Company overview
Leading insurance company of india was launched in june 2008 and is jointly owned by two of the leading public sector banks. the company has access to around 115 million customers and a panindia network of over 9500 branches.
Problem Statement
Customer has ISNP application rearchitected as microservices based.
the main challenge was to containerise the application.
also, there would be weekly new features addition, bug fixes etc. hence ci/cd automation was also required.
also, due to variable load, auto-scaling was required in order to scale out and scale in resources to meet variable workload and yet control the cost.
Our Solution
Converg-IN TEAM HAS DEPLOYED ISNP PROJECT FOR CUSTOMER’S UAT AND PRODUCTION ENV. BOTH.
THE DEPLOYMENT WAS DONE ON KUBERNETES CLUSTER ON AWS EKS. APPLICATION WAS HOSTED ON THE PODS.
EKS CONTROL PLANE DEPLOYED IN PUBLIC SUBNETS.
THE ENVIRONMENT DEPLOYED IN AWS MUMBAI REGION.
WE CONFIGURED LOAD BALANCER TO PROVIDE ACCESS TO END USERS OVER THE INTERNET.
CONFIGURED THE AUTOSCALING GROUP TO HANDLE THE HIGH LOAD.
EKS WAS USED FOR DOCKER CONTAINERS MANAGEMENT AND ORCHESTRATION.
JENKINS USED FOR PULLING THE CODE, BUILD THE PACKAGE AND DEPLOYING DOCKER CONTAINERS.
PREPARE CLOUD (SUBSCRIPTION, STORAGE, NETWORK)
SET UP THE CONFIGURATION SERVER
TEST DEPLOYMENT
RISK MITIGATION PLANNING.
Converg-IN HAS HELPED CUSTOMER CONTAINERISE THE APPLICATION AND DEPLOYED BLUE-GREEN ARCHITECTURE TO DEVTEST & PRODUCTION.
PROFICIENTLY USING ELASTIC KUBERNETES SERVICES AND HELM CHARTS AUTOMATED CI/CD FOR THIS BUSINESS CRITICAL APPLICATION.
HELM CHARTS CHANGES COMMITTED TO VERSION CONTROL SYSTEM (BITBUCKET) AND DEPLOYMENT PIPELINE CONFIGURED TO USE THE GIT REPOSITORY FOR DEPLOYMENT.
COMMIT HELM CHART. HELM CHARTS CHANGES COMMITTED TO VERSION CONTROL SYSTEM (BITBUCKET) AND DEPLOYMENT PIPELINE CONFIGURED TO USE THE GIT REPOSITORY FOR DEPLOYMENT.
CLUSTER AUTOSCALER NEEDED AS TO HELP CLUSTER AUTO SCALE ON LOAD AND BE ABLE TO CREATE NEW NODES.
METRIC SERVER HELPS GET SUITABLE METRICS FOR AUTOSCALING TO WORK.
UPDATE AUTOSCALING FOR NON PROD CLUSTER HELM CHARTS. AUTOSCALING CONFIGURATION UPDATED TO HELP SYSTEM SCALE ON DEMAND.
UPDATING OF BUILD PIPELINE TO AUTO TRIGGER DEPLOYMENT. BUILD PIPELINE UPDATE TO AUTO TRIGGER DEPLOYMENT ON SUCCESSFUL BUILD. CREATION OF IAM SERVICE PRINCIPLE FOR CLUSTER ACCESS IN AUTOMATION PIPELINE.
SERVICE PRINCIPLE BASED ACCESS TO KUBERNETES CLUSTER AS DEFAULT ACCESS IS ONLY WITH THE USER CREATING THE CLUSTER. ALSO THIS HELPS IN AUTOMATION WITH HELP OF ROLES.
UPDATING PRODUCTION HELM CHARTS AS PER APPLICATION PORTS, LIMITS ETC. HELM CHARTS NEEDED TO BE MODIFIED TO MAKE APPLICATION ACCESSIBLE.
CREATE NAMESPACE IN PRODUCTION CLUSTER. SEGREGATE RESOURCES USING NAMESPACE.
UPDATING OF BUILD PIPELINE TO AUTO TRIGGER DEPLOYMENT.
