Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/xobotyi/node-tagged-cache
✍️In-memory, performance-targeted caching module with ttl and tagging support.
https://github.com/xobotyi/node-tagged-cache
cache inmemory inmemory-cache node tagged tagged-cache
Last synced: 1 day ago
JSON representation
✍️In-memory, performance-targeted caching module with ttl and tagging support.
- Host: GitHub
- URL: https://github.com/xobotyi/node-tagged-cache
- Owner: xobotyi
- License: mit
- Created: 2018-12-07T19:11:06.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2023-01-07T05:05:06.000Z (almost 2 years ago)
- Last Synced: 2024-04-24T16:04:46.282Z (8 months ago)
- Topics: cache, inmemory, inmemory-cache, node, tagged, tagged-cache
- Language: TypeScript
- Homepage:
- Size: 1.44 MB
- Stars: 1
- Watchers: 2
- Forks: 0
- Open Issues: 14
-
Metadata Files:
- Readme: README.md
- Funding: .github/FUNDING.yml
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
node-tagged-cache
### Fast NodeJS in-memory caching with tagging
Probably the fastest NodeJS caching system yet.
It is much like memcached or redis but provides you with ability to set tags for cache entries and drop them all together when you need it.
>Note: **this module does not clone stored arrays/objects and operates only with their refs.**
### Performance
As the entries store `node-tagged-cache` uses [Map](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Map) due to it's total overperforming regular objects.
All below performance tests made on Win10, Intel I7 5820K, 16GB DDR4 RAM.
All numbers are OPS (operations per second).
500000 rounds peer test.
storage | set | get | check | delete
--------|-----|-----|-------|-------
obj | 3,846,153 | 12,820,512 | 14,285,714 | 11,111,111
objV2 | 4,237,288 | 13,888,888 | 14,285,714 | 15,151,515
map | 3,816,793 | 55,555,555 | 50,000,000 | 7,246,376
The other thing that appeared to be a real showstopper in terms of performance - `Date.now()` calls, which been used in cache itself and in cache controller. So intermediate timestamp cache was implemented, performance impact of which you can see below.
And, furthermore, below you can see performance comparision with two most popular caching libs i know about.
>Note: node-cache is in `useClones: false,` mode, but cache-base always copies it's values;
module | set | mset | get | mget | has | mhas
-------|-----|------|-----|------|-----|-----
Tagged Cache (w/o timestamp) (w/o tags) | 1,179,245 | 2,463,054 | 1,510,574 | 2,717,391 | 1,533,742 | 2,392,344
Tagged Cache (w/o timestamp) | 1,302,083 | 2,392,344 | 1,146,788 | 1,298,701 | 1,070,663 | 1,388,888
Tagged Cache (w/o tags) | 2,631,578 | 3,333,333 | 5,208,333 | 2,906,976 | 5,494,505 | 3,355,704
Tagged Cache | 2,976,190 | 3,378,378 | 1,845,018 | 1,436,781 | 1,976,284 | 1,529,051
node-cache | 284,900 | none | 1,312,335 | 1,098,901 | none | none
cache-base | 507,614 | none | 892,857 | none | 5,319,148 | none
You can simply run performance tests on your own machine.
```bash
node benchmarks/storage.js
node benchmarks/cache.js
```
### Install
```bash
npm i node-tagged-cache
```
### Usage
```javascript
import TaggedCache, {enableTimestampCache, disableTimestampCache} from "node-tagged-cache";
// default module's export is already created instance of cache with default options
// {
// defaultTTL: 600000, // 10m
// cleanupInterval: 60000, // 1m
// }
enableTimestampCache(); // this is not straight requirement but needed for performance improvement
TaggedCache.mset({
foo: "bar",
baz: "bax"
},
3600000, // 1h
["awesomeTag", "moreAwesomeTag"]);
TaggedCache.mget(["foo", "baz"]); // { foo: "bar", baz: "bax" }
TaggedCache.tags.drop("awesomeTag");
TaggedCache.mhas(["foo", "baz"]); // { foo: false, baz: false }
disableTimestampCache(); // before the script should exit;
```