Peer-to-peer Networking and Applications, Oct 26, 2021
In recent years, cloud is being widely used to host numerous distributed applications. The expand... more In recent years, cloud is being widely used to host numerous distributed applications. The expanding usage of cloud has introduced greater sensitivity in the environment. Therefore, most of the applications require that an effective fault tolerant mechanism must be in place. A fault tolerant mechanism involves detection as well as recovery from failures; traditionally checkpointing has been used to serve the purpose. The conventional checkpointing methods have also been tried in cloud e.g. , periodic checkpointing and application based checkpointing; however, the periodic checkpointing is time inefficient and the application based checkpointing is space inefficient. Secondly, the above methods have been implemented using synchronous approach, which is inherently message inefficient, less scalable and has high synchronization latency. Furthermore, the asynchronous approaches are practically not viable owing to their inability to detect failures. In addition, the cloud entails massive scalability, thus we have proposed a quasi-synchronous checkpointing algorithm for cloud based distributed applications that exhibits better space efficiency while keeping latency under strict control. Our claims have been substantiated with static analysis and suitable simulation experiments.
MapReduce is the renowned distributed and parallel programming model with an extensive support fo... more MapReduce is the renowned distributed and parallel programming model with an extensive support for large scale computing. However, the performance of MapReduce is currently limited by the default scheduler as it is not suitable for heterogeneous environments and it does not consider certain user constraints such as deadlines. The paper proposes a deadline-aware scheduling algorithm that selectively uses speculative execution when the job approaches its deadline in order to expedite job's execution. The algorithm is implemented on the heterogeneous Hadoop cluster and the evaluation shows significant improvement in the performance. The performance improvement was observed as the number of jobs that miss the deadlines as well as the overall execution time for different workloads was minimized.
The smart-phones are computing devices used to execute resource intensive applications. Despite c... more The smart-phones are computing devices used to execute resource intensive applications. Despite considerable progress in the past years, smart-phones still have low computing power with some inbuilt problems such as limited battery, resource deficiency, and limited storage. Mobile Cloud Computing (MCC) improves the overall performance of smart-phones by providing extra resources of remote cloud server. Application Offloading is a primary technique used in MCC which increases the capabilities of SMDs. The proposed offloading framework divides the application into tasks, thus, offloading them on remote cloud server for execution. The offloading process in cloud computing extends battery duration and conserves energy. However, it is critical to evaluate the cases where offloading ensure advantages in terms of data transfer, computing power needed, costs of using the clouds, a determined Quality of service (QoS). The paper presents a fault-tolerant energy efficient framework for application offloading with minimal energy consumption and response time.
Wireless Sensor Networks (WSNs) have extensively been worked upon due to their ability to monitor... more Wireless Sensor Networks (WSNs) have extensively been worked upon due to their ability to monitor various physical phenomenon. Time synchronization is the most fundamental and most important challenge faced by WSNs. Time Synchronization in WSN can be achieved either by gaining consensus amongst all the nodes or by a reference node which broadcasts a single unit of time to the entire network. The paper describes a novel technique to achieve fully distributed cluster clock synchronization, which synchronizes all the nodes in the network to a common consensus value. The algorithm employs clustering and confidence weighted running average (CWA) method for offset and skew compensations. The efficiency of protocol is validated with simulation results presented in the paper.
Energy efficiency of a MapReduce system has become an essential part of infrastructure management... more Energy efficiency of a MapReduce system has become an essential part of infrastructure management in the field of big data analytics. Here, Hadoop scheduler plays a vital role in order to ensure the energy efficiency of the system. A handful of MapReduce scheduling algorithms have been proposed in the literature for slot-based Hadoop system (i.e., Hadoop 0.x and Hadoop 1.x) to minimize the overall energy consumption. However, YARN-based Hadoop schedulers have not been discussed much in the literature. In this paper, we design a scheduling model for Hadoop YARN architecture and formulate the energy efficient scheduling problem as an Integer Program. To solve the problem, we propose a Greedy scheduler which selects the best job with minimum energy consumption in each iteration. We evaluate the performance of the proposed algorithm against the FAIR and Capacity schedulers and find out that our greedy scheduler shows better results for both CPU- and I/O intensive workloads.
