Quicksight

Supported data sources:
- Redshift
- Aurora
- Athena
- RDS
  - myslq,postgres..., mysql, sql server (no oracle)
- on prem
  - mysql ,postgres, mssql.
- from files (on prem or s3)
  - csv, excel, log format..
Standard edition VS Enterprise edition
- enterprise has:
  - Active directory integration
  - encryption at rest
with standard ed, inviting:
- an IAM user
- invite by email, which creates quicksight only account.
- automatic notigfication of access
with enterprise can also:
- use microsoft AD groups in AWS Directorsy Service
- no automatic notification of quicksight access (for security reasons)
all data in transit is encrypted, but at rest only in enterprise
SPICE
- super-fast, parallel in-memory calculation engine
- measured in GB
- 10GB per user
- Hihghly available and durabilty (for data imported to SPICE)
Data preparation:
- change field names
- calculated filed
- SQL query
- join tables -change data types
Steps:
- From a data source, create a dataset
- create an analysis
  - can create a dashboard of an analysis (can't save filters)

VisualTypes:

- Autograph (automatically selected by quicksight)
- Bar chart 
    - single measure for one dimension
    - two or more

Story:

- like a slideshow,

Dashboard:

- read-only snapshot of an analysis (viewer can filter but not save filters)

Big data visualization

notes simplified
Zeppelin
- Zeppelin similar to jupyter notebook.
- Zeppelin can be selected from EMR app list
- can share notebooks (and real-time collaboration like google docs)
- use cases:
  - use with spark sql and mlib, for exploratory data analsysis
Jupyter notebook:
- can contain live code, can share notebooks
- used for data sci.
- install in EMR using bootstrap action