Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Subject: search.filters.subject.Scheduling

Search Results

Now showing 1 - 8 of 8
  • No Thumbnail Available
    Item
    Optimal scheduling of computational task in cloud using Virtual Machine Tree
    (2012) Achar, R.; Santhi Thilagam, P.; Shwetha, D.; Pooja, H.; Roshn; I Andrea
    The increasing demand in computing resources and widespread adaptation of Service Oriented Architecture (SOA) has made cloud as a new IT delivery mechanism. In cloud, computing resources are provided to the requester as a service, which include Software as a Service (SaaS), Platform as a Service (PaaS) and Infrastructure as a Service (IaaS). Cloud Computing is still in developing stage and faces many challenges. Out of the various issues, scheduling plays a very important role in determining the efficient execution of tasks in cloud environment. In this paper we present a scheduling algorithm which uses tree based data structure called Virtual Machine Tree (VMT) for efficient execution of tasks. The proposed algorithm is tested using CloudSim simulator and the results shows that algorithm gives better performance compared to other traditional scheduling algorithms. © 2012 IEEE.
  • No Thumbnail Available
    Item
    Analysis of MapReduce scheduling and its improvements in cloud environment
    (Institute of Electrical and Electronics Engineers Inc., 2015) D'Souza, S.; Chandrasekaran, K.
    MapReduce has become a prominent Parallel processing model used for analysing large scale data. MapReduce applications are increasingly being deployed in the cloud along with other applications sharing the same physical resources. In this scenario, efficient scheduling of MapReduce applications is of utmost importance. Also, MapReduce has to consider various other parameters like energy efficiency and meeting SLA goals besides achieving performance when executing jobs in cloud environments. In this work, we have classified MapReduce Scheduling as Cluster based Scheduling and Objective based Scheduling. We then summarize and analyse the different class of schedulers highlighting the strong points and limitations of each of the scheduling approaches. The Adaptive scheduling techniques provide dynamic resource management and meet performance goals. The Energy efficient scheduling techniques aim to cut data centre costs by using different approaches. Finally, we discuss the current challenges and future work. © 2015 IEEE.
  • No Thumbnail Available
    Item
    Analysis of emerging workflow scheduling algorithms in cloud
    (Institute of Electrical and Electronics Engineers Inc., 2016) Raghavan, S.; Chandrasekaran, K.
    Workflow has a great impact on several diverse fields of science such as Physics, Bioinformatics, Astronomy etc. The computational requirements of the world is increasing exponentially and so the use of workflows. These requirements are expected to increase by many folds. Thus workflow can be considered as one of the important areas to be explored. Workflow is used extensively by the scientific community for automation. Similarly these workflows have a significant impact on grid computing, cloud computing and other distributed environments. Moreover it is because of grid computing technology that power of workflow is realizable. Similarly cloud computing also supports workflows. Workflow scheduling is considered to be the core when it comes to workflow thus it is highly important to know what are workflow scheduling algorithms available for scheduling workflow in cloud. This paper gives an overview about workflow, the impact of workflows on cloud and other important issues related to it are presented. This paper primarily focuses on Workflow scheduling algorithms in cloud and analysis of the same is presented. © 2015 IEEE.
  • No Thumbnail Available
    Item
    Optimization-Aware scheduling in cloud computing
    (Association for Computing Machinery acmhelp@acm.org, 2016) Binu, A.; George, N.; Chandrasekaran, K.
    Cloud computing allows delivery of computational resources via internet. Some of the computational resources include signals, codes and physical resources. Cloud users require computational resources for execution of their tasks. The computational elements, when thoroughly assigned to the cloud users according to requirement facilitate an efficient scheduling mechanism. The efficiency is dependent on the parameters chosen to promote the assignment of task to resources. Here, computational cost is chosen as the optimization factor, and the solution with the least cost is selected as the best assignment. A Cuckoo Search inspired technique is used, and task assignment to resources is done by validating the solution with the least value of cost. © 2016 ACM.