2017 8th International Conference on Computing, Communication and Networking Technologies (ICCCNT), 2017
MapReduce is the renowned distributed and parallel programming model with an extensive support fo... more MapReduce is the renowned distributed and parallel programming model with an extensive support for large scale computing. However, the performance of MapReduce is currently limited by the default scheduler as it is not suitable for heterogeneous environments and it does not consider certain user constraints such as deadlines. The paper proposes a deadline-aware scheduling algorithm that selectively uses speculative execution when the job approaches its deadline in order to expedite job's execution. The algorithm is implemented on the heterogeneous Hadoop cluster and the evaluation shows significant improvement in the performance. The performance improvement was observed as the number of jobs that miss the deadlines as well as the overall execution time for different workloads was minimized.
2017 8th International Conference on Computing, Communication and Networking Technologies (ICCCNT), 2017
Wireless Sensor Networks (WSNs) have extensively been worked upon due to their ability to monitor... more Wireless Sensor Networks (WSNs) have extensively been worked upon due to their ability to monitor various physical phenomenon. Time synchronization is the most fundamental and most important challenge faced by WSNs. Time Synchronization in WSN can be achieved either by gaining consensus amongst all the nodes or by a reference node which broadcasts a single unit of time to the entire network. The paper describes a novel technique to achieve fully distributed cluster clock synchronization, which synchronizes all the nodes in the network to a common consensus value. The algorithm employs clustering and confidence weighted running average (CWA) method for offset and skew compensations. The efficiency of protocol is validated with simulation results presented in the paper.
2016 Fourth International Conference on Parallel, Distributed and Grid Computing (PDGC), 2016
The smart-phones are computing devices used to execute resource intensive applications. Despite c... more The smart-phones are computing devices used to execute resource intensive applications. Despite considerable progress in the past years, smart-phones still have low computing power with some inbuilt problems such as limited battery, resource deficiency, and limited storage. Mobile Cloud Computing (MCC) improves the overall performance of smart-phones by providing extra resources of remote cloud server. Application Offloading is a primary technique used in MCC which increases the capabilities of SMDs. The proposed offloading framework divides the application into tasks, thus, offloading them on remote cloud server for execution. The offloading process in cloud computing extends battery duration and conserves energy. However, it is critical to evaluate the cases where offloading ensure advantages in terms of data transfer, computing power needed, costs of using the clouds, a determined Quality of service (QoS). The paper presents a fault-tolerant energy efficient framework for applica...
Checkpointing is a popular fault-tolerance technique in mobile computing. It is a way to record t... more Checkpointing is a popular fault-tolerance technique in mobile computing. It is a way to record the current status of information, which may be required while recovery after failures. Thus, in case of failure, computation may be restarted from the saved checkpoint instead from the beginning. Checkpointing-based rollback recovery is used in various domains like applied sciences, database management, computer networks and many more. Mobile Adhoc Networks (MANETs) consists of a number of mobile hosts (MHs) connected with each other via wireless links. Here, the challenge is limited power supply and storage capacity. Therefore, checkpointing is a major confront in such dynamic environment. This survey paper presents a comprehensive survey of checkpointing algorithms along with an analysis of their performance characteristics. .
