een data-driven preconditioner om complexe simulaties te optimaliseren en kosten te verlagen.
Digital optical computing platform for neural networks
DOLORES aims to develop a digital optical neural network processor to overcome current optical computing limitations, revolutionizing AI and deep learning applications across various sectors.
method for smart and affordaBle Evaluation of simultaneous faults in heating and cooling sYstems based ON compresseD vapor technology
This project aims to develop a cost-effective, knowledge-based method for comprehensive fault detection and diagnosis in small vapor compression systems, enhancing efficiency and environmental performance.
Scalable Hardware for Large-Scale Quantum Computing
Developing a scalable, fault-tolerant quantum computer using advanced cryo-CMOS technology to enhance precision and efficiency in processing complex data across various fields.
Smart Development & Maintenance Vertical Integrated Platform
Dit project ontwikkelt een innovatief softwareplatform voor virtuele machine- en productieprocesontwikkeling, waardoor kosten en tijd voor fysieke prototypes aanzienlijk worden verminderd.