  • No Thumbnail Available
    Item
    Construing microservice architectures: State-of-the-art algorithms and research issues
    (Springer Verlag service@springer.de, 2019) Nene, A.V.; Joseph, C.T.; Chandrasekaran, K.
    Cloud Computing is one of the leading paradigms in the IT industry. Earlier, cloud applications used to be built as single monolithic applications, and are now built using the Microservices Architectural Style. Along with several advantages, the microservices architecture also introduce challenges at the infrastructural level. Five such concerns are identified and analysed in this paper. The paper presents the state-of-art in different infrastructural concerns of microservices, namely, load balancing, scheduling, energy efficiency, security and resource management of microservices. The paper also suggests some future trends and research domains in the field of microservices. © Springer Nature Switzerland AG 2019.
  • No Thumbnail Available
    Item
    Priority Based Efficient User Scheduling in Quadcopter Assisted 5G Network
    (Institute of Electrical and Electronics Engineers Inc., 2024) Manohar, S.; Manjappa, M.
    In the days to come the Fifth Generation (5G) communication is expected to replace Long Term Evolution (LTE) technology and provide seamless efficiency in terms of throughput and latency. 5G uses New Radio (NR) standards which makes use of higher frequency bands to achieve enormous data rate. In times of need 5G communication could be assisted by unmanned aerial vehicles (UAV) during emergency situations such as flood or earthquakes to provide temporary mobile communication service. Due to its ease of maneuverability, UAVs that are equipped with 5G antennas can be made to hover over the geographical areas where mobile communication service is required. But due to the power consideration of UAVs, they can take a payload of limited weight, which makes it to carry a small sized 5G antenna. Because of this it can serve less number of users leaving behind those users who actually need communication the most. And hence there is a need for scheduling prioritized users so that they get the emergency services first. The prioritized users may be the first responders during a disaster or the one who is running a high priority application. A priority based scheduling algorithm which is built on the existing Round Robin MAC scheduler is proposed in this article. More resource blocks (RB) are allotted to prioritized users so that they can send more data with low latency. Simulation is done to calculate throughput and latency which proves that the proposed priority based scheduling gives better performance when compared to the aforementioned existing scheduler. © 2024 IEEE.
  • No Thumbnail Available
    Item
    Implementation and performance comparison of queuing disciplines for Best-Effort and Time-Sensitive Network
    (Institute of Electrical and Electronics Engineers Inc., 2024) Hagargund, A.G.; Shet, N.S.V.; Kulkarni, M.
    Real-time applications like industrial automation, autonomous vehicles and critical infrastructure systems require deterministic networking strategies to provide efficient data delivery. The networking protocols play an important role in such applications where data transmission of devices is done using TCP/IP based networks. In this study we concentrate on Layer 2 standards of TCP/IP stack to test and validate the scheduling strategies. We examine and compare the performance of legacy scheduling strategies like Round Robin, First In First Out, Token Bucket with Time-sensitive networking standard like IEEE 802.1Qbv. We present the result of all the scheduling strategies and discuss the suitability of these standards for real time and time critical applications. We use Mininet emulator and Linux queuing disciples to validate the standards. The results for delay is discussed for each case. © 2024 IEEE.
  • No Thumbnail Available
    Item
    Adaptive Workload Management for Enhanced Function Performance in Serverless Computing
    (Association for Computing Machinery, Inc, 2025) Birajdar, P.A.; Harsha, V.; Satpathy, A.; Addya, S.K.
    Serverless computing streamlines application deployment by removing the need for infrastructure management, but fluctuating workloads make resource allocation challenging. To solve this, we propose an adaptive workload manager that intelligently balances workloads, optimizes resource use, and adapts to changes with auto-scaling, ensuring efficient and reliable serverless performance. Preliminary experiments demonstrate an ≈ 0.6X% and 2X% improvement in execution time and resource utilization compared to the First-Come-First-Serve (FCFS) scheduling algorithm. © 2025 Copyright held by the owner/author(s).