BUILD PIPELINE UPDATE TO AUTO TRIGGER DEPLOYMENT ON SUCCESSFUL BUILD.
CREATE INGRESS CONTROLLER IN PRODUCTION CLUSTER.
INGRESS CONTROLLER TO LOAD BALANCE THE SERVICES AND CREATE COMMON END POINT FOR BOTH SERVICES.
MAKE INGRESS CONTROLLER AS PRIVATE NLB. MAKING LOAD BALANCER PRIVATE USING VALUES FILES.
CREATION OF DEPLOYMENT PIPELINE FOR PRODUCTION.
COMMIT HELM CHART FOR PRODUCTION. HELM CHARTS CHANGES
COMMITTED TO VERSION CONTROL SYSTEM (BITBUCKET) AND DEPLOYMENT PIPELINE CONFIGURED TO USE THE GIT REPOSITORY FOR DEPLOYMENT.
JENKINS DECLARATIVE DEPLOYMENT PIPELINE WITH HELM AND KUBECTL. AWS INTEGRATION USING SERVICE PRINCIPLE.
UPDATING PROD BUILD PIPELINE TO AUTO TRIGGER DEPLOYMENT. BUILD PIPELINE UPDATE TO AUTO TRIGGER DEPLOYMENT ON SUCCESSFUL BUILD.
INSTALL CLUSTER AUTOSCALER IN PRODUCTION. CLUSTER AUTOSCALER NEEDED AS TO HELP CLUSTER AUTO SCALE ON LOAD AND BE ABLE TO CREATE NEW NODES.
INSTALL METRIC SERVER IN PRODUCTION. METRIC SERVER HELPS GET SUITABLE METRICS FOR AUTOSCALING TO WORK.
UPDATE AUTOSCALING FOR PROD HELM CHARTS. AUTOSCALING CONFIGURATION UPDATED TO HELP SYSTEM SCALE ON DEMAND.
ASSIST WITH API GATEWAY INTEGRATION WITH NLB VIA VPC LINK. API GATEWAY NEED TO BE USED TO ACCESS PRIVATE NLB USING VPC LINK. HENCE KEEPING ORIGIN SECURE AND PROVIDE WAF INTEGRATION.
CREATION OF HELM CHARTS FOR PRODUCTION. HELM CHARTS CREATED FOR PRODUCTION SPECIFIC CHANGES
NEW SUBNET CREATION TO AVOID CLASH WITH EXISTING PROD CLUSTERS. NEW SUBNETS WERE CREATED TO AVOID CLASH WITH ALREADY EXISTING CLUSTERS IN EARLIER SUBNETS.
RECREATE EKS CLUSTER AND DEPLOY THE NODE GROUPS. EKS WAS REACREATED AND ALL ACTIVITIES SET UP AGAIN POST DELETION AND RECREATION OF CLUSTER.
Migration Benefits
- Achieves 99.99% Uptime For Mission-critical Application
- Auto Scale Instances When 60% Threshold Is Exceeded
- Rightsized Instances For Optimal Compute Cost.
- Automatic Backups With Rds.
- Decrease Infrastructure Administrative Cost.
- Receives Technical Support Within Minutes Of Raising The Query.
Outcome/Roadmap ahead
This AWS cloud solution deployment by Converg-IN is driving a lot of cost optimization.
Amazon web services has been around for approximately two decades and it holds a large portion of the cloud infrastructure business. also, an increasing number of tech-savvy businesses and industry leaders are recognising the many benefits of aws cloud computing services.
But more than that, they are using aws for hosting various applications and technology more efficiently to run their organisations, better serve their customers, and dramatically increase their overall efficiency and profit margins.
Driven by flexible cloud capabilities, scalable-infrastructure performance, customer now has better control on infrastructure costs along with business continuity and data availability.
This solution has helped customer towards improving productivity by automating the cicd pipelines using helm-charts, tools like jenkin, gitlab and computing capabilities as per their application requirements.
Converg-IN private limited can also provide 24 x 7 managed support & monitoring provided which drives quarterly optimization on cloud expenditure.
Not sure what fits your business needs?
Get free consultation!
Experience you can trust, service you can count on.
