Adaptive per-user per-object cache consistency management for mobile data access in wireless mesh networks

Abstract

We propose and analyze an adaptive per-user per-object cache consistency management (APPCCM) scheme for mobile data access in wireless mesh networks. APPCCM supports strong data consistency semantics through integrated cache consistency and mobility management. The objective of APPCCM is to minimize the overall network cost incurred due to data query/update processing, cache consistency management, and mobility management. In APPCCM, data objects can be adaptively cached at the mesh clients directly or at mesh routers dynamically selected by APPCCM. APPCCM is adaptive, per-user and per-object as the decision regarding where to cache a data object accessed by a mesh client is made dynamically, depending on the mesh client’s mobility and data query/update characteristics, and the network’s conditions. We develop analytical models for evaluating the performance of APPCCM and devise a computational procedure for dynamically calculating the overall network cost incurred. We demonstrate via both model-based analysis and simulation validation that APPCCM outperforms non-adaptive cache consistency management schemes that always cache data objects at the mesh client, or at the mesh client’s current serving mesh router for mobile data access in wireless mesh networks.

Introduction

Wireless mesh networks (WMNs) are emerging in recent years as a promising standard for next-generation broadband wireless networks and are regarded as a cost-effective solution for providing last-mile broadband wireless Internet connectivity [19]. A WMN consists of two types of components: wireless mesh routers (MRs) and mesh clients (MCs) [1]. Each MR serves as both a router that forwards packets and a wireless access point for MCs. The group of MRs within a WMN forms a wireless mesh backbone that provides last-mile broadband Internet connectivity to MCs. A WMN is interconnected to the Internet through the gateway functionality of MRs. MCs are devices that have wireless access capability, e.g., laptops, smartphones, PDAs, etc.

In this paper, we investigate the problem of data caching and cache consistency management for mobile data access in WMNs. Mobile data access, which is fundamental to client–server computing in mobile environments [11], is challenging due to the intrinsic characteristics of mobile computing: mobility and resource constraints of clients, and bandwidth constraints of wireless communications [9], [20]. An additional challenge as a result of these characteristics is that mobile data access must cope with voluntary or involuntary disconnection of mobile clients. That is, mobile data access must support disconnected operations [13]. Caching is a key technique for improving the performance of mobile data access because it alleviates constraints such as highly variable mobile connectivity and wireless bandwidth limitations, and it significantly reduces the latency for answering a query. Caching is also the basis for disconnected operations in mobile data access.

Data caching is particularly beneficial for mobile Internet data access in WMNs for a major reason. Specifically, because WMNs are a cost-effective solution for last-mile broadband Internet access, Internet traffic contributed considerably by mobile Internet data access is expected to dominate network traffic in WMNs. Because Internet traffic always passes through the gateway, the gateway is potentially the bottleneck under conditions of heavy Internet traffic and network congestion is highly probably to happen around the gateway. Caching is an efficient solution for mitigating the problem because it can significantly reduce the number of uplink and downlink messages passing through the gateway in client–server mobile Internet data access, thereby mitigating the performance bottleneck at the gateways in WMNs [6].

A central issue closely related to caching in mobile data access is cache consistency management because clients may not be able to keep cached data synchronized due to mobility and disconnection. We propose and analyze an adaptive per-user per-object cache consistency management (APPCCM) scheme for mobile data access in WMNs. APPCCM supports strong data consistency semantics through integrated cache consistency and mobility management. The objective of APPCCM is to minimize the overall network cost incurred collectively by data query/update processing, cache consistency management, and mobility management. APPCCM provides two data access and caching modes: a data object can be cached either directly at the MC, or at a data proxy running on an MR dynamically selected by APPCCM. We use the terms client-cache mode (CCM) and data-proxy mode (DPM) to refer to these two modes throughout the paper. CCM is based on existing asynchronous stateful-based cache invalidation schemes [2], [21] augmented with the capability of integrated cache consistency and mobility management such that MCs can perform data access, data caching and cache consistency management while roaming in a WMN. DPM is distinct from traditional approaches by exploiting the capability of increasingly powerful MRs to perform data caching in addition to routing. The rationale of using data proxies is that it is beneficial to cache a data object at the data proxy rather than at the MC under certain operational and environmental conditions which we aim to identify in the paper.

APPCCM is adaptive, per-user and per-object because for each individual MC, the decision of where to cache data objects is made dynamically and independently for each object based on the MC’s mobility and data query/update characteristics, and the WMN conditions. We develop a computational procedure for dynamically calculating the overall communication cost in APPCCM, given parameters characterizing the MC’s mobility and data query/update characteristics, and the WMN conditions. For DPM, APPCCM determines when a cached data object should be migrated between two data proxies due to MC mobility. We use a threshold on the number of location changes of the MC denoted by $K_{\mathrm{optimal}}$ to model the optimal time point at which the data object should be migrated such that the overall communication cost is minimized. When the threshold is reached, the data object will be migrated to the data proxy on the MC’s current serving MR.

We develop analytical models for evaluating the performance of APPCCM and devise a computational procedure for dynamically calculating the overall network cost incurred. We demonstrate via both model-based analysis and simulation validation that APPCCM outperforms non-adaptive cache consistency management schemes that always cache data objects at the mesh client, or at the mesh client’s current serving mesh router for mobile data access in wireless mesh networks.

The rest of the paper is organized as follows. Section 2 surveys existing work and contrasts our scheme with existing schemes. Section 3 presents the system model and assumptions made in the paper. The proposed APPCCM scheme is presented in Section 4. In Section 5, we develop mathematical models for evaluating APPCCM. Performance analysis and comparison are carried out in Section 6. The paper concludes with Section 7.