The information and communication technology (ICT) sector has grown exponentially in the recent y... more The information and communication technology (ICT) sector has grown exponentially in the recent years. An essential component of the ICT organizations is constituted by the data centers that are densely populated with redundant servers and communicational links to ensure the provision of 99.99 % availability of services; a fact responsible for the heavy energy consumption by data centers. For energy economy, the redundant elements can be powered off based on the current workload within the data center. We propose a Data Center-wide Energy-Efficient Resource Scheduling framework (DCEERS) that schedules data center resources according to the current workload of the data center. Minimum subset of resources to service the current workload are calculated by solving the Minimum Cost Multi Commodity Flow (MCMCF) using the Benders decomposition algorithm. The Benders decomposition algorithm is scalable: it solves the MCMCF problem in linear time for large data center environments. Our simulated results depict that DCEERS achieves better energy efficiency than the previous data center resource scheduling methodologies.
IEEE Transactions on Cloud Computing, Sep 19, 2013
Data centers being an architectural and functional block of cloud computing are integral to the I... more Data centers being an architectural and functional block of cloud computing are integral to the Information and Communication Technology (ICT) sector. Cloud computing is rigorously utilized by various domains, such as agriculture, nuclear science, smart grids, healthcare, and search engines for research, data storage, and analysis. A Data Center Network (DCN) constitutes the communicational backbone of a data center, ascertaining the performance boundaries for cloud infrastructure. The DCN needs to be robust to failures and uncertainties to deliver the required Quality of Service (QoS) level and satisfy Service Level Agreement (SLA). In this paper, we analyze robustness of the state-of-the-art DCNs. Our major contributions are: (a) we present multi-layered graph modeling of various DCNs; (b) we study the classical robustness metrics considering various failure scenarios to perform a comparative analysis; (c) we present the inadequacy of the classical network robustness metrics to appropriately evaluate the DCN robustness; and (d) we propose new procedures to quantify the DCN robustness. Currently, there is no detailed study available centering the DCN robustness. Therefore, we believe that this study will lay a firm foundation for the future DCN robustness research.
Data centers are growing exponentially (in number and size) to accommodate the escalating user a... more Data centers are growing exponentially (in number and size) to accommodate the escalating user and application demands. Likewise, the concerns about the environmental impacts, energy needs, and electricity cost of data centers are also growing. Network infrastructure being the communication backbone of the data center plays a pivotal role in the data center’s scalability, performance, energy consumption, and cost. Research community is endeavoring hard to overcome the challenges faced by the legacy Data Center Networks (DCNs). Serious efforts have been made to handle the problems in various DCN areas. This survey presents significant insights to the state-of-the-art research conducted pertaining to the DCN domain along with a detailed discussion of the energy efficiency aspects of the DCNs. The authors explored: (a) DCN architectures (electrical, optical, and hybrid), (b) network traffic management and characterization, (c) DCN performance monitoring, (d) network-aware resource allocation, (e) DCN experimentation techniques, and (f) energy efficiency. The survey presents an overview of the ongoing research in the broad domain of DCNs and highlights the challenges faced by the DCN research community.
Data Center Networks (DCNs) constitute the communication backbone for the cloud computing paradig... more Data Center Networks (DCNs) constitute the communication backbone for the cloud computing paradigm. Recently, network connectivity analysis in terms of reliability has received attention from the network research community. The traditional network features are useful; however, they are insufficient to determine how well-connected or well-designed a DCN is against the node or link removals. In this letter, we present a connectivity analysis of three well-known DCN architectures, namely: (a) ThreeTier, (b) FatTree, and (c) DCell. Our analysis reveals that the classic connectivity measures are inadequate for evaluating DCN connectivity. Therefore, we propose µ-A2TR, a novel metric to characterize network connectivity in the case of node or link failures. Experimental results reveal that the DCNs exhibit a moderate level of connectivity in the case of random node removals. However, connectivity decays abruptly when considering the targeted nodes removal. Moreover, the connectivity analysis depicts significant differences among the considered DCNs.
In this paper we analyze different TCP models and propose a hybrid solution that can significantl... more In this paper we analyze different TCP models and propose a hybrid solution that can significantly improve the performance of network when reordering of packets or short term congestion occurs. The key idea here is to delay congestion response by dynamic DUPTHRESH adjustment so that local recovery can be done, new transmission should be after every two DUPACK to minimize receiver buffer requirement and at the same time maintains basis ACK clocking. We also introduce linear decrease algorithm for CWND adjustment. The Simulation results show that our model can achieve significant performance improvement as compare to previous work with minimal overhead. We carried out a simple modification to the existing solutions therefore it is easy to deploy.
Internet is one of the most extensively used communications medium in the world, which makes the ... more Internet is one of the most extensively used communications medium in the world, which makes the whole world a global village. The Internet is growing at an annualized rate of 18% and now has one billion users. Second billion users will follow in the next ten years. Unfortunately there is very little work done in order to use the IP network in the field of emergency services. These services are mostly based on PSTN (public switched telephone network). In this paper we enhance the previous work that has been done in this regard and tried to improved the performance by bringing some changes in its architecture, which ensures the LI (location Information) stipulation which is a the most challenging aspect using IP flat address model. We have used the concept of DHCP relay (dynamic host configuration protocol) in order to ensure the LI from the caller side. Along with this, our technique also simplifies the architecture of the system.
