Moonju Park

Non-preemptive Fixed Priority Scheduling of Hard Real-Time Periodic Tasks

This paper addresses the problem of scheduling periodic tasks on a uniprocessor using static priority assignment without preemption. The problem of non-preemptive fixed priority scheduling has received little attention until recently, while real-life applications are often based on non-preemptive systems especially in case of embedded systems. In this paper, we show that Rate Monotonic priority assignment is optimal for non-preemptive scheduling when each tasks relative deadline is equal to its period. We have derived a schedulability bound for non-preemptive Rate Monotonic scheduling by using period ratio of the tasks to provide a guarantee that tasks will meet their deadlines. Since the obtained bound is relatively small comparing with the preemptive scheduling, we also propose a method for designing high-utilization non-preemptive system to enhance the utilization bound.

Quantum-Based Fixed Priority Scheduling

Fixed priority schedulers are widely used for real-time systems, and there were efforts to improve the schedulability. For example, preemption threshold scheduling increases the schedulability by introducing non-preemptiveness during task execution. In this paper, we study another scheduling model that can improve the schedulability of tasks: quantum-based scheduling. When quantum-based scheduling is combined with priority-based scheduling, the quantum introduces a partially non-preemptive region which causes a priority inversion. We show that the non-preemptiveness introduced by quantum-based scheduling can improve the schedulability of fixed priority tasks, and there is a task set which is not schedulable by preemption threshold scheduling but is schedulable by quantum-based scheduling. This paper also presents a method to calculate the worst case response time of tasks in quantum-based scheduling.

An Automated Test Code Generation Method for Web Applications using Activity Oriented Approach

Automated tests are important for Web applications as they grow more complex day by day. Web application testing frameworks have emerged to help satisfy this need. However, used without a model that is designed for system evolution and realization, maintaining test code becomes cumbersome and inefficient. This paper describes an activity oriented approach to engineer automated tests for Web applications with reference to a Web application developed for grant funding agencies. In this approach, the developer defines a domain model to represent application state, and uses a test activity graph comprised of self-validating user interactions to verify application correctness. We have implemented a test code generator called iTester using activity-oriented approach.

An efficient visitation algorithm to improve the detection speed of high-interaction client honeypots

Drive-by-download attacks are client-side attacks that originate from web servers that are visited by web browsers. While many web browsers are vulnerable to the drive-by-download attacks, the cost of detecting malicious web pages that launch drive-by-download attacks is expensive. High-interaction client honeypots are security devices capable of detecting malicious web pages; however, their slow and expensive operations in web page visiting have been considered as a problem. The high-interaction client honeypots employ a visitation algorithm to pinpoint which page has made an unauthorized change of system state when any unauthorized change of the system state occurred after visiting suspicious web pages. To improve the performance of the high-interaction client honeypots, we propose a new visitation algorithm, logarithmic divide-and-conquer (LDAC), for identifying malicious web pages. The LDAC algorithm is an enhanced version of the existing binary divide-and-conquer (BDAC) algorithm. If any system state is abnormally changed after having visited k suspicious web pages concurrently, our LDAC algorithm divides the buffer of k pages into [log2k] pieces and recursively visits the pieces until the malicious page or pages are identified, while the BDAC splits the buffer into k/2 portions. Experimental results show that the LDAC has improved performance of the system up to 15 percent compared to the BDAC algorithm.

Efficient Detection of Malicious Web Pages Using High-Interaction Client Honeypots *

Drive-by-download attacks are client-side attacks that originate from web servers clients visit. High-interaction client honeypots identify malicious web pages by directly visiting the web pages and are very useful. However, they still have shortcomings that must be addressed: long inspection time and possibility of not detecting certain attacks such as time bombs. To address these problems, we propose a new detection method to identify web pages with time bombs. The proposed method introduces a pattern-based static analysis for detecting time bombs efficiently. A high-interaction client honeypot performs the static analysis before carrying out execution-based dynamic analysis. The static analysis classifies sample web pages into two groups, the first one assumed to be time-bombs and the second one assumed to be no time-bombs. We then perform dynamic analysis for the first using sequential visitation algorithm with long classification delay and for the second using divide-and-conquer visitation algorithm with short classification delay. Experimental results demonstrate that our method is more accurate and costs less than conventional methods.

