Ehsan Ullah Munir

SDBATS: A Novel Algorithm for Task Scheduling in Heterogeneous Computing Systems

A heterogeneous computing system (HCS) efficiently utilizes the heterogeneity of diverse computational resources interconnected with high speed networks to execute a group of compute intensive tasks. These are typically represented by means of a directed acyclic graph (DAG) with varied computational requirements and constraints. The optimal scheduling of the given set of precedence-constrained tasks to available resources is a core concern in HCS and is known to be NP-complete problem. Task prioritization has been a major criterion for achieving high performance in HCS. This paper presents a SD-Based Algorithm for Task Scheduling (SDBATS) which uses the standard deviation of the expected execution time of a given task on the available resources in the heterogeneous computing environment as a key attribute for assigning task priority. This new approach takes into account the task heterogeneity and achieves a significant reduction in the overall execution time of a given application. The performance of the proposed algorithm has been extensively studied under a variety of conditions on standard task graphs from Graph Partition Archive as well as on some real world application DAGs such as Gaussian Elimination and Fast Fourier Transformation application DAGs. Our results show that SDBATS outperforms well known existing DAG scheduling algorithms in terms of schedule length (make span) and speedup.

PEGA: A Performance Effective Genetic Algorithm for Task Scheduling in Heterogeneous Systems

Task scheduling has vital importance in heterogeneous systems because efficient task scheduling can enhance overall system performance considerably. This paper addresses the task scheduling problem by effective utilization of evolution based algorithm. Genetic algorithms are promising to provide near optimal results even in the large problem space but at the same time the time complexity of Genetic Algorithms are higher. The proposed algorithm, Performance Effective Genetic Algorithm (PEGA) not only provides near optimal schedule but also has a low time complexity. The PEGA efficiently finds the best solution from the search space; PEGA is performance effective due to effective utilization of genetic operators (crossover and mutation) through rigorous search. In addition the chromosome encoding with b-level introduces simplicity with efficiency. The performance is compared through extensive simulations with standard genetic algorithm (SGA). The comparison of results proved that the PEGA outperforms SGA in providing near optimal schedules with considerable less run time.

Data-Intensive Workflow Optimization Based on Application Task Graph Partitioning in Heterogeneous Computing Systems

Stream based data processing model is proven to be an established method to optimize data-intensive applications. Data-intensive applications involve movement of huge amount of data between execution nodes that incurs large costs. Data-streaming model improves the execution performance of such applications. In the stream-based data processing model, performance is usually measured by throughput and latency. Optimization of these performance metrics in heterogeneous computing environment becomes more challenging due to the difference in the computing capacity of execution nodes and variations in the data transfer capability of communication links between these nodes. This paper presents a dual objective Partitioning based Data-intensive Workflow optimization Algorithm (PDWA) for heterogeneous computing systems. The proposed PDWA provides significantly reduced latency with increase in the throughput. In the proposed algorithm, the application task graph is partitioned such that the interpartition data movement is minimal. Such optimized partitioning enhances the throughput. Each partition is mapped to the execution node that gives minimum execution time for that particular partition. PDWA also exploits partial task duplication to reduce the latency. We evaluated the proposed algorithm with synthesized benchmarks and workflows from the real-world workloads, and the proposed algorithm shows 60% reduced latency with 47% improvement in the throughput as compared to the approach when workflows are not partitioned.

HADCC: Hybrid Advanced Distributed and Centralized Clustering Path Planning Algorithm for WSNs

Designing and development of energy effective path planning algorithm is very key research domain in order to tackle the issues of limited life-time for Wireless Sensor Networks (WSNs). So in WSNs, energy efficiency is major concern of researchers. Overall advancement in routing protocols prove that clustering is much better approach as compared to flat and location-based energy efficient routing protocols. Due to better performance, multiple energy efficient clustering routing protocols have been proposed. But existing clustering algorithms are centralized or distributed, which are not intelligent enough and do not produce hybrid cluster-head selection. In this paper, we propose a cluster structured path planning algorithm named, Hybrid Advance Distributed Centralized Clustering (HADCC) path planning energy efficient algorithm. HADCC proposed model is fascinated with hybrid cluster head selection algorithm. This hybrid algorithm makes decision of cluster head selection of nodes. In order to execute proposed model we have also proposed an advance network topology, in which the whole network region is divided into two physical levels. First physical level consists of a circular region, containing homogeneous normal nodes and all important Base Station. While, second physical level is outer region of the circle that contains advanced heterogeneous nodes. Simulation results indicate that HADCC prolongs the network lifetime as compared to existing advanced clustering routing protocols for both homogeneous and heterogeneous WSNs. HADCC outperforms in case of stability and network life time as compared to the existing clustering algorithms.

