Well Architected Framework

 

Operational Excellence

• Ability to run and monitor systems
• Continuously improve supporting processes and procedures
• How code will be deployed, updated, operated
• Logging, reduce defects, perform quick, safe fixes

1. manage and automate changes
2. respond to events
3. successfully manage daily operations

Best practice areas

Prepare -> Operate -> Evolve

Perform Operations as code

• Define your entire workload - applications and infrastructure as Code
• Update it with code
• Implement operational procedures as code
• Limit human error
• Enable consistent responses to events
• Annotations are input to your operations as code

Automate documentation

Create annotated documentation after every build

Make frequent small changes

Enable components to be updated regularly

Refine operation procedures frequently

Identify opportunities to improve

Anticipate failure

Identify potential sources of failure so they can be mitigated

Learn from all operational events and failures

Share across teams and the organization

Security

• protect informations, systems, assets, data, network and compute resources
• identify who can do what
• applies to both confidentiality and integrity
• establish control to detect security events
• protect data in transit and at rest
• classify data into security levels
• tools to identify and investigate security incidents
• enable traceability

Best practice areas: Security, IAM, Detection, Infrastructure & Data protection, Incident response

Design principles

Implement a strong identity foundation

• principle of least privilege
• reduce reliance on long-term credentials

Enable traceability

• monitor and alert actions, changes to your environment in real time
• integrate logs and metrics with systems
• automatically respond and take actions

Apply security at all layers

• edge network, VPC, subnet, load balancer

Automate security best practices

• create secure architectures
• version controlled templates

Protect data in transit and at rest

• classify data into security levels
• encryption, tokenization

Keep people away from data

• reduce chances of human error
• eliminate and reduce direct access or manual processing of data

Prepare for security events

• have incident management process that aligns with organizational requirements
• run incident response simulations
• use tools with automation

Cost Optimization

• Avoid unnecessary costs
• Understand where money is spend
• Select appropriate resources
• Analyze spending trends over time
• Scale if needed without overspending
• iterative process, improved during the lifetime
• consider using managed services where you can
• decommissioning resources
• using correct pricing models to reduce cost

Best Practice Areas

Cloud financial management

Expenditure and usage awareness

Cost-effective resources

Managing demand and supply

Optimize over time

Design Principles

Implement Cloud Financial Management

• use knowledge building, tools, processes programs resources to optimize cloud costs

Adopt a consumption model

• Pay only for what you require
• Increase or decrease usage based on business need

Measure overall efficiency

• Measure business output vs associated costs of the workload

Stop spending money on undifferentiated heavy lifting

• do not focus on building IT infrastructure, let AWS do that

Analyze and attribute expenditure

• attribute cost and system usage to specific workload owners
• measure ROI
• workload owners can optimize and reduce cost

Performance Efficiency

• use resources sparingly
• use IT and compute resources efficiently
• select right resource types and sizes
• monitor performance
• meet efficiency as demand changes and technologies evolve
• gather data on all aspects and use it to select and configure resources
• review your choices periodically
• take advantage of new services
• monitoring for deviance of expected performance
• use tradeoffs to improve performance: compression, relaxed consistency requirements, caching

Best practice areas

• Selection
• Review
• Monitoring
• Tradeofss

Design Principles

Democratize technologies

• Expertise in technologies such as NoSQL databases, machine learning is not available evenly across community, but these are available as services and can be consumed in the cloud

Go global in minutes

• Deploy systems in multiple regions

Use server less architectures

• lower transactional costs
• easier to operate
• managed services at cloud scale

Experiment more often

• comparative testing of different types of instances , storage, configurations

Mechanical sympathy

• use approach that aligns best
• example: consider data access patterns when you select a database technology

Sustainability

• maximize efficiency
• reduce waste
• energy reduction
• minimizing total resources required
• selection of efficient programming language
• adoption of modern algorithms
• efficient data storage technique
• minimize requirements for high powered hardware
• anticipate adoption of new more efficient hardware and software
• understanding your impact

Reliability

• recover from failure
• mitigate disruption
• design distributed systems
• recovery planning
• recover quickly from disruptions
• managing service quota
• data backups

Best practice areas

• Foundations, Workload architecture (Well-planned)
• Change management, Failure management (Monitoring)

Design Principles

Automatically recover from failure

• monitor key performance indicators
• configure systems to trigger automatic recovery when threshold is breached
• automatic notification for failures

Test recovery procedures

• test systems failure
• validate recovery procedures

Scale horizontally to increase aggregate workload availability

• replace one large resource with multiple smaller resources
• distribute requests
• minimize impact of single point of failure

Stop guessing capacity

• monitor demand and system usage
• automate addition and removal of resources

Manage change in automation

• use automation to make changes to infrastructure

AWS Well-Architected Tool

• compares your workloads to the state of the latest AWS best practices in terms of architecture
• delivers step by step action plan for improvement

How it works

1. You define your workload
2. You answer a series of questions in the six areas