Life Sciences

ii High Performance Computing Lens AWS Well-Architected Framework November 2018

• Cover
• Introduction
• Definitions
• General Design Principles
• Scenarios
• Loosely Coupled, High-Throughput Computing
• Batch-Based Architecture
• Queue-Based Architecture
• Traditional Cluster Environment
• Serverless
• Tightly Coupled, High-Performance Computing
• Persistent Cluster
• Ephemeral Cluster
• HPC Microservices
• The Pillars of the Well-Architected Framework
• Operational Excellence Pillar
• Design Principles
• Definition
• Best Practices
• Prepare
• Operate
• Evolve
• Security Pillar
• Design Principles
• Definition
• Best Practices
• Identity and Access Management
• Detective Controls
• Infrastructure Protection
• Data Protection
• Incident Response
• Reliability Pillar
• Design Principles
• Definition
• Best Practices
• Foundations
• Change Management
• Failure Management
• Performance Efficiency Pillar
• Design Principles
• Definition
• Best Practices
• Selection
• Compute
• Storage
• Networking
• Review
• Monitoring
• Trade-offs
• Cost Optimization Pillar
• Design Principles
• Definition
• Best Practices
• Cost-Effective Resources
• Matching Supply and Demand
• Expenditure Awareness
• Optimizing Over Time
• Conclusion
• Contributors
• Further Reading