01
Business needs
DigiValet, an experience platform and smart solutions provider for luxury hotels, branded residences, and reputed hospitals, wanted to foray into a niche vertical. They were looking to build a platform to support tailor-made requirements for managing premium residential/commercial and short/long stay residential properties. The hospitality experience provider was looking for a robust, scalable, performant, and flexible futuristic cloud solution that would:
- Help DigiValet gain early-bird advantage with faster time-to-market
- Facilitate easier onboarding of brands and their properties
- Support existing third-party service providers used by the properties
- Feature a ML-based analytics and prediction for personalization
- Support geographies not accessible to public cloud service providers
02
Technical needs
Since faster time-to-market was one of the critical requirements, the Impetus team architected a modular solution on the AWS cloud powered by diverse AWS Services. The team used Angular with a modularization approach for building a responsive web application, ensuring content optimization for mobile devices.
DigiValet had some specific technical needs for the solution, like:
- Highly configurable to accommodate brand-specific configurations
- Flexible data schema to support a variety of data inputs
- Backend application and platform designed to support white-labelling with brand-specific configuration
- Ability to capture information for multiple verticals from multiple data sources like MongoDB cluster for transactional and booking data, Google Analytics for mobile and web click stream events, and automation log files
- Support for i18n and p13n
03
Solution
Impetus architected and deployed a cost-effective, scalable, multi-tenant, and customizable AWS-based platform. Built on a secure serverless architecture, the hybrid platform could also be implemented on-premises to support geographies not accessible to public cloud service providers. While the platform’s back end was based on a multi-tenant, single-schema architecture, the team took a microservices-based approach to separate out business functions.
Solution highlights:
- Used GraphQL to reduce network bandwidth, optimize response payload, and improve response time
- Used Cognito as the identity manager for the application
- Used PySpark jobs running on AWS Glue to transform raw data
- Used PySpark to transform raw data and upload it to a data warehouse
- Analytics microservice that will pull data from the data warehouse
- Customized widgets at the front-end portal for a single pane view of analytics
- Automated CI-CD for seamless production updates with zero downtime
- To make PII data GDPR compliant,
- Encrypted data-at-rest
- Secured HTTPS data connection for end-to-end security of data-in-transit
- Masked PII data
- Ensured data storage on the desired region as per GDPR compliance
- Enabled auditing of events on cloud infrastructure
AWS stacks used
AWS Web Application Firewall (WAF), Amazon CloudFront, Amazon Simple Storage Service (S3), Amazon Cognito, Amazon API Gateway, AWS Lambda, Amazon Virtual Private Cloud (VPC), AWS Key Management Services (KMS), AWS Secrets Manager, AWS Identity and Access Management (IAM), AWS Glue, Amazon Redshift, Amazon CloudTrail, Amazon CloudWatch, AWS X-Ray, Amazon Social Networking Service (SNS), Amazon Simple Queue Service (SQS), Amazon Simple Email Service (SES)
04
Impact
- Onboarded 3 leading hospitality brands on the platform within a year of launch
- Bagged one of the largest passive house projects in the world
- ~80% reduction in deployment time with thoughtful use of technologies and streamlined processes
- Automated platform to onboard new brands in 1 business day with standard specifications and properties
- 99.99% system availability with support for further scalability
- Platform meets stringent data residency requirements of the European Union (EU), wherein most of the potential business is expected
