Ultra-Reliable Distributed Cloud Network Control with End-to-End Latency Constraints

Abstract

We are entering a rapidly unfolding future driven by the delivery of real-time computation services, such as industrial automation and augmented reality, collectively referred to as augmented information (AgI) services, over highly distributed cloud/edge computing networks. The interaction intensive nature of AgI services is accelerating the need for networking solutions that provide strict latency guarantees. In contrast to most existing studies that can only characterize average delay performance, we focus on the critical goal of delivering AgI services ahead of corresponding deadlines on a per-packet basis, while minimizing overall cloud network operational cost. To this end, we design a novel queuing system able to track data packets' lifetime and formalize the delay-constrained least-cost dynamic network control problem. To address this challenging problem, we first study the setting with average capacity (or resource budget) constraints, for which we characterize the delay-constrained stability region and design a near-optimal control policy leveraging Lyapunov optimization theory on an equivalent virtual network. Guided by the same principle, we tackle the peak capacity constrained scenario by developing the reliable cloud network control (RCNC) algorithm, which employs a two-way optimization method to make actual and virtual network flow solutions converge in an iterative manner. Extensive numerical results show the superior performance of the proposed control policy compared with the state-of-the-art cloud network control algorithm, and the value of guaranteeing strict end-to-end deadlines for the delivery of next-generation AgI services.

Optimal Cloud Network Control with Strict Latency Constraints

“…Due to space limitations, in this paper we describe the proposed approach for the packet routing problem and refer the reader to the longer version in [2] for its cloud network control generalization. 1 Nonetheless, we do use the general cloud network control model for the numerical results in Section V.…”

“…We assume that the input packets of the given commodity 1 The generalization is based on the layered-graph technique [13]. 2 As an extension to this work, the problem formulated under a more realistic scenario with peak link capacity constraint is also studied in [2]. The solution closely follows the principle of the methodology presented in this paper.…”

“…We emphasize that the above problem cannot be addressed by the LDP approach, because (i) the queuing system (9e) allows packet drops by (2), and (ii) we employ the actual flow as the decision variable, i.e., constraint (9d), which are different from the standard formulation.…”

“…From the consumers' perspective, excessive end-to-end delays can significantly impact quality of experience (QoE), especially for delay-sensitive AgI applications where packets must be delivered by a strict deadline in order to be effective. In this context, An extended version of this paper is submitted to the IEEE Transactions on Communications [2].…”

“…In [19], the authors formulate the problem of timely throughput maximization as a constrained Markov decision process (CMDP), and address it by solving the single-packet routing problem separately for each packet; yet, its computational complexity is prohibitive for practical implementation. A more comprehensive literature review is presented in [2].…”