IEEE 2019

Published on Jan 09, 2020

Myself Arun (Black jacket) and Dr Santosh Kumar (Grey jacket) had the opportunity to represent TonkaBI at the 18th International Conference on Information Technology organized by IIIT Bhubaneswar. It was conducted from Dec 19 – 21, under IEEE.

It was a really great experience, we had the opportunity to interacted with fellow industry members and research scholars. This fantastic event gave us the opportunity to share our ideas and the challenges we face while doing our respective research in depth.

We had the privilege to attend several keynote sessions delivered by highly distinguished persons from the industry and academy. The sessions were very knowledgeable and there were a lot of concepts that we could take away to aide in our future research.

With Dr. Ruchi Saxena (Director of Operations, Caerobotics Healthcare Solutions, UK)

Dr. Ruchi Saxena took a session on Drones and data for disasters which was very informative.

As our paper was accepted by IEEE, we had to do a paper presentation in front of the chair. The chair needed to approve our paper before it would be published in IEEE Xplore.

The chairs for my presentation consisted of Dr. Vo Thi Ngoc Chau(Ho Chi Minh City University of Technology, Vietnam) and Dr. Satyananda Champati Rai (Silicon Institute of Technology).

The paper titled ‘Development of Novel Ensemble Machine Learning Architecture for Forecasting Unresponsive Server State’ discusses about the problem wherein large production clusters there comes a point where the server gets overloaded during peak hours and becomes unresponsive. When the server reaches such a state, the only workaround is to force restart the server and thus in the process lose critical data from currently executing applications at that time, as these tasks would be abruptly stopped. Our paper discusses the various machine learning techniques which can be used to predict when the server could reach this unresponsive state, and if so, using a decision support system, kill some tasks which are having low priority and thus ensuring uninterrupted running of the server.

The paper presentation went really well and it was appreciated by both my chairs.

Dr. Vo Thi Ngoc Chau was especially interested to implement the idea in her university. She had also encountered the problem statement many times, the solution to which I discussed in detail in my presentation.

The paper was approved by both the chairs, and will be published in IEEE Xplore very soon.

Collecting Certificate of Appreciation from Dr. Vo Thi Ngoc Chau

In addition to representing TonkaBI at ICIT 2019, we got the chance to speak about Deep learning and AI and also talk about TonkaBI, our vision and the opportunities in the company. It was a very good interaction with the students there, they found the session very interesting and really appreciated the work which we were doing.

With College principal, chairman and student heads.

We are very glad to have received this opportunity to have our paper appreciated and accepted for publishing in IEEE Xplore. Once the paper is published and open, we hope people who face similar problems would find the paper helpful. Also looking to implement the paper in a full fledged production cluster very soon.

Posted in blog on Jan 09, 2020