Conference Papers

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

Browse

Author: search.filters.author.Addya, S.K.Subject: search.filters.subject.Cloud Computing

Search Results

Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    MatchCloud: Service Matching for Multi Cloud Marketplace
    (Institute of Electrical and Electronics Engineers Inc., 2021) Chakma, A.; Kumar, S.; Mahato, P.K.; Satpathy, A.; Addya, S.K.
    The modern applications execute in the cloud via independent executable entities called virtual machines (VMs). In a typical multi-SP market with variable pricing and heterogeneous resource demands of VMs, resource allocation/placement is particularly challenging. To maximize the social welfare of the multi-SP markets, in this paper, we propose a resource allocation technique called MatchCloud formulated as a one-to-many matching game. Owing to the in-applicability of the classical deferred acceptance algorithm (DAA) due to size heterogeneity, we adopt a modified version of the algorithm. Moreover, preference generation is crucial for matching markets. Hence, we also present a simple yet efficient technique to assign preferences to two different stakeholders, i.e., VMs and SPs. Simulation results show that VM proposing RDA performs better compared to when SPs propose. © 2021 IEEE.
  • No Thumbnail Available
    Item
    Container-based Service State Management in Cloud Computing
    (Institute of Electrical and Electronics Engineers Inc., 2021) Nath, S.B.; Addya, S.K.; Chakraborty, S.; Ghosh, S.K.
    In a cloud data center, the client requests are catered by placing the services in its servers. Such services are deployed through a sandboxing platform to ensure proper isolation among services from different users. Due to the lightweight nature, containers have become increasingly popular to support such sandboxing. However, for supporting effective and efficient data center resource usage with minimum resource footprints, improving the containers' consolidation ratio is significant for the cloud service providers. Towards this end, in this paper, we propose an exciting direction to significantly boost up the consolidation ratio of a data-center environment by effectively managing the containers' states. We observe that many cloud-based application services are event-triggered, so they remain inactive unless some external service request comes. We exploit the fact that the containers remain in an idle state when the underlying service is not active, and thus such idle containers can be checkpointed unless an external service request comes. However, the challenge here is to design an efficient mechanism such that an idle container can be resumed quickly to prevent the loss of the application's quality of service (QoS). We have implemented the system, and the evaluation is performed in Amazon Elastic Compute Cloud. The experimental results have shown that the proposed algorithm can manage the containers' states, ensuring the increase of consolidation ratio. © 2021 IFIP.
  • No Thumbnail Available
    Item
    Performance Analysis of Disruptive Instances in Cloud Environment
    (Institute of Electrical and Electronics Engineers Inc., 2024) Nandy, P.; Saha, R.; Satpathy, A.; Chakraborty, S.; Addya, S.K.
    Virtualization enables the service providers (SPs) to logically partition the resources into virtual machines (VM) instances. Real-world SPs such as Amazon, Google, Microsoft Azure, IBM, and Oracle provide different flavors of VM instances, such as on-demand, reserved, and low-cost or spot, depending on the type of application hosted. The on-demand instances are short-term and typically incur a higher cost than reserved instances that are provisioned for a longer duration at a discounted rate. Low-cost or spot instances are cost-effective compared to on-demand but are reclaimable by the SPs. The SPs often claim that the on-demand and spot instances achieve similar performance, but it is far from that. This paper studies the performance of spot instances via rigorous experimentation over commercial SPs such as Amazon AWS and Microsoft Azure. Real-world evaluations affirm that spot instances perform poorly compared to their on-demand counterpart concerning memory, CPU, disk read, and write operations. We identify such instances as disruptive and name them so because it does not fulfill the performance, durability, and flexibility expectations like an on-demand instance having the same configuration. We also perform hypothesis testing over the experimental data obtained to corroborate our claim further. © 2024 IEEE.