China
Africa
Explore chapters and articles related to this topic
A Novel Flexible Framework for Rapidly Integrating Offline Experiment into Remote Laboratory System
Published in Ning Wang, Qianlong Lan, Xuemin Chen, Gangbing Song, Hamid Parsaei, Development of a Remote Laboratory for Engineering Education, 2020
Ning Wang, Qianlong Lan, Xuemin Chen, Gangbing Song, Hamid Parsaei
To compare the performances of different real-time data transmission protocols, we tested 10, 100, 250, and 500 data exchanges per millisecond between the client web module and the server-side middleware in our new system. Each data exchange between the Node.js server and the Chrome browser is a 4K bytes random data string. The Node.js server is running in release model, without debug mode. Meanwhile, the output console messages are minimized for both the server and client. The server is HP Proliant DL380e Gen8. Hardware of the server includes Intel Xeon E5 2.5 GHz processer and 16GB of RAM. The network is the University of Houston’s main campus Wi-Fi network. The download speed is around 45 Mbps and upload speed around 75 Mbps. The Socket.IO is a better data transmission module than WebSocket for real-time data communication without extra plug-ins. Consequently, Socket.IO is the best selection for us to be used for implementing real-time data transmission between the users with the experimental equipment.
NAM: a nearest acquaintance modeling approach for VM allocation using R-Tree
Published in International Journal of Computers and Applications, 2021
Ankita Jiyani, Mehul Mahrishi, Yogesh Meena, Girdhari Singh
The following advances are required during the time span of this test: setting the parameters, programming in CloudSim, and performance assessment. We assessed the execution of our NAM with other four vital conventional heuristic algorithms: Round Robin, Load Balance, Anti Affinity, and VM Allocation Policy Simple (Default Allocation Policy in CloudSim), with same configuration at simulation time. Keeping in mind the end goal to verify the energy efficiency of the proposed algorithm in convoluted condition, the mimicked information focus involves distinctive number of physical hubs. The empirical analysis is conducted on Intel 2.53 GHz Processor with 2GB RAM and Eclipse Neon with Java 1.8 version. We simulate a virtualized datacenter having X86 machine architecture with Linux operating system and Xen as VMM. The datacenter has 100 heterogeneous physical hubs. Some of these physical hubs are HP ProLiant ML110 G5 servers, and the others are HP ProLiant ML110 G4 servers. Every hub is described by the CPU execution characterized in Millions Instructions Per Second (MIPS), the MIPS of the HP ProLiant ML110 G5 and G4 server is 2260 and 1860, separately. The storage of two sorts of servers is 1000 GB and 640 GB. The number of VMs ranges from 50 to 300. Every parameter of physical hubs and Virtual Machine is instated at irregular inside a specific range as indicated by following Table 1. Total number of virtual machine in the datacenter is 15, whereas the total number of hosts is 8.
Energy savings and usability of zero-client computing in office settings
Published in Intelligent Buildings International, 2020
Amanda Farthing, M. Rois Langner, Kim Trenbath
In the RSF data center, VM computation takes place on HP ProLiant blade servers and VM storage is hosted on two Dell EqualLogic server arrays. Table 5 shows the separate pro rata calculations for VM power consumption for these two systems. Averages are based on normal operation and maximums are based on physical space limitations. The average server-based power required for a single VM was calculated to be 18.1 W. To account for the additional power required to maintain the data center, the server-based power was multiplied by the RSF PUE of 1.16. The overall power draw of a single VM in the data center was thus calculated to be 20.99 W.
An effective framework for finding similar cases of dengue from audio and text data using domain thesaurus and case base reasoning
Published in Enterprise Information Systems, 2018
Rajinder Sandhu, Jaspreet Kaur, Vivek Thapar
Now these case bases are compared with each other for finding similarities among them. Similarity finding algorithms are run on HP ProLiant Server with 8GB Ram and Intel core i7 processor with 2.4 GHz computation power. The same algorithm is also run on Amazon EC2 with varying computational power according to the requirement. Figure 10 provides the resource utilization of both Amazon EC2 and local servers. Figure 11 depicts the response time of both cloud and local servers as recorded during the experimental evaluations.