MobiQual: QoS-aware Load Shedding in Mobile CQ Systems Bu˘ gra Gedik #1 , Kun-Lung Wu #2 , Philip S. Yu #3 , Ling Liu *4 # Thomas J. Watson Research Center, IBM Research, 19 Skyline Dr, Hawthorne, NY 10532 1 bgedik@us.ibm.com, 2 klwu@us.ibm.com, 3 psyu@us.ibm.com ∗ College of Computing, Georgia Institute of Technology, 801 Atlantic Drive, Atlanta, GA 30332 4 lingliu@cc.gatech.edu Abstract— Freshness and accuracy are two key measures of quality of service (QoS) in location-based, mobile continual queries (CQs). However, it is often difficult to provide both fresh and accurate CQ results due to (a) limited resources in computing and communication and (b) fast-changing load conditions caused by continuous mobile node movement. Thus a key challenge for a mobile CQ system is: How do we achieve the highest possible quality of the query results, in both freshness and accuracy, with currently available resources under changing load conditions? In this paper, we formulate this problem as a load shedding one, and develop MobiQual - a QoS-aware framework for performing both update load shedding and query load shedding. The design of MobiQual highlights three important features. (1) Differentiated load shedding: Different amounts of query and update load shedding are applied to different groups of queries and mobile nodes, respectively. (2) Per-query QoS specifications: The overall freshness and accuracy of the query results are maximized with individualized QoS specifications. (3) Low-cost adaptation: MobiQual dynamically adapts, with a minimal overhead, to changing load conditions and available resources. We show that, through a careful combination of update and query load shedding, the MobiQual approach leads to much higher freshness and accuracy in the query results in all cases, compared to existing approaches. I. I NTRODUCTION With the ever increasing accessibility of wireless com- munications and the proliferation of mobile devices, we are experiencing a world where we can stay connected while on- the-go. Combined with the availability of low-cost positioning devices (such as GPS sensors), this has created a new class of applications and business opportunities in the area of mobile location-based services (LBSs). Examples include location- aware information delivery and resource management, such as transportation services (NextBus bus locator [1], Google ride finder [2]), fleet management, mobile games, and battlefield coordination. A key challenge for LBSs is a scalable location monitoring system capable of handling large number of mobile nodes and processing complex queries over their positions. Although several mobile continual query (CQ) systems have been pro- posed to handle long-running location monitoring tasks in a scalable manner [3], [4], [5], [6], [7], the focus of these works is primarily on efficient indexing and query processing techniques, not on accuracy or freshness of the query results. Accuracy (inaccuracy) is defined based on the amount of mobile node position errors found in the query results at the time of query re-evaluation. This accuracy measure is strongly tied to the frequency of position updates received from the mobile nodes. Although one can also use a higher level concept to measure accuracy, such as the amount of containment errors found in the query results 1 , including both false positives (inclusion errors) and false negatives (exclusion errors), we argue that using position update errors for accuracy measure will provide higher level of precision. This is primar- ily because by utilizing the amount of node position errors as the accuracy measure, one can easily bound the inaccuracy by a threshold-based position reporting scheme [8], [9]. Freshness (staleness), on the other hand, refers to the age of the query results since the last query re-evaluation. It is dependent on the frequency of query re-evaluations performed at the server. As mobile nodes continue to move, there are further deviations in mobile node positions after the last query re-evaluation. However, such deviations are not attributed to inaccuracy. Hence, freshness can be seen as a metric capturing the post-query-re-evaluation deviations in mobile node posi- tions. It is important to note that higher freshness does not necessarily imply higher accuracy and vice versa. To obtain fresher query results, the CQ server must re- evaluate the continual queries more frequently, requiring more computing resources. Similarly, to attain more accurate query results, the CQ server must receive and process position updates from the mobile nodes in a higher rate, demanding communication as well as computing resources. However, it is almost impossible for a mobile CQ system to achieve 100% fresh and accurate results due to continuously changing positions of mobile nodes. A key challenge therefore is: How do we achieve the highest possible quality of the query results in both freshness and accuracy, in the presence of changing available resources and changing workloads of location up- dates and location queries? In this paper, we propose MobiQual − a resource-adaptive and QoS-aware load shedding framework for mobile CQ systems. MobiQual is capable of providing high-quality query 1 A bound on the amount of containment errors can be approximated by a bound on the position errors, if the distribution of the mobile nodes around the query boundaries is at hand or can be approximated.