In recent years, Network on Chip (NoC) is used to handle the communication issues between cores or processing elements. The major drawback of existing XY routing algorithm for mesh topology is its inability to handle a high traffic load. In this paper, an E-XY (Entropy based XY) routing algorithm is proposed to generate information about the adjacent router based on previously communicated packets. Tests have been carried out on a mesh topology of 8x8 simulated using Omnet++ 4.4.1 using HNOCS version. Different types of traffic have been considered for result computation including uniform, bit complement, neighbour and tornado. The proposed algorithm is compared with other routing algorithms including XY, IX/Y and Odd Even. Results demonstrate that the proposed algorithm is comparable to XY routing algorithm up to the load factor of 0.8 and performs well than the XY, IX/Y and Odd Even routing algorithms with the increase in load.

Distributed systems are efficient means of realizing High-Performance Computing (HPC). They are used in meeting the demand of executing large-scale high-performance computational jobs. Scheduling the tasks on such computational resources is one of the prime concerns in the heterogeneous distributed systems. Scheduling jobs on such systems are NP-complete in nature. Scheduling requires either heuristic or metaheuristic approach for sub-optimal but acceptable solutions. An application can be divided into a number of tasks which can be represented as Direct Acyclic Graph (DAG). To accomplish high performance, it is important to efficiently schedule these dependent tasks on resources with the satisfaction of the constraints defined for schedule generation. Inspired by Quantum computing, this work proposes a Quantum Genetic Algorithm with Rotation Angle Refinement (QGARAR) for optimum schedule generation. In this paper, the proposed QGARAR is compared with its peers under various test conditions to account for minimization of the makespan value of dependent jobs submitted for execution on heterogeneous distributed systems.

Increasing use of IP enabled smart gadgets, rapidly increasing the formation of cyber centric social networks. Any human discussion reflects its emotions and ego centric thoughts among the masses which results the formation of pro and anti-groups. Identification of various categories of groups and communities in social network is very important to eliminate the chances of human created crisis. This paper predicts the migration of individuals from one community to other community and the person who bridges the two communities. The prediction of social networks is carried out by mapping various epidemic models on human created social network. The centrality measurement detects the bridging element between two communities.

Many variants of the mesh interconnection networks have been suggested in past by the placement of the extra links to the existing mesh topology. In this paper, we have searched for the optimal links that should place in the topology so as to reduce the average inter node distance of the resultant topology. This is a heuristic search problem which has the single objective that is to minimize the average inter node distance of the topology. The Improved Environmental Adaptation Method (IEAM) is one of the latest techniques based on phenotypic properties of genes, which prove to be fast in searching the optimal solution to the problem. The paper describes in detail the various operator of IEAM and compares its performance with the existing heuristic technique that is Genetic algorithm. The IEAM obviously proves to be better in finding the results. After finding the optimized topology the resultant is tested for the performance with existing variant of the topology like mesh, center concentrated mesh, T Mesh for performance and proves to give the improvement more than 2 times.

The mesh interconnection network is the most popular topology for the network on chip architecture. The horizontal fat mesh topology is designed to keep all the advantages of the mesh topologies prime of which is the almost wire area and the simplicity of the topology. The designed topology also does not have the non-planar links as present in most of the topologies which requires multiple layers. The performance of the topology has been tested on the uniform traffic for the latency, throughput and the power of the network. The topology has shown the improvement of in latency, throughput and power of network.

Objectives: The high computing demands leads to high communication between the various cores on the chips. This leads to the exploration of various topologies. The Mesh based topologies are widely accepted due to simplicity. This mesh ­topology is modified with objective to reduce inter node distance. Methods/Statistical Analysis: Proposed topology has been generated as the union of the Diagonal connected mesh and T Mesh. To test the performance the proposed topology that is Diagonal connected T mesh is compared with diagonal connected mesh and T Mesh. To test the performance of the topology static routing algorithm is used. The various traffic patterns have been used for the analysis of latency and bandwidth. Findings: The proposed topology has performed better in comparison to Diagonal connected Mesh and T Mesh in case of Uniform traffic and Bit Complement traffic. In case of tornado traffic results are identical to that of Diagonal connected mesh. The hop count analysis shows that there is always a positive improvement. Average Hop count of the diagonal Connected T mesh is 87% less than that of it counterpart. This makes the proposed proved to be better than the existing topologies. Application/Improvements: The proposed topology for network on chip will always be suitable for the applications in which the communication is either uniform among the nodes or is communicating to the nodes having complement traffic match. Keywords: Hop Count, Mesh, Network on Chip, Topology, Traffic Analysis

