This folder contains reading notes for several (old but useful) official documentations and blogs from datastax. It's hard to merge them to one file
Datastax
- CQL
- http://www.datastax.com/dev/blog/cql3-for-cassandra-experts
- Does CQL support dynamic columns / wide rows (2013)
-
Thus, the way to model dynamic cells in CQL is with a compound primary key. For the gory details on things like CompositeType, see my previous post.
-
- Data model
- Storage engine
- Time series
- Basic time series with Cassandra
- server-1-load-20110306, put date in the row key, and only have time in column
- small physical row, no larger than 10MB
- Advanced time series with Cassandra
- use materialized view, for simplying storing integer value, this is useless, for more than of field, this is useful
- meta row for timeline starting points, for query that only have end without a start
- different split factor to share load, the split factor can be different at different time, need meta row
- variable bucket size to avoid big row and sparse row, need meta row as well
- use separated process to handle meta row to avoid race, then coordinate might be needed
- NOTE: I think the security policy of Fedora makes it hard to play with docker ....
- Metric Collection and Storage with Cassandra
- Datastax has OpsCenter for monitoring Cassandra, though it is not open source, and no longer support OSS version of C*
- Getting Started with Time Series Data Modeling
- partition
- reverse order (order by time desc) and let old data expire
- Basic time series with Cassandra
StackOverflow
- http://stackoverflow.com/questions/24949676/difference-between-partition-key-composite-key-and-clustering-key-in-cassandra
- http://stackoverflow.com/questions/15857779/commitlog-and-sstables-in-cassandra-database
- it tells sstable structure
Spark
Other blogs