Please enable JavaScript.
Coggle requires JavaScript to display documents.
Data-intensive applications (Foundations of data systems (Scalability…
Data-intensive applications
Foundations of data systems
Reliability
Tolerate to errors
Hardware faults (network, ram, disc)
Random and independent
Software faults
Bugs
Corelation across nodes
To avoid
Tests
Monitorings
Process isolation
Human errors
Wrong configuration
Corrupted input
To aviod
Easy rollback
Validate inputs
Well designed systems
Sandbox
Scalability
Positivity to deal with high load
Load
Reads from db per second
Number of active users
Requests per second
Usually measures in persentels
Scaling up (vertical)
Scaling out
Share nothing arcgitecture
Elastic systems
Can automatically scale
Maintanability
Productively make changes on system
Data systems (databases and queues)
Data models
Relational
Document
Contains all hierarchy of object
Many to many implements with document refs
Schemales
Schema on read
All data in one locality
Graph
Storage and retrieval