Computing and Information Systems - Research Publications
Permanent URI for this collection
Search Results
1 - 10 of 45
-
ItemA study on the evaluation of HPC microservices in containerized environment(Wiley, 2019-01-01)Containers are gaining popularity over virtual machines as they provide the advantages of virtualization with the performance of near bare metal. The uniformity of support provided by Docker containers across different cloud providers makes them a popular choice for developers. Evolution of microservice architecture allows complex applications to be structured into independent modular components making them easier to manage. High-performance computing (HPC) applications are one such application to be deployed as microservices, placing significant resource requirements on the container framework. However, there is a possibility of interference between different microservices hosted within the same container (intracontainer) and different containers (intercontainer) on the same physical host. In this paper, we describe an extensive experimental investigation to determine the performance evaluation of Docker containers executing heterogeneous HPC microservices. We are particularly concerned with how intracontainer and intercontainer interference influences the performance. Moreover, we investigate the performance variations in Docker containers when control groups (cgroups) are used for resource limitation. For ease of presentation and reproducibility, we use Cloud Evaluation Experiment Methodology (CEEM) to conduct our comprehensive set of experiments. We expect that the results of evaluation can be used in understanding the behavior of HPC microservices in the interfering containerized environment.
-
ItemNo Preview AvailableWattsApp: Power-aware container scheduling(IEEE, 2020-12-01)Containers are popular for deploying workloads. However, there are limited software-based methods (hardware- based methods are expensive) for obtaining the power consumed by containers to facilitate power-aware container scheduling. This paper presents WattsApp, a tool underpinned by a six step software-based method for power-aware container scheduling to minimize power cap violations on a server. The proposed method relies on a neural network-based power estimation model and a power capped container scheduling technique. Experimental studies are pursued in a lab-based environment on 10 benchmarks on Intel and ARM processors. The results highlight that power estimation has negligible overheads - nearly 90% of all data samples can be estimated with less than a 10% error, and the Mean Absolute Percentage Error (MAPE) is less than 6%. The power-aware scheduling of WattsApp is more effective than Intel's Running Power Average Limit (RAPL) based power capping as it does not degrade the performance of all running containers.
-
ItemE-Storm: Replication-based State Management in Distributed Stream Processing Systems(IEEE COMPUTER SOC, 2017-01-01)Apache Storm is a fault-tolerant, distributed inmemory computation system for processing large volumes of high-velocity data in real-time. As an integral part of the fault-tolerance mechanism, Storm's state management is achieved by a checkpointing framework, which commits states regularly and recovers lost states from the latest checkpoint. However, this method involves a remote data store for state preservation and access, resulting in significant overheads to the performance of error-free execution.In this paper, we propose E-Storm, a replication-based state management system that actively maintains multiple state backups on different worker nodes. We build a prototype on top of Storm by extending it with monitoring and recovery modules to support inter-task state transfer whenever needed. The experiments carried out on synthetic and real-world streaming applications confirm that E-Storm outperforms the existing checkpointing method in terms of the resulting application performance, obtaining as much as 9.44 times throughput improvement while reducing the application latency down to 9.8%.
-
ItemDynamic Management of Traffic Signals through Social IoT(Elsevier, 2020-06-04)Traffic congestion is a major threat to transportation sector in every urban city around the world. This causes many adverse effects like, heavy fuel consumption, increased waiting time, pollution, etc. and pose an eminent challenge to the movement of emergency vehicles. To achieve better driving we proceed towards a trending research field called Social Internet of Vehicles (SIoV). A social network paradigm that permits the establishment of social relationships among every vehicle in the network or with any road infrastructure can be radically helpful. This holds as the aim of SIoV, to be beneficial for the drivers, in improving the road safety, avoiding mishaps, and have a friendly-driving environment. In this paper, we propose a Dynamic congestion control with Throughput Maximization scheme based on Social Aspect (D-TMSA) utilizing the social, behavioral and preference-based relationships. Our proposed scheme along with the various social relationship types allocates green signal to maximize the traffic flow passing through an intersection. Simulation results show that the D-TMSA outperforms the existing work by achieving high throughput, lowering the total traveling time and reducing the average waiting time to better the flow of traffic based on their social attributes with each other.
-
ItemSDVADC: Secure Deduplication and Virtual Auditing of Data in Cloud(Elsevier, 2020)Over the last few years, deploying data to cloud service for repository is an appealing passion that avoids efforts on significant information sustenance and administration. In distributed repository utilities, deduplication technique is often exploited to minimize the capacity and bandwidth necesseties of amenities by erasing repetitive data and caching only a solitary duplicate of them. Proof-of-Ownership mechanisms authorize any possessor of the identical information to approve to the distributed repository server that he possess the information in a dynamic way. In repository utilities with enormous information, the repository servers may intend to minimize the capacity of cached information, and the customers may want to examine the integrity of their information with a reasonable cost. We propose Secure Deduplication and Virtual Auditing of Data in Cloud (SDVADC) mechanism that realizes integrity auditing and deduplication of information in cloud. The mechanism supports secure deduplication of information and effective virtual auditing of the documents during the download process. In addition, the proposed mechanism lowers the burden of dataowner to audit documents by himself and there is no need to delegate auditing to the Third Party Auditor (TPA). Experimental results demonstrate that the virtual auditing has low auditing time cost relative to the existing public auditing schemes.
-
ItemJeeva: Enterprise Grid Enabled Web Portal for Protein Secondary Structure Prediction(IEEE, 2008)This paper presents a Grid portal for protein secondary structure prediction developed by using services of Aneka, a .NET-based enterprise Grid technology. The portal is used by research scientists to discover new prediction structures in a parallel manner. An SVM (Support Vector Machine)-based prediction algorithm is used with 64 sample protein sequences as a case study to demonstrate the potential of enterprise Grids.
-
Item
-
Item
-
ItemReliable Provisioning of Spot Instances for Compute-intensive Applications(IEEE, 2012)Cloud computing providers are now offering their unused resources for leasing in the spot market, which has been considered the first step towards a full-fledged market economy for computational resources. Spot instances are virtual machines (VMs) available at lower prices than their standard on-demand counterparts. These VMs will run for as long as the current price is lower than the maximum bid price users are willing to pay per hour. Spot instances have been increasingly used for executing compute-intensive applications. In spite of an apparent economical advantage, due to an intermittent nature of biddable resources, application execution times may be prolonged or they may not finish at all. This paper proposes a resource allocation strategy that addresses the problem of running compute-intensive jobs on a pool of intermittent virtual machines, while also aiming to run applications in a fast and economical way. To mitigate potential unavailability periods, a multifaceted fault-aware resource provisioning policy is proposed. Our solution employs price and runtime estimation mechanisms, as well as three fault tolerance techniques, namely checkpointing, task duplication and migration. We evaluate our strategies using trace-driven simulations, which take as input real price variation traces, as well as an application trace from the Parallel Workload Archive. Our results demonstrate the effectiveness of executing applications on spot instances, respecting QoS constraints, despite occasional failures.
-
Item
- 1 (current)
- 2
- 3
- 4
- 5