Distance Based Pre-cluster Head Selection Scheme for a Chain-Based Protocol

PEGASIS, a chain-based protocol, forms chains from sensor nodes so that each node transmits and receives from a neighbor. In this way, only one node (known as a HEAD) is selected from that chain to transmit to the sink. Although PEGASIS is able to balance the workload among all of the nodes by selecting the HEAD node in turn, a considerable amount of energy may be wasted when nodes which are far away from sink node act as the HEAD. In this study, DERP (Distance-based Energy-efficient Routing Protocol) is proposed to address this problem. DERP is a chain-based protocol that improves the greedy-algorithm in PEGASIS by taking into account the distance from the HEAD to the sink node. The main idea of DERP is to adopt a pre-HEAD (P-HD) to distribute the energy load evenly among sensor nodes. In addition, to scale DERP to a large network, it can be extended to a multi-hop clustering protocol by selecting a “relay node” according to the distance between the P-HD and SINK. Analysis and simulation studies of DERP show that it consumes up to 80% less energy, and has less of a transmission delay compared to PEGASIS.

An Efficient Test Method for Rate Monotonic Schedulability

Rate Monotonic scheduling algorithm has been widely used in real-time systems for its optimality in fixed priority scheduling. Determining the Rate Monotonic schedulability of tasks is an important problem when designing a real-time system. There are exact schedulability test methods for Rate Monotonic scheduling, but the worst case response time analysis on which the exact tests are based is NP-hard. So the exact tests are often too complex to be executed on-line for large numbers of tasks. For practical use, polynomial time sufficient conditions for the Rate Monotonic schedulability have been studied. However, existing polynomial time tests are often too pessimistic. In this paper, we propose a new polynomial time sufficient condition based on the response time analysis, which is less pessimistic than existing ones. Simulation results show that our test significantly outperforms the existing tests and has performance close to the exact test.

Integration of Preemption Threshold and Quantum-Based Scheduling for Schedulability Enhancement of Fixed Priority Tasks

Fixed priority scheduling is an important real-time scheduling scheme widely used in practice. To improve the schedulability of fixed priority scheduling considerable effort has been made such as introduction of preemption threshold or deferred preemption, and quantum-based scheduling. In this paper, we develop a new scheduling scheme by introducing both preemption threshold and quantum into one scheduling framework. The new scheduling method may successfully schedule tasks which are not schedulable either by preemption threshold scheduling or quantum-based scheduling, as well as tasks schedulable by either scheduling, thus improves the schedulability of fixed-priority tasks. We analyze the schedulability of the new scheduler by computing the worst case response time of tasks. Based on the analysis, we developed an algorithm for assignment of preemption threshold and quantum sizes.

Multi-level QoS Support with Variable Window Size in Weakly Hard Real-Time Systems

Weakly hard real-time model has been successfully used for scheduling in firm real-time systems where some deadline misses are tolerable. The tolerable deadline misses are specified using a (m, k) tuple, which means that tasks are desired to meet m deadlines in any k consecutive task invocations. This model is adopted to provide QoS in many applications such as control systems or multimedia applications. In this paper, we show that multiple QoS levels can be provided using (m, k) constraint with variable window size k. Even though the size of window k is changed at run time, it is shown that the QoS level with minimum ratio of m/k is guaranteed for evenly distributed pattern. Using this guarantee, we show that higher utilization with equal or higher QoS can be achieved in overload management.

Analysis on quantum-based fixed priority scheduling of real-time tasks

Fixed priority schedulers are widely used for real-time systems, and there were efforts to improve the schedulability. Preemption threshold scheduling is one of such efforts with a dual priority scheme. It increases the schedulability by introducing priority inversion during task execution. In this paper, we analyze another scheduling model, quantum-based scheduling. When quantum-based scheduling is combined with priority-based scheduling, the quantum introduces a partially non-preemptive region. We show that the non-preemptiveness introduced by quantum-based scheduling can improve the schedulability of fixed priority tasks, and there is a task set which is not schedulable by preemption threshold scheduling but is schedulable by quantum-based scheduling. This paper presents a method to calculate the worst case response time of tasks in quantum-based scheduling. Based on the previous research results for preemptive and non-preemptive scheduling, we present a priority assignment method for quantum-based scheduling also.