# db

* [Couchbase vs CouchDB](#couchbase-vs-couchdb)
* [API Query](#api-query)
* [key-value](#key-value)
* [Time Series DBs](#time-series-dbs)
* [TimescaleDB](#timescaledb)
* [Riak TS](#riak-ts)
* [Firebase](#firebase)
* [Supabase](#supabase)

## Couchbase vs CouchDB

<https://www.couchbase.com/couchbase-vs-couchdb>

|| Couchbase Server|Apache CouchDB| |---|---| |Topology|Distributed|Replicated| |Automatic failover|Yes|No| |Integrated cache|Yes|No| |Memcached compatible|Yes|No| |Query language|Yes, N1QL (SQL for JSON)|No|

## API Query

<http://graphql.org/>

## key-value

<https://github.com/dgraph-io/badger>

## Time Series DBs

Why relational database instead of NoSQL: <https://blog.timescale.com/time-series-data-why-and-how-to-use-a-relational-database-instead-of-nosql-d0cd6975e87c>

Elastic, InfluxDB, MongoDB, Cassandra, Couchbase, Graphite, Prometheus, ClickHouse, OpenTSDB, DalmatinerDB, KairosDB, RiakTS.

## TimescaleDB

<https://github.com/timescale/timescaledb\\>
packaged as a PostgreSQL extension

<http://docs.timescale.com/v0.8/getting-started/installation/linux/installation-docker>

```
docker run -d --name timescaledb -p 5432:5432 timescale/timescaledb
```

<https://blog.timescale.com/when-boring-is-awesome-building-a-scalable-time-series-database-on-postgresql-2900ea453ee2>

<http://docs.timescale.com/v0.8/getting-started/creating-hypertables>

```
CREATE EXTENSION IF NOT EXISTS timescaledb CASCADE;

SELECT create_hypertable('conditions', 'time');
-- backgroud:
CREATE INDEX conditions_time_idx
    ON public.conditions USING btree
    ("time" DESC)
    TABLESPACE pg_default;

-- additionally partition the data on another
--   dimension (what we call 'space partitioning').
-- E.g., to partition `location` into 4 partitions:
SELECT create_hypertable('conditions', 'time', 'location', 4);

SELECT time_bucket('5 minutes', time) AS time_range,
    location, COUNT(*),
    MAX(temperature) AS max_temp,
    MAX(humidity) AS max_hum
FROM conditions
WHERE time > NOW() - interval '3 hours'
GROUP BY time_range, location
ORDER BY time_range DESC, max_temp DESC;

    time_range       | location | count | max_temp | max_hum 
------------------------+----------+-------+----------+---------
2018-02-23 17:00:00+00 | office   |     3 |       70 |      50
2018-02-23 16:35:00+00 | garage   |     1 |       77 |    65.2
2018-02-23 16:35:00+00 | office   |     2 |     70.1 |    50.1
2018-02-23 16:35:00+00 | basement |     1 |     66.5 |      60
2018-02-23 16:25:00+00 | office   |     1 |       70 |      50
(5 rows)
```

## Riak TS

<http://docs.basho.com/riak/ts/\\>
Riak TS is a distributed NoSQL key/value store optimized for time series data.

## Firebase

Realtime NoSQL: <https://firebase.google.com/pricing>

## Supabase

Firebase alternative, PostgreSQL

* OSS: <https://github.com/supabase/supabase>
* <https://supabase.com/pricing>