Content Delivery Network is a large distributed system consisting of multiple servers deployed in much geographical location that delivers the content to end user with high performance and high availability. But the increase load of content delivery server and network degrades the quality of service. Also 70% of unintentional failures are Single link failure which creates potential failure along the delivery chains. To get over the problem some mirror server are placed in the network where a request is transferred to one of the mirror servers such that connectivity between user and server is within small hop count even during the failure. This paper targets the stability of network reliability by providing hop length stability and network connectivity. We use the stable matching approach on the network to find the vital link to be protected so that user can access the server within small hop count even if non - vital links fails. Moreover, we prove that the problem can solved in polynomial time conditioned when hop count is determined to be one.

the network topology of interconnection networks plays a fundamental role in the performance of computer architecture. The most commonly used network topologies are Mesh, Torus and their extended versions. Various performance parameters such as latency, throughput, and cost can be used to measure the efficiency of a network. In this paper, the aim is to compare four topologies on basis of average latency versus throughput. Keeping in mind the cost and node degree factors we have proposed a modified version of CCTorus namely, Centrally Connected Partial Torus (CCPTorus). We have evaluated these topologies using OMNeT++ at different values of applied load. Simulation results show that CCTorus has better average throughput than Mesh, C(2)Mesh and CCPTorus.

The topology of interconnection networks is the cardinal factor influencing the performance of computer architecture. The most popularly used network topologies are Mesh and its modified versions. Various performance parameters used to measure the network efficiency are latency, throughput, and cost. In this paper, the main objective is to modify C(2)Mesh in order to enhance performance throughput and reduce the average latency significantly. We also compare three topologies on basis of average latency versus throughput. Keeping in mind the cost and node degree factors we have proposed a modified version of C(2)Mesh namely, Efficient-Modified Centre Connected Mesh (EMC(2)Mesh). We have evaluated these topologies using OMNeT++ at different values of applied load. Simulation results show that EMC(2)Mesh has better average throughput and lower cost than C(2)Mesh.

Complex problems consisting of interdependent subtasks are represented by a direct acyclic graph (DAG). Subtasks of this DAG are scheduled by the scheduler on various grid resources. Scheduling algorithms for grid strive to optimize the schedule. Nowadays a lot of grid resources are attached by P2P approach. Grid systems and P2P model both are newfangled distributed computing approaches. Combining P2P model and grid systems we get P2P grid systems. P2P grid systems require fully decentralized scheduling algorithm, which can schedule interreliant subtasks among nonuniform computational resources. Absence of central scheduler caused the need for decentralized scheduling algorithm. In this paper we have proposed scheduling algorithm which not only is fruitful in optimizing schedule but also does so in fully decentralized fashion. Hence, this unconventional approach suits well for P2P grid systems. Moreover, this algorithm takes accurate scheduling decisions depending on both computation cost and communication cost associated with DAG’s subtasks.

The major applications of multiprocessor system-on-a-chip (SoC) comprises of heterogeneous processors on a single chip. In this kind of integration, major problem is the cache coherence problem i.e maintenance of data integrity. In this paper, we propose a record based methodology to remove the cache coherence problem in heterogeneous multiprocessor platforms with shared and private memories. As shown in our simulation results, a performance improvement up to 18.75% is achieved while using proposed approach in four different benchmarks on LIMES simulator as compared to three other snoopy cache coherence protocols Dragon, Berkeley and MESI.

Load Balancing is very important requirement of any Distributed System. Migration of task is one of the eminent methodologies for the same purpose. Till now utilization is the only key based on which source and destination processor of victim task are being selected. This paper evaluates information theoretic based entropy factor that works similar to utilization factor and its better performance in task migration methodology. We have computed the performance of our system by calculating the Average scheduling latency, Deadline missing rate, Migration rate and finally the Execution ratio of total tasks for 5 and 10 numbers of processors of RTDS.

In Real Time System, the achievement of deadline is the main target of every scheduling algorithm. Earliest Deadline First (EDF), Rate Monotonic (RM), and least Laxity First are some renowned algorithms that work well in their own context. As we know, there is a very common problem Domino's effect in EDF that is generated due to overloading condition (EDF is not working well in overloading situation). Similarly, performance of RM is degraded in underloading condition. We can say that both algorithms are complements of each other. Deadline missing in both events happens because of their utilization bounding strategy. Therefore, in this paper we are proposing a new scheduling algorithm that carries through the drawback of both existing algorithms. Joint EDF-RM scheduling algorithm is implemented in global scheduler that permits task migration mechanism in between processors in the system. In order to check the improved behavior of proposed algorithm we perform simulation. Results are achieved and evaluated in terms of Success Ratio (SR), Average CPU Utilization (ECU), Failure Ratio (FR), and Maximum Tardiness parameters. In the end, the results are compared with the existing (EDF, RM, and D_R_EDF) algorithms. It has been shown that the proposed algorithm performs better during overloading condition as well in underloading condition.

