Power-Aware Allocation of Graph Jobs in Geo-Distributed Cloud Networks

TL;DR

This paper presents a comprehensive framework for power-aware graph job allocation in geo-distributed cloud networks, addressing NP-hard challenges with scalable solutions for various network sizes and incorporating online learning for strategy optimization.

Contribution

It introduces scalable algorithms for graph job allocation considering power consumption, including convex programming, distributed methods, and cloud crawlers for large-scale networks.

Findings

Efficient solutions for small, medium, and large-scale GDCNs.

Effective power-aware allocation strategies under different pricing schemes.

Successful application of online learning for strategy adaptation.

Abstract

In the era of big-data, the jobs submitted to the clouds exhibit complicated structures represented by graphs, where the nodes denote the sub-tasks each of which can be accommodated at a slot in a server, while the edges indicate the communication constraints among the sub-tasks. We develop a framework for efficient allocation of graph jobs in geo-distributed cloud networks (GDCNs), explicitly considering the power consumption of the datacenters (DCs). We address the following two challenges arising in graph job allocation: i) the allocation problem belongs to NP-hard nonlinear integer programming; ii) the allocation requires solving the NP-complete sub-graph isomorphism problem, which is particularly cumbersome in large-scale GDCNs. We develop a suite of efficient solutions for GDCNs of various scales. For small-scale GDCNs, we propose an analytical approach based on convex programming. For medium-scale GDCNs, we develop a distributed allocation algorithm exploiting the processing power of DCs in parallel. Afterward, we provide a novel low-complexity (decentralized) sub-graph extraction method, based on which we introduce cloud crawlers aiming to extract allocations of good potentials for large-scale GDCNs. Given these suggested strategies, we further investigate strategy selection under both fixed and adaptive DC pricing schemes, and propose an online learning algorithm for each.