In recent years, cloud is being widely used to host numerous distributed applications. The expand... more In recent years, cloud is being widely used to host numerous distributed applications. The expanding usage of cloud has introduced greater sensitivity in the environment. Therefore, most of the applications require that an effective fault tolerant mechanism must be in place. A fault tolerant mechanism involves detection as well as recovery from failures; traditionally checkpointing has been used to serve the purpose. The conventional checkpointing methods have also been tried in cloud e.g. , periodic checkpointing and application based checkpointing; however, the periodic checkpointing is time inefficient and the application based checkpointing is space inefficient. Secondly, the above methods have been implemented using synchronous approach, which is inherently message inefficient, less scalable and has high synchronization latency. Furthermore, the asynchronous approaches are practically not viable owing to their inability to detect failures. In addition, the cloud entails massive scalability, thus we have proposed a quasi-synchronous checkpointing algorithm for cloud based distributed applications that exhibits better space efficiency while keeping latency under strict control. Our claims have been substantiated with static analysis and suitable simulation experiments.
The phenomenal success of certain crowdsourced online platforms, such as Wikipedia, is accredited... more The phenomenal success of certain crowdsourced online platforms, such as Wikipedia, is accredited to their ability to tap the crowd's potential to collaboratively build knowledge. While it is well known that the crowd's collective wisdom surpasses the cumulative individual expertise, little is understood on the dynamics of knowledge building in a crowdsourced environment. A proper understanding of the dynamics of knowledge building in a crowdsourced environment would enable one in the better designing of such environments to solicit knowledge from the crowd. Our experiment on crowdsourced systems based on annotations shows that an important reason for the rapid knowledge building in such environments is due to variance in expertise. First, we used as our test bed, a customized Crowdsourced Annotation System (CAS) which provides a group of users the facility to annotate a given document while trying to understand it. Our results showed the presence of different genres of prof...
Proceedings of the International Conference on Advances in Computing and Artificial Intelligence - ACAI '11, 2011
Abstract The problem of agreement has been widely studied by many research groups in crash-stop a... more Abstract The problem of agreement has been widely studied by many research groups in crash-stop as well as Byzantine model for distributed transaction systems. In one of our previous works [12], we assumed a trustworthy Transaction Manager, TM amenable to the ...
Peer-to-peer Networking and Applications, Oct 26, 2021
In recent years, cloud is being widely used to host numerous distributed applications. The expand... more In recent years, cloud is being widely used to host numerous distributed applications. The expanding usage of cloud has introduced greater sensitivity in the environment. Therefore, most of the applications require that an effective fault tolerant mechanism must be in place. A fault tolerant mechanism involves detection as well as recovery from failures; traditionally checkpointing has been used to serve the purpose. The conventional checkpointing methods have also been tried in cloud e.g. , periodic checkpointing and application based checkpointing; however, the periodic checkpointing is time inefficient and the application based checkpointing is space inefficient. Secondly, the above methods have been implemented using synchronous approach, which is inherently message inefficient, less scalable and has high synchronization latency. Furthermore, the asynchronous approaches are practically not viable owing to their inability to detect failures. In addition, the cloud entails massive scalability, thus we have proposed a quasi-synchronous checkpointing algorithm for cloud based distributed applications that exhibits better space efficiency while keeping latency under strict control. Our claims have been substantiated with static analysis and suitable simulation experiments.
MapReduce is the renowned distributed and parallel programming model with an extensive support fo... more MapReduce is the renowned distributed and parallel programming model with an extensive support for large scale computing. However, the performance of MapReduce is currently limited by the default scheduler as it is not suitable for heterogeneous environments and it does not consider certain user constraints such as deadlines. The paper proposes a deadline-aware scheduling algorithm that selectively uses speculative execution when the job approaches its deadline in order to expedite job's execution. The algorithm is implemented on the heterogeneous Hadoop cluster and the evaluation shows significant improvement in the performance. The performance improvement was observed as the number of jobs that miss the deadlines as well as the overall execution time for different workloads was minimized.
