Industry
Retail
Challenges
• Recurring issues • System stability • Cost and performance tuning
Solution
• Consultation • Establishment of processes • Performance improvements • Cost reduction
Business Value
• 30% cost reduced • Improved performance • Continuous improvement initiative
Technology
• Microservices Dockers • Kubernetes • Erlang • Mongo • PostgreSQL • Elastic Map Reduce • Bamboo
Challenge
Nearbuy is India’s first hyper-local online platform that enables customers and local merchants to discover and engage with each other. Nearbuy is present in 35+ cities, 18+ categories and have over 50,000 merchants across 100,000+ unique locations.
Nearbuy was looking for a partner that can work with them to design and deliver a comprehensive technology strategy, instead of vanilla managed services. The first identified area of action was to improve system reliability and to find optimization opportunities. So far the managed services at Nearbuy were focused mainly on the resolution of urgent issues causing application downtime and productivity loss.
They wanted to implement a preemptive process that would provide better monitoring, prevent issues and reduce downtime.
Needless to say, Nearbuy’s platform involves an amalgamation of cutting edge & complex technology stack, on the cloud with an absolute necessity of high availability. This did not make it easier!
Business value delivered
BluePi and Nearbuy teams worked together to resolve the challenges and lay the foundation of a continuous improvement framework. The engagement started with in-depth workshops and interviews with the stakeholders to identify functional and non-functional requirements. The journey so far has been fruitful and a great learning experience for both the teams. So far we have made significant improvements in overall system stability and performance, Here are a few highlights
-
Improved monitoring capabilities: We started with the establishment of SLAs, and well-defined processes based on industry best practices. This was followed by setting up of a monitoring system for critical resources to build better visibility. Can’t improve what you can’t measure!
A robust monitoring system was deployed which provides complete health statistics of all the resources running on AWS, including the spot instances. The results are accessible through an intuitive UI. The platform also includes customized alerts and notifications, which are necessary for any proactive system.
-
Cost optimization: Deployment of the monitoring system as a first step helped us identify opportunities for improvements. The next step was the optimization of services leading to better performance and also an improvement in ROI.
AWS Spot instances deserve a special mention here because they were a key factor in reducing costs. The cost optimization initiative led to a 30% cost savings month on month.
-
Performance improvement: One of the learnings for the team after monitoring the system for a couple of weeks, was the recurrence of some defects and niggles in the system, which were impacting the system stability. These defects were eating up a lot of time and effort for the team.
The next level of improvement came from process automation that detected and resolved the defects with minimal intervention. This led to a significant reduction in downtime and improvement in the overall stability of the system.
-
Continuous improvement initiative: This is not the end of the journey, both teams are not satisfied with the improvements made so far, so we have set up regular review meetings to monitor and discuss the system performance.
These review meetings are a great help in identifying the next improvement areas and plan for the next steps.