CONFIGURE STORE WORK SCHEDULE AS A SERVERLESS SYSTEM
Customer Is a large Japanese enterprise that run a convenience store chain with a total of over 5000 stores.
MINISTOP Co., Ltd.
01. Customer conditions
・ Customer want to manage the work schedule management of each customer’s store in common
・ Since there is a lot of processed data, it is necessary to process 3GB to 5GB of data every month.
・ Infrequent use of services
* Monthly job schedule is used main before and after the deadline. Average Using 3 times a day
02. Our Solutions
・ Utilization mainly for managed services
・ “API Gateway, Lambda, RDS” combination to process daily job schedules.
・ Using AWS Elastic Container Service to process the data due to lots of processing data for the monthly job schedule
・ By using serverless construction instead of EC2 (virtual server), operating cost was reduced by 50%
・ Daily ・ Monthly job schedule is automatically executed in AWS CloudWatch Event
Monitoring / Auditing:
・ Perform regular monitoring with AWS CloudWatch Logs
04. Development team
・ Japan Onsite: 4 MM
・ Offshore: 20 MM
・ Development period: 4 months (from requirement definition to release)
B2B Mobile application
B2C mobile application, serving 40,000 users in first phase and millions in later phase
Using microservices architecture leveraging Kubernetes on AWS for nearly unlimited scalability
EKS, Aurora DB, Prometheus, Grafana (monitoring), Splunk (logging)