DiPET: Distributed Stream Processing on Fog and Edge Systems via Transprecise Computing

DiPET：通过 Transprecise 计算在雾和边缘系统上进行分布式流处理

• 批准号: EP/T022345/1
• 负责人: Hans Vandierendonck
• 金额: $ 33.27万
• 依托单位: Queen's University Belfast
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2020
• 资助国家: 英国
• 起止时间: 2020 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FT022345%2F1
• 关键词: DiPET; Distributed; Stream; Processing; Fog

The DiPET project investigates models and techniques that enable distributed stream processing applications to seamlessly span and redistribute across fog and edge computing systems. The goal is to utilize devices dispersed through the network that are geographically closer to users to reduce network latency and to increase the available network bandwidth. However, the network that user devices are connected to is dynamic. For example, mobile devices connect to different base stations as they roam, and fog devices may be intermittently unavailable for computing. In order to maximally leverage the heterogeneous compute and network resources present in these dynamic networks, the DiPET project pursues a bold approach based on transprecise computing. Transprecise computing states that computation need not always be exact and proposes a disciplined trade-off of precision against accuracy, which impacts on computational effort, energy efficiency, memory usage and communication bandwidth and latency. Transprecise computing allows to dynamically adapt the precision of computation depending on the context and available resources. This creates new dimensions to the problem of scheduling distributed stream applications in fog and edge computing environments and will lead to schedules with superior performance, energy efficiency and user experience. The DiPET project will demonstrate the feasibility of this unique approach by developing a transprecise stream processing application framework and transprecision-aware middleware. Use cases in video analytics and network intrusion detection will guide the research and underpin technology demonstrators.

DiPET项目研究了分布式流处理应用程序在雾计算和边缘计算系统之间无缝跨越和重新分布的模型和技术。目标是利用分散在网络中的设备，这些设备在地理上更接近用户，以减少网络延迟并增加可用的网络带宽。但是，用户设备所连接的网络是动态的。例如，移动设备在漫游时连接到不同的基站，雾设备可能间歇性地无法进行计算。为了最大限度地利用这些动态网络中的异构计算和网络资源，DiPET项目采用了一种基于透明计算的大胆方法。透明计算表明，计算不需要总是精确的，并提出了精确与准确性之间有纪律的权衡，这对计算工作量、能源效率、内存使用、通信带宽和延迟都有影响。透明计算允许根据上下文和可用资源动态调整计算精度。这为在雾和边缘计算环境中调度分布式流应用程序的问题创造了新的维度，并将导致具有卓越性能、能源效率和用户体验的调度。DiPET项目将通过开发透明流处理应用程序框架和透明感知中间件来演示这种独特方法的可行性。视频分析和网络入侵检测的用例将指导研究和支撑技术演示。

项目成果

SAPCo Sort: optimizing Degree-Ordering for Power-Law Graphs

• DOI: 10.1109/ispass55109.2022.00015
• 发表时间: 2022-05
• 期刊: 2022 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)
• 作者: Mohsen Koohi Esfahani;Peter Kilpatrick;Hans Vandierendonck

Exploiting in-Hub Temporal Locality in SpMV-based Graph Processing

• DOI: 10.1145/3472456.3472462
• 发表时间: 2021-08
• 期刊: Proceedings of the 50th International Conference on Parallel Processing
• 作者: Mohsen Koohi Esfahani;Peter Kilpatrick;Hans Vandierendonck

The Effects of Weight Quantization on Online Federated Learning for the IoT: A Case Study

• DOI: 10.1109/access.2024.3349557
• 发表时间: 2024
• 期刊: IEEE Access
• 影响因子: 3.9
• 作者: Nil Llisterri Giménez;JunKyu Lee;Felix Freitag;Hans Vandierendonck

Resource-Efficient Deep Learning: A Survey on Model-, Arithmetic-, and Implementation-Level Techniques

资源高效的深度学习：模型、算术和实现级技术的调查

• DOI: 10.48550/arxiv.2112.15131
• 发表时间: 2021
• 作者: Lee J

Locality Analysis of Graph Reordering Algorithms

• DOI: 10.1109/iiswc53511.2021.00020
• 发表时间: 2021-11
• 期刊: 2021 IEEE International Symposium on Workload Characterization (IISWC)
• 作者: Mohsen Koohi Esfahani;Peter Kilpatrick;Hans Vandierendonck