Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/akumuli/Akumuli
Time-series database
https://github.com/akumuli/Akumuli
c-plus-plus database metrics time-series time-series-database tsdb
Last synced: 13 days ago
JSON representation
Time-series database
- Host: GitHub
- URL: https://github.com/akumuli/Akumuli
- Owner: akumuli
- License: apache-2.0
- Created: 2014-01-28T21:31:20.000Z (almost 11 years ago)
- Default Branch: master
- Last Pushed: 2022-08-07T01:52:15.000Z (over 2 years ago)
- Last Synced: 2024-07-31T22:52:42.535Z (3 months ago)
- Topics: c-plus-plus, database, metrics, time-series, time-series-database, tsdb
- Language: C++
- Homepage: http://akumuli.org
- Size: 5.1 MB
- Stars: 836
- Watchers: 44
- Forks: 85
- Open Issues: 63
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
- awesome-repositories - akumuli/Akumuli - Time-series database (C++)
- awesome-dataops - Akumuli - Can be used to capture, store and process time-series data in real-time. (Database / Time Series Database)
- awesome-data-engineering - Akumuli - Akumuli is a numeric time-series database. It can be used to capture, store and process time-series data in real-time. The word "akumuli" can be translated from esperanto as "accumulate". (Databases)
README
README [](https://travis-ci.org/akumuli/Akumuli) [](https://scan.coverity.com/projects/akumuli)
======
**Akumuli** is a time-series database for modern hardware.
It can be used to capture, store and process time-series data in real-time.
The word "akumuli" can be translated from Esperanto as "accumulate".
Features
-------
* Column-oriented storage.
* Based on novel [LSM and B+tree hybrid datastructure](http://akumuli.org/akumuli/2017/04/29/nbplustree/) with multiversion concurrency control (no concurrency bugs, parallel writes, optimized for SSD and NVMe).
* Supports both metrics and events.
* Fast and effecient compression algorithm that outperforms 'Gorilla' time-series compression.
* Crash safety and recovery.
* Fast aggregation without pre-configured rollups or materialized views.
* Many queries can be executed without decompressing the data.
* Compressed in-memory storage for recent data.
* Can be used as a server application or embedded library.
* Simple API based on JSON and HTTP.
* Fast range scans and joins, read speed doesn't depend on database cardinality.
* Fast data ingestion:
* 5.4M writes/sec on DigitalOcean droplet with 8-cores 32GB of RAM (using only 6 cores)
* 4.6M writes/sec on DigitalOcean droplet with 8-cores 32GB of RAM (6 cores with enabled WAL)
* 16.1M writes/sec on 32-core Intel Xeon E5-2680 v2 (c3.8xlarge EC2 instance).
* Queries are executed lazily. Query results are produced as long as client reads them.
* Compression algorithm and input parsers are fuzz-tested on every code change.
* Grafana [datasource](https://github.com/akumuli/akumuli-datasource) plugin.
* Fast and compact inverted index for time-series lookup.
Roadmap
------
|Storage engine features |Current version|Future versions|
|-------------------------------|---------------|---------------|
|Inserts |In order |Out of order |
|Updates |- |+ |
|Deletes |- |+ |
|MVCC |+ |+ |
|Compression |+ |+ |
|Tags |+ |+ |
|High-throughput ingestion |+ |+ |
|High cardinality |+ |+ |
|Crash recovery |+ |+ |
|Incremental backup |- |+ |
|Clustering |- |+ |
|Replication |- |+ |
|ARM support |+ |+ |
|Windows support |- |+ |
|Query language features |Current version|Future versions|
|-------------------------------|---------------|---------------|
|Range scans |+ |+ |
|Merge series |+ |+ |
|Aggregate series |+ |+ |
|Merge & aggregate |+ |+ |
|Group-aggregate |+ |+ |
|Group-aggregate & merge |+ |+ |
|Join |+ |+ |
|Join & merge |- |+ |
|Join & group-aggregate |- |+ |
|Join & group-aggregate & merge |- |+ |
|Filter by value |+ |+ |
|Filter & group-aggregate |+ |+ |
|Filter & join |+ |+ |
Gettings Started
----------------
* You can find [documentation](https://akumuli.gitbook.io/docs) here
* [Installation & build instructions](https://akumuli.gitbook.io/docs/getting-started)
* [Getting started guide](https://akumuli.gitbook.io/docs/getting-started)
* [Writing data](https://akumuli.gitbook.io/docs/writing-data)
Supported Platforms
-------------------
Akumuli supports 64 and 32-bit Intel processors. It also works on 64 and 32-bit ARM processors but these architectures are not covered by continous integration.
Pre-built [Debian/RPM packages](https://packagecloud.io/Lazin/Akumuli) for the following platforms
are available via packagecloud:
* AMD 64 Ubuntu 14.04
* AMD 64 Ubuntu 16.04
* AMD 64 Ubuntu 18.04
* AMD 64 Debian Jessie
* AMD 64 Debian Stretch
* AMD 64 CentOS 7
* ARM 64 Ubuntu 16.04
* ARM 64 Ubuntu 18.04
* ARM 64 CentOS 7
Docker image is availabe through [Docker Hub](https://hub.docker.com/r/akumuli/akumuli/tags/).
Tools for monitoring
--------------------
Akumuli supports OpenTSDB telnet-style API for writing. This means that many collectors works with it
without any trouble, for instance `netdata`, `collectd`, and `tcollector`. Grafana
[datasource](https://github.com/akumuli/akumuli-datasource) plugin is availabe as well.
Akumuli can be used as a long-term storage for Prometheus using [akumuli-prometheus-adapter](https://github.com/akumuli/akumuli-prometheus-adapter).
[Google group](https://groups.google.com/forum/#!forum/akumuli)