The smart-phones are computing devices used to execute resource intensive applications. Despite c... more The smart-phones are computing devices used to execute resource intensive applications. Despite considerable progress in the past years, smart-phones still have low computing power with some inbuilt problems such as limited battery, resource deficiency, and limited storage. Mobile Cloud Computing (MCC) improves the overall performance of smart-phones by providing extra resources of remote cloud server. Application Offloading is a primary technique used in MCC which increases the capabilities of SMDs. The proposed offloading framework divides the application into tasks, thus, offloading them on remote cloud server for execution. The offloading process in cloud computing extends battery duration and conserves energy. However, it is critical to evaluate the cases where offloading ensure advantages in terms of data transfer, computing power needed, costs of using the clouds, a determined Quality of service (QoS). The paper presents a fault-tolerant energy efficient framework for application offloading with minimal energy consumption and response time.
Wireless Sensor Networks (WSNs) have extensively been worked upon due to their ability to monitor... more Wireless Sensor Networks (WSNs) have extensively been worked upon due to their ability to monitor various physical phenomenon. Time synchronization is the most fundamental and most important challenge faced by WSNs. Time Synchronization in WSN can be achieved either by gaining consensus amongst all the nodes or by a reference node which broadcasts a single unit of time to the entire network. The paper describes a novel technique to achieve fully distributed cluster clock synchronization, which synchronizes all the nodes in the network to a common consensus value. The algorithm employs clustering and confidence weighted running average (CWA) method for offset and skew compensations. The efficiency of protocol is validated with simulation results presented in the paper.
Energy efficiency of a MapReduce system has become an essential part of infrastructure management... more Energy efficiency of a MapReduce system has become an essential part of infrastructure management in the field of big data analytics. Here, Hadoop scheduler plays a vital role in order to ensure the energy efficiency of the system. A handful of MapReduce scheduling algorithms have been proposed in the literature for slot-based Hadoop system (i.e., Hadoop 0.x and Hadoop 1.x) to minimize the overall energy consumption. However, YARN-based Hadoop schedulers have not been discussed much in the literature. In this paper, we design a scheduling model for Hadoop YARN architecture and formulate the energy efficient scheduling problem as an Integer Program. To solve the problem, we propose a Greedy scheduler which selects the best job with minimum energy consumption in each iteration. We evaluate the performance of the proposed algorithm against the FAIR and Capacity schedulers and find out that our greedy scheduler shows better results for both CPU- and I/O intensive workloads.
2017 8th International Conference on Computing, Communication and Networking Technologies (ICCCNT), 2017
MapReduce is the renowned distributed and parallel programming model with an extensive support fo... more MapReduce is the renowned distributed and parallel programming model with an extensive support for large scale computing. However, the performance of MapReduce is currently limited by the default scheduler as it is not suitable for heterogeneous environments and it does not consider certain user constraints such as deadlines. The paper proposes a deadline-aware scheduling algorithm that selectively uses speculative execution when the job approaches its deadline in order to expedite job's execution. The algorithm is implemented on the heterogeneous Hadoop cluster and the evaluation shows significant improvement in the performance. The performance improvement was observed as the number of jobs that miss the deadlines as well as the overall execution time for different workloads was minimized.
2017 8th International Conference on Computing, Communication and Networking Technologies (ICCCNT), 2017
Wireless Sensor Networks (WSNs) have extensively been worked upon due to their ability to monitor... more Wireless Sensor Networks (WSNs) have extensively been worked upon due to their ability to monitor various physical phenomenon. Time synchronization is the most fundamental and most important challenge faced by WSNs. Time Synchronization in WSN can be achieved either by gaining consensus amongst all the nodes or by a reference node which broadcasts a single unit of time to the entire network. The paper describes a novel technique to achieve fully distributed cluster clock synchronization, which synchronizes all the nodes in the network to a common consensus value. The algorithm employs clustering and confidence weighted running average (CWA) method for offset and skew compensations. The efficiency of protocol is validated with simulation results presented in the paper.