The information and communication technology (ICT) sector has grown exponentially in the recent y... more The information and communication technology (ICT) sector has grown exponentially in the recent years. An essential component of the ICT organizations is constituted by the data centers that are densely populated with redundant servers and communicational links to ensure the provision of 99.99 % availability of services; a fact responsible for the heavy energy consumption by data centers. For energy economy, the redundant elements can be powered off based on the current workload within the data center. We propose a Data Center-wide Energy-Efficient Resource Scheduling framework (DCEERS) that schedules data center resources according to the current workload of the data center. Minimum subset of resources to service the current workload are calculated by solving the Minimum Cost Multi Commodity Flow (MCMCF) using the Benders decomposition algorithm. The Benders decomposition algorithm is scalable: it solves the MCMCF problem in linear time for large data center environments. Our simulated results depict that DCEERS achieves better energy efficiency than the previous data center resource scheduling methodologies.
IEEE Transactions on Cloud Computing, Sep 19, 2013
Data centers being an architectural and functional block of cloud computing are integral to the I... more Data centers being an architectural and functional block of cloud computing are integral to the Information and Communication Technology (ICT) sector. Cloud computing is rigorously utilized by various domains, such as agriculture, nuclear science, smart grids, healthcare, and search engines for research, data storage, and analysis. A Data Center Network (DCN) constitutes the communicational backbone of a data center, ascertaining the performance boundaries for cloud infrastructure. The DCN needs to be robust to failures and uncertainties to deliver the required Quality of Service (QoS) level and satisfy Service Level Agreement (SLA). In this paper, we analyze robustness of the state-of-the-art DCNs. Our major contributions are: (a) we present multi-layered graph modeling of various DCNs; (b) we study the classical robustness metrics considering various failure scenarios to perform a comparative analysis; (c) we present the inadequacy of the classical network robustness metrics to appropriately evaluate the DCN robustness; and (d) we propose new procedures to quantify the DCN robustness. Currently, there is no detailed study available centering the DCN robustness. Therefore, we believe that this study will lay a firm foundation for the future DCN robustness research.
Data centers are growing exponentially (in number and size) to accommodate the escalating user a... more Data centers are growing exponentially (in number and size) to accommodate the escalating user and application demands. Likewise, the concerns about the environmental impacts, energy needs, and electricity cost of data centers are also growing. Network infrastructure being the communication backbone of the data center plays a pivotal role in the data center’s scalability, performance, energy consumption, and cost. Research community is endeavoring hard to overcome the challenges faced by the legacy Data Center Networks (DCNs). Serious efforts have been made to handle the problems in various DCN areas. This survey presents significant insights to the state-of-the-art research conducted pertaining to the DCN domain along with a detailed discussion of the energy efficiency aspects of the DCNs. The authors explored: (a) DCN architectures (electrical, optical, and hybrid), (b) network traffic management and characterization, (c) DCN performance monitoring, (d) network-aware resource allocation, (e) DCN experimentation techniques, and (f) energy efficiency. The survey presents an overview of the ongoing research in the broad domain of DCNs and highlights the challenges faced by the DCN research community.
Data Center Networks (DCNs) constitute the communication backbone for the cloud computing paradig... more Data Center Networks (DCNs) constitute the communication backbone for the cloud computing paradigm. Recently, network connectivity analysis in terms of reliability has received attention from the network research community. The traditional network features are useful; however, they are insufficient to determine how well-connected or well-designed a DCN is against the node or link removals. In this letter, we present a connectivity analysis of three well-known DCN architectures, namely: (a) ThreeTier, (b) FatTree, and (c) DCell. Our analysis reveals that the classic connectivity measures are inadequate for evaluating DCN connectivity. Therefore, we propose µ-A2TR, a novel metric to characterize network connectivity in the case of node or link failures. Experimental results reveal that the DCNs exhibit a moderate level of connectivity in the case of random node removals. However, connectivity decays abruptly when considering the targeted nodes removal. Moreover, the connectivity analysis depicts significant differences among the considered DCNs.
In this paper we analyze different TCP models and propose a hybrid solution that can significantl... more In this paper we analyze different TCP models and propose a hybrid solution that can significantly improve the performance of network when reordering of packets or short term congestion occurs. The key idea here is to delay congestion response by dynamic DUPTHRESH adjustment so that local recovery can be done, new transmission should be after every two DUPACK to minimize receiver buffer requirement and at the same time maintains basis ACK clocking. We also introduce linear decrease algorithm for CWND adjustment. The Simulation results show that our model can achieve significant performance improvement as compare to previous work with minimal overhead. We carried out a simple modification to the existing solutions therefore it is easy to deploy.
Internet is one of the most extensively used communications medium in the world, which makes the ... more Internet is one of the most extensively used communications medium in the world, which makes the whole world a global village. The Internet is growing at an annualized rate of 18% and now has one billion users. Second billion users will follow in the next ten years. Unfortunately there is very little work done in order to use the IP network in the field of emergency services. These services are mostly based on PSTN (public switched telephone network). In this paper we enhance the previous work that has been done in this regard and tried to improved the performance by bringing some changes in its architecture, which ensures the LI (location Information) stipulation which is a the most challenging aspect using IP flat address model. We have used the concept of DHCP relay (dynamic host configuration protocol) in order to ensure the LI from the caller side. Along with this, our technique also simplifies the architecture of the system.
Uploads