Multihop Offloading of Multiple DAG Tasks in Collaborative Edge Computing

Abstract

Collaborative edge computing has become a popular paradigm where edge devices collaborate by sharing resources. Data dissemination is a fundamental problem in CEC to decide what data is transmitted from which device and how. Existing works on data dissemination have not focused on coflow scheduling in CEC, which involves deciding the order of flows within and across coflows at network links. Coflow implies a set of parallel flows with a shared objective. The existing works on coflow scheduling in data centers usually assume a non-blocking switch and do not consider congestion at different links in the multihop path in CEC, leading to increased coflow completion time (CCT). Furthermore, existing works do not consider multiple flow sources that cannot be ignored, as data can have duplicate copies at different edge devices. This work formulates the multisource coflow scheduling problem in CEC, which includes jointly deciding the source and flow ordering for multiple coflows to minimize the sum of CCT. This problem is shown to be NPhard and challenging as each flow can have multiple dependent conflicts at multiple links. We propose a source and coflowaware search and adjust (SCASA) heuristic that first provides an initial solution considering the coflow characteristics. SCASA further improves the initial solution using the source search and adjust heuristic by leveraging the knowledge of both coflows and network congestion at links. Evaluation done using simulation experiments shows that SCASA leads to up to 83% reduction in the sum of CCT compared to benchmarks without a joint solution.

Recent Advances in Collaborative Scheduling of Computing Tasks in an Edge Computing Paradigm

“…Its high-performance solution methods are highly valuable, while its exact global optimal solution cannot be obtained in general for sizable problems. Although there have been many studies on collaborative scheduling of computing tasks in edge computing [47,62,63], the following issues should be addressed:…”

Reliability‐based offloading strategy for deadline‐constrained taskflows in vehicular edge computing environments

“…At each scheduling step, the DLS algorithm selects the next task to be scheduled and the machine to execute it by finding the ready task and machine pair with the highest dynamic level. Based on the Latest‐Finish‐Time scheduling algorithm (BLFT): 39 this algorithm can be regarded as a general name of a class of algorithms, mainly to solve the time‐constrained task flow scheduling problem. It first sets the latest finish time for each task and then selects the resource with the smallest latest finish time for each task as the execution location. Joint dependent task offloading and flow scheduling heuristic algorithm (JDOFH): 20 this algorithm considers the dependencies of network flows in DAG and the start time of network flows, a joint dependency task offloading and flow scheduling heuristic is proposed to minimize the average completion time of a task in a collaborative edge computing scenario.…”

“…• Joint dependent task offloading and flow scheduling heuristic algorithm (JDOFH): 20 this algorithm considers the dependencies of network flows in DAG and the start time of network flows, a joint dependency task offloading and flow scheduling heuristic is proposed to minimize the average completion time of a task in a collaborative edge computing scenario.…”

“…Wang et al 19 proposed an imitation learning online task scheduling algorithm in a VEC environment that minimized the time delay of tasks. Sahni et al 20 studied task flow offloading and scheduling in collaborative edge computing scenarios by considering the dependencies between subtask directed acyclic graphs and the start time of network flows, and proposes a joint dependent task offloading and flow scheduling heuristic algorithm that aims to minimize the average completion time of tasks.…”

A novel multi-application offloading method formobile intelligent vehicles in a cloud-edge hybridenvironment

“…Ying et al [23] considered task scheduling for MEC applications based on directed acyclic graphs(DAG) and proposed a maximum reliability offloading algorithm that decomposes for a given constraint and debits new dynamic adjustments to maximize execution reliability for given energy consumption and delay constraint. Sahni et al [24] proposed a joint dependent task offloading and flow scheduling heuristic algorithm for minimizing task completion time considering the dependencies between tasks and conflicts of network flows. Zhang et al [25] mitigate the concurrent request scheduling problem based on directed acyclic graphs in an online manner.…”