A Dual Ring Connectivity Model for VANET Under Heterogeneous Traffic Flow

Vehicular ad-hoc network (VANET) is characterized as a highly dynamic wireless network due to the dynamic connectivity of the network nodes. To achieve better connectivity under such dynamic conditions, an optimal transmission strategy is required to direct the information flow between the nodes. Earlier studies on VANET’s overlook the characteristics of heterogeneity in vehicle types, traffic structure, flow for density estimation, and connectivity observation. In this paper, we have proposed a heterogeneous traffic flow based dual ring connectivity model to enhance both the message disseminations and network connectivity. In our proposed model the availability of different types of vehicles on the road, such as, cars, buses, etc., are introduced in an attempt to propose a new communication structure for moving vehicles in VANETl under cooperative transmission in heterogeneous traffic flow. The model is based on the dual-ring structure that forms the primary and secondary rings of vehicular communication. During message disseminations, Slow speed vehicles (buses) on the secondary ring provide a backup path of communication for high speed vehicles (cars) moving on the primary ring. The Slow speed vehicles act as the intermediate nodes in the aforementioned connectivity model that helps improve the network coverage and end-to-end data delivery. For the evaluation and the implementation of dual-ring model a clustering routing scheme warning energy aware cluster-head is adopted that also caters for the energy optimization. The implemented dual-ring message delivery scheme under the cluster-head based routing technique does show improved network coverage and connectivity dynamics even under the multi-hop communication system.

A Reversible Watermarking Technique for Social Network Data Sets for Enabling Data Trust in Cyber, Physical, and Social Computing

Social network data are being mined for extracting interesting patterns. Such data are collected by different researchers and organizations and are usually also shared via different channels. These data usually have huge volume because there are millions of social network users throughout the world. In this context, ownership protection of such data sets with huge volume becomes relevant. Digital watermarking is a more demanding solution than any other technique for ensuring rights protection and integrity of the original data sets. The objective of this paper is to devise a reversible watermarking technique for the social network data to prove ownership rights and also provide a mechanism for data recovery. Robustness of the proposed technique is evaluated through attack analysis using experimental study. In this paper, Z notation-based formal specification is also provided to show the working of the proposed reversible watermarking technique for social network data sets for enabling data trust in Cyber, Physical, and Social Computing (CPSCom).

HETS: Heterogeneous Edge and Task Scheduling Algorithm for Heterogeneous Computing Systems

Widely used computing systems are heterogeneous in nature, comprising of interconnected resources which differ in computational capability of processing nodes and network bandwidth. Due to this diversity, an efficient heuristic is required to achieve high performance in heterogeneous computing system. In our proposed scheduling algorithm, Heterogeneous Edge and Task Scheduling (HETS), we schedule the communication between the tasks of application graph onto the network links of varying bandwidth, and schedule these tasks of different computation on the network processors after considering the computational capability of the available processors. In HETS, the prioritization is done by calculating the edge priority as well as the node priority. HETS algorithm selects the task after all its incoming edges are scheduled. The proposed algorithm minimizes the communication overhead of the application graph edges and obtains reduced schedule length in terms of the over allexecution time. Performance of the proposed algorithm is studied by varying parameters of the standard task graphs as well as on real world directed acyclic graphs (DAGs) application, such as Cybershake, Gaussian Elimination, and Montage. Extensive simulation results show the effectiveness of HETS algorithm interms of reduced makespan and improved Schedule Length Ratio(SLR) for the given tasks.

A Segmented Approach for DAG Scheduling in Heterogeneous Environment

Task scheduling optimization is crucial in order to achieve maximum advantage out of available resources having diverse characteristics. In heterogeneous environment scheduling set of dependent tasks involve two dimensional considerations. Tasks are supposed to be assigned to best suited machines while avoiding the extra overhead of communication cost which should ultimately enhance the performance mostly in terms of minimizing the completion time of a job. Extensive research work has been done addressing the same problem domain and number of well-known heuristics has been proposed. In this paper a new heuristic is proposed which assign priorities to the set of dependent tasks based on three different parameters which are average computation cost, average communication cost and mean of both. A segmented approach is introduced which schedules tasks based on nature set of tasks in terms of computation cost and there precedence constraints. The experimental results show the better performance of proposed heuristic.

DS-Index: Ranking Authors Distinctively in an Academic Network

The impact and productivity of researchers are assessed using bibliometric parameters, such as the number of publications and citation analysis. A number of indices exist that use these parameters, but almost all of them overlook citation pattern of the researchers, which results in assigning the same index value to the two different authors with different citation patterns. In this paper, a new index called DS-index is proposed, which differentiates among the authors having even a very small change in the citation pattern of their publications. It uniquely identifies the different index values and thus the proper ranking order for authors. The index is applied to the self-developed large DBLP data set having publication data of over 50 years. The results compared with the existing indices using the standard performance evaluation measures confirm that the proposed index performs better by ranking the authors in a distinctive order.

List-Based Task Scheduling for Cloud Computing

Cloud computing model provides global and ondemand access to resources in a seamless manner with minimal interaction with the service provider. A typical cloud data center consists of several computational resources interconnected with each other through high-speed networks. In cloud the program execution can be visualized as a collection of multiple tasks represented by Directed Acyclic Graph (DAG) that execute in their logical sequence. Prioritization of these tasks plays an important role to achieve high performance and improved efficiency in a cloud environment. In this paper, we propose a novel task scheduling algorithm named Median Deviation based Task Scheduling (MDTS), which uses Median Absolute Deviation (MAD) of the Expected Time to Compute (ETC) of a task as a major attribute to calculate ranks of the given tasks. We use coefficient-of-variation (COV) based technique that considers task and machine heterogeneity to estimate the ETC of a particular DAG. The proposed algorithm is evaluated under various conditions using synthetic DAGs and real world applications. Our evaluation shows that the proposed MDTS algorithm produces high quality schedules and significantly reduces the makespan of an application by up to 25.01%.