Performance analysis of Software is one of the important consideration. Performance Engineering m... more Performance analysis of Software is one of the important consideration. Performance Engineering methodology is used to examine the performance of a system by taking the performance parameters at different stages of SDLC. In this paper we model and simulate different algorithm namely Graph Analysis Algorithm, Queuing Model, Open Queuing Network model, and Closed Queuing Network Model. The Performance parameters, list and type of nodes that are used in the Execution Graph are provided to the Tool to compute and estimate the performance metrics like Response time, Throughput, Residence time, Utilization. Queue Length of the software and also describe how the distribution of mean service time affects the performance metrics for each scenario is also addressed. The Tool also calculates and estimates the performance metrics for the observed intervals of the software and results are compared.
Resource allocation is one of the main issues for providing services in an efficient way on distr... more Resource allocation is one of the main issues for providing services in an efficient way on distributed databases. Improving the performance is one of the key research issues by proper design of efficient distributed database and proper use of resources in information technology. The cost of each computational service depends on the amount of computations. The system resources have to be allocated in order to handle the workload and minimize the cost of computing by using proper allocation strategy. Performance is strongly related to the allocation of resources and data fragments in a distributed environment. The query requires data to be accessed from one or multiple sites. The required data are fragmented and placed in various data centers, where resources exist. In this paper, we propose an algorithm to minimize the total data-transfer cost required for processing the queries by proper allocation of resources. We are finding the optimum allocation strategy, which predicts the performance by estimating the cost.
Cloud computing is a fast-emerging area with so many benefits for the end users. Many leading fir... more Cloud computing is a fast-emerging area with so many benefits for the end users. Many leading firms make use of cloud mainly to store their huge volume of data and other such purposes. When speaking about cloud, we need to focus on the data centers. This is because the virtual machines (VM) present in the cloud are mapped to the physical devices that are present in the data centers. This seems to be a simple process, but the tricky part is when the number of users increases. Number of users send the request for a particular or more than one service to the cloud. The service providers have to handle the entire request and also concentrate on the satisfaction of the end user along with the quality of service (QoS). Most important thing is that the service provider must give a quick response to the cloud consumers, i.e., the delay must be minimized. Only then the QoS provided could be maintained. In this paper, we are going to see about how the tasks are being scheduled and allocated optimally to the VMs present in the cloud. We are going to delineate the various algorithms that are implemented so far for this process. In this paper, we have described an overview of cloud, task scheduling and VM allocation.
Performance analysis of Software is one of the important consideration. Performance Engineering m... more Performance analysis of Software is one of the important consideration. Performance Engineering methodology is used to examine the performance of a system by taking the performance parameters at different stages of SDLC. In this paper we model and simulate different algorithm namely Graph Analysis Algorithm, Queuing Model, Open Queuing Network model, and Closed Queuing Network Model. The Performance parameters, list and type of nodes that are used in the Execution Graph are provided to the Tool to compute and estimate the performance metrics like Response time, Throughput, Residence time, Utilization. Queue Length of the software and also describe how the distribution of mean service time affects the performance metrics for each scenario is also addressed. The Tool also calculates and estimates the performance metrics for the observed intervals of the software and results are compared.
Resource allocation is one of the main issues for providing services in an efficient way on distr... more Resource allocation is one of the main issues for providing services in an efficient way on distributed databases. Improving the performance is one of the key research issues by proper design of efficient distributed database and proper use of resources in information technology. The cost of each computational service depends on the amount of computations. The system resources have to be allocated in order to handle the workload and minimize the cost of computing by using proper allocation strategy. Performance is strongly related to the allocation of resources and data fragments in a distributed environment. The query requires data to be accessed from one or multiple sites. The required data are fragmented and placed in various data centers, where resources exist. In this paper, we propose an algorithm to minimize the total data-transfer cost required for processing the queries by proper allocation of resources. We are finding the optimum allocation strategy, which predicts the performance by estimating the cost.
Cloud computing is a fast-emerging area with so many benefits for the end users. Many leading fir... more Cloud computing is a fast-emerging area with so many benefits for the end users. Many leading firms make use of cloud mainly to store their huge volume of data and other such purposes. When speaking about cloud, we need to focus on the data centers. This is because the virtual machines (VM) present in the cloud are mapped to the physical devices that are present in the data centers. This seems to be a simple process, but the tricky part is when the number of users increases. Number of users send the request for a particular or more than one service to the cloud. The service providers have to handle the entire request and also concentrate on the satisfaction of the end user along with the quality of service (QoS). Most important thing is that the service provider must give a quick response to the cloud consumers, i.e., the delay must be minimized. Only then the QoS provided could be maintained. In this paper, we are going to see about how the tasks are being scheduled and allocated optimally to the VMs present in the cloud. We are going to delineate the various algorithms that are implemented so far for this process. In this paper, we have described an overview of cloud, task scheduling and VM allocation.
Uploads
Papers by JAGANNATHA S