Please enable JavaScript.

Coggle requires JavaScript to display documents.

AWS Cloud (Pricing and Support (TrustedAdvisor (Checks: Cost Optimization,…

- - - - Internal or External
  - - - CPU, Memory, Network Performance
    - - Magnetic, General purpose SSD, provisioned IOPS
    - - mySQL, Amazon Aurora, SQL Server, Oracle, MariaDB, Postgre SQL
    - - Able to configure so that a standby copy of the DB is replicated to another AZ on another subnet in your VPC
        
        Synchronous replication
        
        If master fails, AWS automatically loads standby copy. Don't need to change anything because url is same
    - - Asynchronous
      - Offload read queries from master DB
      - Can be manually promoted to master, but requires intervention
      - For Disaster recovery or apps with high read needs
  - - - Changes to an S3 bucket can trigger, respond to HTTP requests, respond to API calls
    - - NodeJS, Java, Python
    - - disc space 512 MB
      - max 5 minute execution
      - Constrained by deploy pkg size/max file descriptors
      - Memory allocation 128-1528 MB
      - request & response body payload <= 6MB
      - Event request body <= 128 kb
      - Limit to the # of concurrent executions which can be increased if requested
    - - Automated Backups
      - Processing objects uploaded to S3
      - Event-driven log analysis
      - Event-driven transformations
      - Internet of Things
      - Operating serverless websites
      - ETL Pipeline
  - - - instance type
      - database
      - auto-scaling
      - Access server log files
      - Enable HTTPS on load balancer
  - - - Set security - who can send, who subscribe?
  - - - CPU Utilization
      - Status checks (failure)
      - Page view count
      - Several services provide free metrics (EC2, EBS, RDS)
        
        Can publish own app metrics for additional fee
    - - Monitors logs for specific values, patterns
      - Includes installable agent
      - Monitor logs from Ec2 in real time
      - Monitor CloudTrail logged events
      - Archive log data
      - Admins view logs in Mgt Console
        
        Even resources across different regions
    - - Email or SMS
      - Monitor single metric
      - Sustained state changes only
    - - Perform actions based on threshold or trend over time periods
        
        EC2 action
        
        Auto scaling
        
        Notification to SNS
    - - Automatically invoke Lambda function to update DNS entries when an event notifies that EC2 instance enters running state
      - Direct specific API records from CloudTrail to a Kinesis stream for analysis
      - Take a snapshot of an EBS volume on schedule
      - Log S3 object-level events
    - - Fixed rate
      - CRON expression
    - - Customized views
      - Create dashboards using console, CLI, or PutDashboard API
  - - - Save money due to lower rates on data transfer out
      - Make customers happier
        
        cache content close to them
    - - Speed up delivery of static and dynamic content .css, .html, .php, image files
      - Live streaming
        
        RTMP used for live streaming Adoble Flash Media
      - Static asset cacheing
      - Security and DDOS protection
      - API acceleration
  - - - JSON or YAML
      - Parameters and conditions allow you to use same template to create different environments
      - Infrastructure as Code
      - If template has an error, then CloudFormation will stop and roll back by default
    - - Use the "DependsOn" parameter to specify order of creation
      - Whoever is using template must have permissions to all the resources in the template
- - - - Terminal window
        
        Copy to S3: aws s3 cp <filename> <s3 arn>
        
        Sync: aws s3 sync <folder> <s3 arl>
        
        Can sync in reverse from Amazon to local server
        
        Copy from S3: aws s3 cp <s3 arn>
  - - - Redundantly stored across multiple AWS facilities in region
  - - - Can be configured to support cross-region replication
- - - - Auto Scaling Group
        
        Where deployment takes place
        
        VPC/Subnets; load balancer, min/max instances; desired capacity
      - Auto Scaling Policy
        
        When to launch new EC2 instances
        
        Scheduled; On demand; scale out and scale in policy
      - Launch Configuration
        
        Amazon Machine Image AMI, instance type, security group used when new EC2 launched
- - - - Edge Locations - host Amazon CloudFront - CDN content delivery network
- - - - Security Design Principles:
        Security at all layers
        Enable trace-ability
        Principle of least privilege
        secure your system
        automate
    - - Foundation
        
        handle changes in demand/requirements
        
        detect failure
        
        heal itself
      - change mgt
        
        Understand and be aware of how change affects system
      - Reliability design
        
        Test recovery procedures
        
        Automatically recover
        
        Scale horizontally
        
        Stop guessing capacity
        
        manage change in automation
    - - Performance Efficiency Design
        
        Democratize advanced technologies - push the knowledge into cloud vendor's domain if you don't have the skillset
        
        Going global in minutes
        
        Use a serverless architecture
        
        Experiment more often
        
        Have mechanical sympathy - use the technologocial approach that best aligns with the goal you're trying to achieve
    - - Cost Optimization design principles
        
        Consumption model - pay for what you use
        
        Measure overall efficiency
        
        Reduce spending on data center operations
        
        Analyze and attribute expenditure
        
        Use managed services