guillotina cache implementation using redis + lru in-memory cache
Project description
guillotina_rediscache implements redis into guillotina with an additional in-memory layer cache.
In order to coordinate invalidating the in-memory cache, guillotina_rediscache utilizes the pub/sub feature redis provides.
Configuration
app_settings for this:
{
"databases": {
"db": {
...
"cache_strategy": "redis"
...
}
},
"redis": {
'host': 'localhost',
'port': 6379,
'ttl': 3600,
'memory_cache_size': 1000,
'pool': {
'minsize': 5,
'maxsize': 100
}
}
}
Run measures
Using guillotina run command:
./bin/g run --script=measures/serialize.py -c measures/config.yaml
With profiling:
./bin/g run --script=measures/serialize.py -c measures/config.yaml --line-profiler --line-profiler-matcher="*serialize*"
1.3.0 (2018-03-19)
Provide redis file manager [vangheem]
1.2.0 (2018-03-14)
Upgrade to work with guillotina 2.4.x [vangheem]
1.1.6 (2018-03-01)
Handle errors while canceling task init task [vangheem]
1.1.5 (2018-03-01)
Handle errors while canceling task [vangheem]
1.1.4 (2018-02-15)
Fix cache stats endpoint [vangheem]
1.1.3 (2018-01-22)
Be able to disable deleting group of cache keys together with cluster_mode option. [vangheem]
1.1.2 (2018-01-17)
Fix using redis.delete when keys are of length 0 [vangheem]
1.1.1 (2018-01-17)
batch all cache deletes into one request [vangheem]
1.1.0 (2018-01-12)
Push cache updates to redis subscriber. This should improve cache hits dramatically [vangheem]
1.0.14 (2018-01-10)
Only run invalidation task if we have keys to invalidate [vangheem]
1.0.13 (2017-12-15)
Improve request performance [vangheem]
Change the way we’re using the redis pool so it reuses connections [vangheem]
1.0.12 (2017-11-30)
Missing await statement for self.get_redis() [vangheem]
1.0.11 (2017-11-08)
Handle CancelledError [vangheem]
1.0.10 (2017-11-06)
upgrade for guillotina 2.0.0 [vangheem]
1.0.9 (2017-10-23)
Fix handling connection objects and releasing back to pool [vangheem]
1.0.8 (2017-10-23)
Fix use of pool [vangheem]
1.0.7 (2017-10-23)
Use pickle instead of json from load/dumps because it is much faster [vangheem]
1.0.6 (2017-10-19)
Use ujson [vangheem]
1.0.5 (2017-10-02)
Track all keys needing invalidation and do invalidation in an async task so the request can finish faster. [vangheem]
1.0.4 (2017-05-29)
Test fixes [vangheem]
1.0.3 (2017-05-26)
Fix delete not properly invalidating cache [vangheem]
1.0.2 (2017-05-15)
Fix channel publishing invalidations [vangheem]
1.0.1 (2017-05-15)
Fix release
1.0.0 (2017-05-15)
initial release
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for guillotina_rediscache-1.3.0.tar.gz
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | ddf376a400689d2c4d48ebc74b267621f96df67d2c393c7cdba2a643f2f3553a |
|
| MD5 | c2df94bb2854cdfc831d29b667132cbf |
|
| BLAKE2b-256 | 46a66afdfad6592f965a2a7f65391e048d74826307f3e30a2d5d3c598c64beb3 |
Hashes for guillotina_rediscache-1.3.0-py3-none-any.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 0339148997383f0c67583205f54fbf1d1ba2a288ef841be784a655740c3c098b |
|
| MD5 | ba71e18472cb64469e35d92b7315d327 |
|
| BLAKE2b-256 | 77d40687213760dfcbe77507c02539299f27dbe360e7469157c8dc231b995f84 |
