Well Architected Framework WP - Performance Efficiency Flashcards
Performance Efficiency Pillar
Efficient use of computing to meet requirements and maintain that efficiency as demand and technology change
Design Principles
Democratize advanced technologies. ie AWS lets anyone use dynamoDB without being a NoSQL expert
Go global in minutes
User server less architectures
Experiment more often
Mechanical Sympathy
Performance Efficiency Definition
2017 whitepaper 4 areas that compose Performance Efficiency Selection Review Monitoring Trade-Offs
===== 2016 whitepaper compute storage database space-time trade off
Compute
Key AWS Service for elastic compute solutions
AWS lets you change type of server, or go server-less
Instances - default
Containers - improve utilization
Functions - for event driven or parallel tasks
Key service is auto-scaling to match supply and demand
Compute questions
How do you select appropriate instance type
how do you ensure you continue to have most appropriate instance type as new ones introduced
how do you monitor instances
how do you ensure quantity of instance matches demand
Storage best practices
Optimal storage solution depends on several factors:
access method (block, file, object) random or sequential access throughput requirements frequency of access and updates available and durability constraints
Storage Characteristics to consider
/
2016 WhitePaperStorage questions
Ability to share, file size, cache size, latency, throughput, persistence of data. Match these to appropriate AWS storage: S3, Glacier, EBS, EFS, Instance Store
=====
2016 white paper storage questions
how do you select appropriate storage solutions
how do you ensure continue to have most appropriate storage solution
how do you monitor storage for performance
how do you ensure capacity and throughput matches demand
Databases - questions
Consider using different database services for different types of data
=======
2016 Whitepaper
How do you select appropriate DB solution for your system
How do you ensure you continue to have most appropriate DB solution as new systems evolve
How do you ensure capacity and throughput matches demand
space time tradeoff
use services like RDS to add read replicas
Use direct connect for predictable latency
use global infrastructure to put copies of environment close to customers
space time trade off questions
how do you select right proximity and caching solutions
how do you ensure you continue to have most appropriate proximity and caching solutions
how do you monitor proximity and caching solutions and make sure they match demand
Key AWS services
Compute: AutoScaling
storage: EBS, S3, Glacier
Database: RDS, DynamoDB, Redshift
Space-time: CloudFront, ElastiCache, direct connect, read replicas
exam tips - 4 areas for performance efficiency
compute
storage
databases
space time tradeoff
exam tips
review the questions for each category
Key AWS Services for Storage
S3
also:
EBS, EFS, EC2 Instance Store, Glacier
