Making edge-computing resilient

Harchol, Yotam; Mushtaq, Aisha; Fang, Vivian; McCauley, James; Panda, Aurojit; Shenker, Scott

doi:10.1145/3419111.3421278

Cited by 8 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, Chaterji et al [64] presents the resilience of Cyber Physical System (CPS) and discusses two techniques resilience-by-design and resilienceby-reaction. Harchol et al [65] proposed a framework to improve edge-computing resilience for session-oriented applications. They utilized message replay and checkpoint-based mechanisms to make client-edge-server systems more tolerant to edge failures and client mobility.…”

Section: Related Workmentioning

confidence: 99%

Resilient Edge: Building an Adaptive and Resilient Multi-Communication Network for IoT Edge Using LPWAN and WiFi

Kumar

Yadav

Indrusiak

2023

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

Resilient Edge: Building an Adaptive and Resilient Multi-Communication Network for IoT Edge Using LPWAN and WiFi

Kumar

Yadav

Indrusiak

2023

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

show abstract

“…Fail-over time, however, was measured to be roughly 3 orders of magnitudes higher than normal case latencies. CESSNA [59] presents a solution to the problem of proper checkpointing, request logging, and replay in a distributed edge computing scenario where information about the order and status of requests are dispersed in the edge, as is often the case in IoT systems.…”

Section: Redundancy Mechanismsmentioning

confidence: 99%

A Survey on Resilience in the IoT

et al. 2021

View full text Add to dashboard Cite

Internet-of-Things (IoT) ecosystems tend to grow both in scale and complexity, as they consist of a variety of heterogeneous devices that span over multiple architectural IoT layers (e.g., cloud, edge, sensors). Further, IoT systems increasingly demand the resilient operability of services, as they become part of critical infrastructures. This leads to a broad variety of research works that aim to increase the resilience of these systems. In this article, we create a systematization of knowledge about existing scientific efforts of making IoT systems resilient. In particular, we first discuss the taxonomy and classification of resilience and resilience mechanisms and subsequently survey state-of-the-art resilience mechanisms that have been proposed by research work and are applicable to IoT. As part of the survey, we also discuss questions that focus on the practical aspects of resilience, e.g., which constraints resilience mechanisms impose on developers when designing resilient systems by incorporating a specific mechanism into IoT systems.

show abstract

“…Typically, a cloud node locates at a remote place, which is accessible from a LAN through the Internet. A lower data transmission between the LAN and the cloud server reduces the data transmission over the network core hence mitigating the Internet congestion [39]. Transferring the intermediate results of a DNN to the cloud curtails the data transmission between the LAN and the cloud.…”

Section: B Per-image Communication Overheadmentioning

confidence: 99%

Dynamic DNN Decomposition for Lossless Synergistic Inference

Zhang¹,

Tian²,

Zhang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Deep neural networks (DNNs) sustain high performance in today's data processing applications. DNN inference is resource-intensive thus is difficult to fit into a mobile device. An alternative is to offload the DNN inference to a cloud server. However, such an approach requires heavy raw data transmission between the mobile device and the cloud server, which is not suitable for mission-critical and privacy-sensitive applications such as autopilot. To solve this problem, recent advances unleash DNN services using the edge computing paradigm. The existing approaches split a DNN into two parts and deploy the two partitions to computation nodes at two edge computing tiers. Nonetheless, these methods overlook collaborative deviceedge-cloud computation resources. Besides, previous algorithms demand the whole DNN re-partitioning to adapt to computation resource changes and network dynamics. Moreover, for resourcedemanding convolutional layers, prior works do not give a parallel processing strategy without loss of accuracy at the edge side. To tackle these issues, we propose D 3 , a dynamic DNN decomposition system for synergistic inference without precision loss. The proposed system introduces a heuristic algorithm named horizontal partition algorithm to split a DNN into three parts. The algorithm can partially adjust the partitions at run time according to processing time and network conditions. At the edge side, a vertical separation module separates feature maps into tiles that can be independently run on different edge nodes in parallel. Extensive quantitative evaluation of five popular DNNs illustrates that D 3 outperforms the state-of-the-art counterparts up to 3.4× in end-to-end DNN inference time and reduces backbone network communication overhead up to 3.68×.

show abstract

Making edge-computing resilient

Cited by 8 publications

References 23 publications

Resilient Edge: Building an Adaptive and Resilient Multi-Communication Network for IoT Edge Using LPWAN and WiFi

Resilient Edge: Building an Adaptive and Resilient Multi-Communication Network for IoT Edge Using LPWAN and WiFi

A Survey on Resilience in the IoT

Dynamic DNN Decomposition for Lossless Synergistic Inference

Contact Info

Product

Resources

About