Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal The Eastasouth Journal of Information System and Computer Science
Shubh Prabhat
Architect, GlobalLogic Inc., Cary, North Carolina, USA
Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Event-Driven Decomposition Strategies for Migrating Monolithic Enterprise Platforms to Resilient Microservice Architectures Shubh Prabhat
The Eastasouth Journal of Information System and Computer Science Vol. 1 No. 02 (2023): The Eastasouth Journal of Information System and Computer Science (ESISCS)
Publisher : Eastasouth Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58812/esiscs.v1i02.1309

Abstract

Enterprise platforms are often operated in large monoliths. They are moved to microservices by teams to achieve scale, quicker releases and isolation of failures. Most of these moves do not provide resilience since services make long synchronous chains among themselves. The paper suggests an event-based decomposition plan to such migrations. There are five phases of the strategy. They consist of domain mapping, cut scoring, event contract design, strangler extraction and fault injection testing. The work is directed by two scores. A cut score is measured to assess the effectiveness of a service in retaining its calls and data within the boundary of the service. An event conversion ratio calculates the number of cross service interactions that consume events as compared to direct call. Design decisions are related to resilience through a path availability equation and an event lag equation. The test is carried out on a virtual order management system with 1.4 million code lines and 236 database tables. The site is divided into two services into eight parts. One of them is synchronous calls. The remaining uses the suggested strategy. The foundation of a monolith is the basis. The event driven version supports 1,000 orders/minute with a p95 latency of 360 milliseconds compared to 655 milliseconds in the synchronous version. The one service outage completes 91 percent against 47 percent of orders. The recovery time decreases from 22 to 9 minutes. The variations also decrease from 3.9 to 1.6 per 10,000 orders. After release, one of eight service boundaries had to change. The event driven version takes 38 person months compared to 31. The cuts should be influenced by event design.
Graph-Based Dependency Intelligence for Predictive Failure Analysis in Large-Scale Distributed Systems Shubh Prabhat
The Eastasouth Journal of Information System and Computer Science Vol. 2 No. 02 (2024): The Eastasouth Journal of Information System and Computer Science (ESISCS)
Publisher : Eastasouth Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58812/esiscs.v2i02.1310

Abstract

Distributed systems with many services fail via the distributed processes of dependent services, and most monitoring systems do not respond to problems until the first symptom occur. This article introduces a graph-based dependency intelligence architecture, which is used to predict failures that are going to impact users. The framework constructs a weighted service dependency graph using traces, metrics and logs, and trains the graph in each time window, and uses a graph neural network to make a prediction about the likelihood of each service failing within the next fifteen minutes. The edge weight is defined by three equations, the message passing rule and the probability of failure and a fourth equation of the evaluation score. The paper provides a survey of the literature on microservice faults, root cause localization and graph learning, and a comparison of six recent diagnosis studies with the prediction gap. A benchmark of 41 services and 126 dependency edges on a cluster of 12 nodes was utilised in the experiment. The team infused 300 failure instances between four classes over 30 days and evaluated the models on 120 unknown instances. The proposed model achieved precision that was equal to 0.89, a recall of 0.87 and an F1 score that was 0.88. A model that was a static graph model achieved 0.79, service level LSTM reached 0.69 and threshold alerting reached 0.56. The suggested model provided a mean warning lead time of 8.7 minutes, reduced the false alarms to 2.1 per day and the true root service was ranked in top three ranked services in 81 percent of episodes. Findings indicate that the dynamic dependency weights provide predictive power over and above the service level signals.
Co-Authors