Browsing by Author "Meena, D."

Now showing 1 - 2 of 2

DCRDA: deadline-constrained function scheduling in serverless-cloud platform
(Springer, 2025) Birajdar, P.A.; Meena, D.; Satpathy, A.; Addya, S.K.
The serverless computing model frees developers from operational and management tasks, allowing them to focus solely on business logic. This paper addresses the computationally challenging function-container-virtual machine (VM) scheduling problem, especially under stringent deadline constraints. We propose a two-stage holistic scheduling framework called DCRDA targeting deadline-constrained function scheduling. In the first stage, the function-to-container scheduling is modeled as a one-to-one matching game and solved using the classical Deferred Acceptance Algorithm (DAA). The second stage addresses the container-to-VM assignment, modeled as a many-to-one matching problem, and solved using a variant of the DAA, the Revised-Deferred Acceptance Algorithm (RDA), to account for heterogeneous resource demands. Since matching-based strategies require agent preferences, a Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) ranking mechanism is employed to prioritize alternatives based on execution time, deadlines, and resource demands. The primary goal of DCRDA is to maximize the success ratio (SR), defined as the ratio of functions executed within the deadline to the total functions. Extensive test-bed validations over commercial providers such as Amazon EC2 show that the proposed framework significantly improves the success ratio compared to baseline approaches. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.
Employing Differentiable Neural Computers for Image Captioning and Neural Machine Translation
(Elsevier B.V., 2020) Sharma, R.; Kumar, A.; Meena, D.; Pushp, S.
In the history of artificial neural networks, LSTMs have proved to be a high-performance architecture at sequential data learning. Although LSTMs are remarkable in learning sequential data but are limited in their ability to learn long-term dependencies and representation of certain data structures because of the lack of external memory. In this paper, we tackled two main tasks, one is language translation and other is image captioning. We approached the problem of language translation by leveraging the capabilities of the recently developed DNC architectures. Here we modified the DNC architecture by including dual neural controllers instead of one and an external memory module. Inside our controller, we employed a neural network with memory-augmentation which differs from the original differentiable neural computer, we implemented a dual controller's system in which one controller is for encoding the query sequence whereas another controller is for decoding the translated sequences. During the encoding cycle, new inputs are read and the memory is updated accordingly. In the decoding cycle, the memory is protected from any writing from the decoding controller. Thus, the decoder phase generates a translated sequence at a time step. Therefore, the proposed dual controller neural network with memory-augmentation is then trained and tested on the Europarl dataset. For the image captioning task, our architecture is inspired by an end-to-end image captioning model where CNN's output is passed to RNN as input only once and the RNN generates words depending on the input. We trained our DNC captioning model on 2015 MSCOCO dataset. In the end, we compared and shows the superiority of our architecture as compared to conventionally used LSTM and NTM architectures. Â© 2020 The Authors. Published by Elsevier B.V.