Mesh and its variants are most simple and popular interconnection networks in the research community. Based on the 2D diagonal mesh a new interconnection topology called a modified diagonal mesh interconnection network (MDMIN) has been proposed. The proposed topology performs better than the simple 2D Mesh and diagonal (toroidal) mesh interconnection networks and also overcomes its drawbacks.

This paper describes the transition of a male-pessimal matching set to optimal when it is a man-oriented approach by deleting a pair from matching set considering the score based approach. A descriptive explanation of the proposed algorithm both in a sequential and parallel manner is given. The comparison based theoretical analysis shows that the best case of the algorithm is lower bound of n3.

Interdisciplinary approach is very promising in the modern communication era. An interdisciplinary approach combines more than one discipline and creates something novel that enriches the overall educational experience. This paper is about applying information theoretic concepts of entropy in a real time distributed system (RTDS). Entropy is a very well known concept in thermodynamics, image processing, information theory and many other fields, since utilisation is a very important factor for all real time scheduling algorithms that govern the dynamics of a particular processor. Due to arrival or execution of tasks, the value of this utilisation factor varies, which creates uncertainty in the processor. Therefore, here a calculated entropy value tells the amount of uncertainty present in the particular processor as well as in the entire system. We have calculated entropy (uncertainty) values of the processor in per unit time with the aid of information provided by real time tasks. The resultant graph of these values with respect to time surprisingly ends up showing one-to-one mapping between utilisation and entropy. From these encouraging results, we thought of replacing the utilisation factor with entropy. To the best of our knowledge, this paper is the first that completely implements the role of entropy instead of utilisation in RTDS. After the accomplishment of entropy, based on success ratio, failure ratio, efficiency and tardiness, we compare the performance of the system. Lastly, we acquire similar or healthier results comparatively.

With the implausible rise in microprocessor technology having high speed processors and enhanced the processor-memory speed gap, the design of on chip memory hierarchy is a momentous concern in embedded systems. This paper describes a simulation based performance evaluation of typical cache vs. scratch-pad memory (SPM) design issues in embedded systems such as associativity and replacement policies. The evaluation in paper is done using Simple Scalar simulation tools. in addition, we propose the use of new cache replacement policy, called tag based dual cache replacement policy in SPM and access the best approach for replacement in SPM for embedded systems. According to our simulation results, proposed replacement approach in SPM effectively decreases the data miss ratio of the cache. Comparisons against a SPM solution using dual replacement policy show remarkable advantage of up to 20% in cache miss rate for designs of the same memory size.

Due to monetary limitation, small organizations cannot afford high end supercomputers to solve highly complex tasks. P2P (peer to peer) grid computing is being used nowadays to break complex task into subtasks in order to solve them on different grid resources. Workflows are used to represent these complex tasks. Finishing such complex task in a P2P grid requires scheduling subtasks of workflow in an optimized manner. Several factors play their part in scheduling decisions. The genetic algorithm is very useful in scheduling DAG (directed acyclic graph) based task. Benefit of a genetic algorithm is that it takes into consideration multiple criteria while scheduling. In this paper, we have proposed a precedence level based genetic algorithm (PLBGSA), which yields schedules for workflows in a decentralized fashion. PLBGSA is compared with existing genetic algorithm based scheduling techniques. Fault tolerance is a desirable trait of a P2P grid scheduling algorithm due to the untrustworthy nature of grid resources. PLBGSA handles faults efficiently.

This paper presents a new fault sustainable interconnection network (IN) called as Irregular Augmented Shuffle Exchange Network–3 (IASEN–3) and its routing algorithm. The Performance of IASEN–3 has been evaluated and compared with existing IASEN–2. The experimental results shows that IASEN–3 is more efficient than IASEN–2 in terms of throughput and processor utilization. These results are analyzed in faulty and non–faulty network environments.

There are various states of affairs in Natural Language Processing (NLP), Thermodynamics and Image processing, which illustrates the maximum entropy model (MEM) at different levels. An application of the principle of maximum entropy for approximation of fundamental probability distribution can depend on the variables that used for recitation of system. In Real Time Distributed System (RTDS) arrival time, worst-case execution time and deadline are some fundamental attributes of real time tasks that elucidate the activities of entire system. As we all know that for the acceptance test (in real time scheduling algorithms) of task scheduling and in order to govern the dynamics (task migration and duplication), utilization is the only parameter that brings into play in RTDS. All these methodologies need current utilization value (=1) of all participant processors of the system. To the best of our knowledge, this paper first time introduces one more contender that promises great affinity to replace utilization factor. This paper introduces MEM in RTDS arena that works analogous to the utilization factor with some additional advantages in terms of scalability and processor information. With the help of some mathematical derivations and theoretical examples, we would like to broadcast that entropy can be another parameter that is able to govern the dynamics in RTDS parallel to utilization factor.