Geek Sync | Designing Data Intensive Cloud Native Applications
Download as PPTX, PDF1 like213 views
The document presents insights on designing data-intensive cloud-native applications, focusing on the transition from monolithic architectures to microservices. It discusses various topics such as cloud migration strategies, polyglot persistence, and the importance of resilience and scalability in microservices. Additionally, it covers key architectural patterns including event-driven architecture, the circuit breaker design pattern, and serverless databases for improved efficiency.
Geek Sync | Designing Data Intensive Cloud Native Applications
- 1. Designing Data Intensive Cloud Native Applications July 10th, 2019 Samir Behara, System Architect, EBSCO
- 2. SAMIR BEHARA System Architect, EBSCO Samir Behara builds software solutions using cutting edge technologies. Has a Bachelor Degree in Computer Science with 13 years of IT experience. Frequent Speaker at Technical Conferences. Author of www.dotnetvibes.com
- 3. • Monolith vs Microservices • Design Principles of Microservice Architecture • How to break a Monolithic Database • Cloud Migration Strategy • Cloud Native Architecture • Polyglot Persistence • Database Scaling – Horizontal and Vertical • Event Driven Architecture • Circuit Breaker Design Pattern • Serverless Database
- 4. Monolithic Architecture Order Management Service Monolithic Database Large Codebase Difficult to Scale Longer Development Cycle Complicated Deployments Fixed Technology stack Performance Issues Tight Coupling
- 5. Applying Domain Driven Design
- 6. Monolith Architecture – Centralized Database
- 7. Databases are private to each Microservice
- 8. Emergence of Microservices Shorter Development Cycle Faster Deployments Highly Scalable Right Technology Stack Business Domain Driven Resiliency & Observability High Cohesion & Loose Coupling
- 9. Order Service Title Service Customer Service Monolithic Legacy Application Order Service Customer Service Title Service Monolithic to Service Oriented to Microservices Journey Shared Database Independent DatabasesMonolithic Database
- 11. Cloud Native Architecture DevOps Microservices Containers Continuous Delivery Cloud Native is a combination of various patterns like DevOps, CI/CD, Containers, Orchestration, Microservices and Cloud Infrastructure
- 12. Cloud Migration Strategy Migrate Modernize Lift and Shift No Code Changes required Fast Deployment Can’t leverage the true potential of Cloud Expensive on long run Containers and Microservice Architecture Requires application refactoring Time Consuming Leverage benefits of Cloud Native Cost Optimized solution in Cloud
- 13. 6 R’s of Cloud Migration RETIRE Time to Migrate --- > CostandEffort---> RETAIN REHOST REPLATFORM REFACTOR REPURCHASE
- 14. Polyglot Persistence Service A Service B Service C Service D Service E Service F Service G Relational Time Series Document Key-Value Graph Search In-Memory Leverage multiple data storage technologies to handle the varied requirements of your services
- 15. Polyglot Persistence Selecting the Right Database technology for the workload
- 16. CAP Theorem Consistency Availability Partition Tolerance A distributed database system can only have 2 of the 3 properties – • Consistency • Availability • Partition Tolerance
- 18. Primary DB Instance Read Replicas Handling Read-Heavy Workloads using Read Replicas Read Write Read Read Read Application Auto Scaling Replication Lag Load Balancer
- 19. Multi AZ Deployments for High Availability Availability Zone 1 Availability Zone 2 Standby ReplicaPrimary Instance Data Replication
- 20. Master Instance Read Replica Read Replica Read Replica Read Replica Read Replica Availability Zone 1 Availability Zone 2 Availability Zone 3 Multi AZ Deployments for Read Replicas
- 21. Cust A-D Cust E-J Cust K-N Cust O-S Cust T-Z Database Sharding based on Range Strategy Database Sharding Shards
- 22. Shard A Database Sharding based on Hash Strategy Shard B Shard C Hash Strategy Order Service Hash_Key % 3
- 23. Caching to Improve Database Performance Order Service CACHE 1. Is data in Cache? 2. If Yes, Read it 3. If No, fetch from DB, Save in Cache and Read it • Keep frequently accessed data in Memory • Memory is faster than disk • Reduced load on Database
- 29. Circuit Breaker Design Pattern Design for Failures Handle Failures gracefully Avoid Cascading failures Avoid single point of failure Design Resilient and Fault Tolerant Microservices
- 30. Event Driven Architecture Tight Coupling between components in your architecture Loose Coupling between components for better performance
- 31. Design for Scalability Order Service Web Server Order Service Order Service Order Service Order Service Web Server
- 32. Evolution of Serverless Architecture Moving from legacy monolithic on-premise infrastructure to microservices based in public cloud
- 33. Serverless Database • Resolves capacity issues with Over Provisioning or Under Provisioning of Databases • Serverless Computing for your databases • Do not pay for unused resources • Automatically starts the database, scales based on usage and shuts down when not in use.
- 34. Use Cases of Serverless Databases Test Databases Variable Workloads New Database Development Infrequently Used Databases Cost Effective Solution Unpredictable Workloads
- 35. Serverless Database in action Seamlessly Scale capacity up or down based on application usage Automatic Start up and Shut Down