Skip to content

Installation

pip install blissdata

Packages: PyPI and Conda

Repository: Gitlab

Redis setup

Blissdata needs a running Redis database with:

  • Redis server 7
  • ReJson extension
  • RediSearch extension

The easiest way is to use Redis Stack distribution which include Redis Server and the most common extensions. 

docker run -p 6379:6379 redis/redis-stack:7.2.0-v5

Redis server configuration

The following options must be set on Redis before running blissdata:

maxmemory 4gb # or the amount you want
maxmemory-policy noeviction # memory management is done by memory tracker
save "" # disable file backup

This can be done either by editing redis.conf file, or from the command line.

Blissdata index initialization

Blissdata needs to configure the database once at redis server start. This is mostly for RediSearch to know what to index. 

from blissdata.redis_engine.store import DataStore

_ = DataStore("redis://localhost:6379", init_db=True)

Memory tracker

Blissdata relies on a memory tracker to clear space when necessary. This is a daemon process that monitor the database.

For default configuration, run: 

memory_tracker --redis-url redis://localhost:6379
2023-10-31 15:23:16,158 INFO: INIT | Redis max. memory: 1GB
2023-10-31 15:23:16,158 INFO: INIT | Mem. usage to trigger cleanup: 80% (819.2MB)
2023-10-31 15:23:16,158 INFO: INIT | Protected history length: 180 seconds
2023-10-31 15:23:16,159 INFO: Memory usage: 1.857MB