2016 Fourth International Conference on Parallel, Distributed and Grid Computing (PDGC), 2016
The smart-phones are computing devices used to execute resource intensive applications. Despite c... more The smart-phones are computing devices used to execute resource intensive applications. Despite considerable progress in the past years, smart-phones still have low computing power with some inbuilt problems such as limited battery, resource deficiency, and limited storage. Mobile Cloud Computing (MCC) improves the overall performance of smart-phones by providing extra resources of remote cloud server. Application Offloading is a primary technique used in MCC which increases the capabilities of SMDs. The proposed offloading framework divides the application into tasks, thus, offloading them on remote cloud server for execution. The offloading process in cloud computing extends battery duration and conserves energy. However, it is critical to evaluate the cases where offloading ensure advantages in terms of data transfer, computing power needed, costs of using the clouds, a determined Quality of service (QoS). The paper presents a fault-tolerant energy efficient framework for applica...
Checkpointing is a popular fault-tolerance technique in mobile computing. It is a way to record t... more Checkpointing is a popular fault-tolerance technique in mobile computing. It is a way to record the current status of information, which may be required while recovery after failures. Thus, in case of failure, computation may be restarted from the saved checkpoint instead from the beginning. Checkpointing-based rollback recovery is used in various domains like applied sciences, database management, computer networks and many more. Mobile Adhoc Networks (MANETs) consists of a number of mobile hosts (MHs) connected with each other via wireless links. Here, the challenge is limited power supply and storage capacity. Therefore, checkpointing is a major confront in such dynamic environment. This survey paper presents a comprehensive survey of checkpointing algorithms along with an analysis of their performance characteristics. .
In recent years, cloud is being widely used to host numerous distributed applications. The expand... more In recent years, cloud is being widely used to host numerous distributed applications. The expanding usage of cloud has introduced greater sensitivity in the environment. Therefore, most of the applications require that an effective fault tolerant mechanism must be in place. A fault tolerant mechanism involves detection as well as recovery from failures; traditionally checkpointing has been used to serve the purpose. The conventional checkpointing methods have also been tried in cloud e.g. , periodic checkpointing and application based checkpointing; however, the periodic checkpointing is time inefficient and the application based checkpointing is space inefficient. Secondly, the above methods have been implemented using synchronous approach, which is inherently message inefficient, less scalable and has high synchronization latency. Furthermore, the asynchronous approaches are practically not viable owing to their inability to detect failures. In addition, the cloud entails massive scalability, thus we have proposed a quasi-synchronous checkpointing algorithm for cloud based distributed applications that exhibits better space efficiency while keeping latency under strict control. Our claims have been substantiated with static analysis and suitable simulation experiments.
The phenomenal success of certain crowdsourced online platforms, such as Wikipedia, is accredited... more The phenomenal success of certain crowdsourced online platforms, such as Wikipedia, is accredited to their ability to tap the crowd's potential to collaboratively build knowledge. While it is well known that the crowd's collective wisdom surpasses the cumulative individual expertise, little is understood on the dynamics of knowledge building in a crowdsourced environment. A proper understanding of the dynamics of knowledge building in a crowdsourced environment would enable one in the better designing of such environments to solicit knowledge from the crowd. Our experiment on crowdsourced systems based on annotations shows that an important reason for the rapid knowledge building in such environments is due to variance in expertise. First, we used as our test bed, a customized Crowdsourced Annotation System (CAS) which provides a group of users the facility to annotate a given document while trying to understand it. Our results showed the presence of different genres of prof...
Proceedings of the International Conference on Advances in Computing and Artificial Intelligence - ACAI '11, 2011
Abstract The problem of agreement has been widely studied by many research groups in crash-stop a... more Abstract The problem of agreement has been widely studied by many research groups in crash-stop as well as Byzantine model for distributed transaction systems. In one of our previous works [12], we assumed a trustworthy Transaction Manager, TM amenable to the ...
Uploads
Papers by Poonam Saini