ArcticDB is a high performance, serverless DataFrame database built for the Python Data Science ecosystem.
Launched in March 2023, it is the successor to Arctic.
Use of ArcticDB in production (including business or commercial environments) or for a Database Service requires a paid for license from ArcticDB Limited . Please contact [email protected] for further details.
ArcticDB offers an intuitive Python-centric API enabling you to read and write Pandas DataFrames to S3 or LMDB utilising a fast C++ data-processing and compression engine.
ArcticDB allows you to:
- Pandas in, Pandas out: Read and write Pandas DataFrames, NumPy arrays and native types to S3 and LMDB without leaving Python.
- Built for time-series data: Efficiently index and query time-series data across billions of rows
- Time travel: Travel back in time to see previous versions of your data and create customizable snapshots of the database
- Schemaless Database: Append, update and modify data without being constrained by the existing schema
- Optimised for streaming data: Built in support for efficient sparse data storage
- Powerful processing: Filter, aggregate and create new columns on-the-fly with a Pandas-like syntax
- C++ efficiency: Accelerate analytics though concurrency in the C++ data-processing engine
ArcticDB handles data that is big in both row count and column count, so a 20-year history of more than 400,000 unique securities can be stored in a single symbol. Each symbol is maintained as a separate entity with no shared data which means ArcticDB can scale horizontally across symbols, maximising the performance potential of your compute, storage and network.
ArcticDB is designed from the outset to be resilient; there is no single point of failure, and persistent data structures in the storage mean that once a version of a symbol has been written, it can never be corrupted by subsequent updates. Pulling compressed data directly from storage to the client means that there is no server to overload, so your data is always available when you need it.
Quickstart
Prebuilt binary availability (for the current version)
Storage compatibility
| Linux | Windows | Mac |
|---|
| S3 | ✔️ | ✔️ | ✔️ |
| LMDB | ✔️ | ✔️ | ✔️ |
| Azure Blob Storage | ✔️ | ✔️ | ✔️ |
We have tested against the following S3 backends:
- AWS S3
- Ceph
- MinIO on Linux
- Pure Storage S3
- Scality S3
- VAST Data S3
Installation
Install ArcticDB:
$ pip install arcticdb
or using conda-forge
$ conda install -c conda-forge arcticdb
Import ArcticDB:
>>> import arcticdb as adb
Create an instance on your S3 storage (with or without explicit credentials):
# Leave AWS to derive credential information
>>> ac = adb.Arctic('s3://MY_ENDPOINT:MY_BUCKET?aws_auth=true')
# Manually specify creds
>>> ac = adb.Arctic('s3://MY_ENDPOINT:MY_BUCKET?region=YOUR_REGION&access=ABCD&secret=DCBA')
Or create an instance on your local disk:
>>> ac = adb.Arctic("lmdb:///<path>")
Create your first library and list the libraries in the instance:
>>> ac.create_library('travel_data')
>>> ac.list_libraries()
Create a test dataframe:
>>> import numpy as np
>>> import pandas as pd
>>> NUM_COLUMNS=10
>>> NUM_ROWS=100_000
>>> df = pd.DataFrame(np.random.randint(0,100,size=(NUM_ROWS, NUM_COLUMNS)), columns=[f"COL_{i}" for i in range(NUM_COLUMNS)], index=pd.date_range('2000', periods=NUM_ROWS, freq='h'))
Get the library, write some data to it, and read it back:
>>> lib = ac['travel_data']
>>> lib.write("my_data", df)
>>> data = lib.read("my_data")
To find out more about working with data, visit our docs
Documentation
The source code for the ArcticDB docs are located in the docs folder, and are hosted at docs.arcticdb.io.
License
ArcticDB is released under a Business Source License 1.1 (BSL)
BSL features are free to use and the source code is available, but users may not use ArcticDB for production use or for
a Database Service, without agreement with Man Group Operations Limited.
Use of ArcticDB in production or for a Database Service requires a paid for license from ArcticDB Limited
and is licensed under the ArcticDB Software License Agreement. For more information please contact [email protected].
This is the beginning of the README. Read the rest at the original. Read the full README ↗
