The Fastest Time Series Object Storage for AI Infrastructure
High-performance time series object storage, written in Rust, designed for speed, simplicity and robustness. Handle data of any size - images, files, text, vibration data and more. Store billions of blobs with time indexing and AI labels.
Simplify Your Infrastructure
Merge blob and time series functionalities, reducing the need for multiple databases.
Stay In Control Of Your Data
Customize real-time data retention policies and replication strategies.
Handle Large Data Volumes
Store billions of time-stamped blobs with AI labels and access them with low latency.
Get the Best Performance
Outperform other databases with a customized solution for time-series object data.
- vs TimescaleDB
- vs MongoDB
- vs MinIO
Record Size | Read Speed (%) | Write Speed (%) |
---|---|---|
1 MB | +671% | +1604% |
100 KB | +603% | +924% |
10 KB | +313% | +297% |
1 KB | +28% | +198% |
Time Series Blob Storage
Capture and access blob data as time series, tailored for edge computing, computer vision, and IoT.
No Size Limit for Blobs
ReductStore handles blob data without size limits; your disk capacity is the only boundary.
Real-Time FIFO Quota
Ensure optimal storage management with FIFO quotas, preventing disk space shortages in real-time.
Data Labeling & Filtering
Manage your time-series blob data with ease: annotate, filter, and save AI labels or meta-data.
Advanced HTTP(S) API
Integrate and communicate with ReductStore using our feature-rich and secure API.
Efficient Data Batching
Minimize network overhead in areas with high latency by fetching records in batched HTTP responses.
Data Replication
Synchronize data across buckets with replication for high availability and disaster recovery.
Iterative Data Querying
Efficiently queries large datasets with minimal load for real-time and historical data processing.
Token Authorization
Secure data access with token-based authorization to protect your data from unauthorized access.
- Python
- JavaScript
- C++
- Rust
- cURL
import time
import asyncio
from reduct import Client, Bucket
async def main():
client = Client('http://127.0.0.1:8383')
bucket: Bucket = await client.create_bucket("my-bucket", exist_ok=True)
ts = time.time_ns() / 1000
await bucket.write("entry-1", b"Hey!!", ts)
async with bucket.read("entry-1", ts) as record:
data = await record.read_all()
print(data)
loop = asyncio.get_event_loop()
loop.run_until_complete(main())
Web Console
ReductStore has an integrated web console that allows you to easily manage your data and access to it.
CLI Client
You can customize data retention and replication policies using the ReductStore CLI client.
Typical Use Cases
The Fastest Time Series Object Store for Edge AI
Learn how to store and manage data for edge computing and AI applications.
High Frequency Vibration Data Storage
Learn how to store and manage high-frequency vibration data.
Computer Vision Applications
Learn how to store data for computer vision applications.
Frequently Asked Questions
ReductStore is built for scalability, designed to efficiently manage large data volumes typical in edge computing and AI/ML workflows. It combines blob and time-series data functionalities, enabling the system to handle billions of time-stamped blobs. Scaling up primarily involves expanding disk capacity. The system's architecture supports low-latency access with iterator and range query capabilities, ensuring optimal performance at scale. More information about the architecture can be found in the How Does It Work section of the documentation.
Yes, ReductStore includes several customizable features for managing data, one of which is the First In, First Out (FIFO) approach. This method is specifically designed to handle data based on volume intervals, ensuring that as new entries are added, older ones are systematically removed to prevent disk space issues. Additionally, AI labels can be attached to blobs as metadata, helping in the identification and replication of key data, thus ensuring vital information is retained. These functionalities are all accessible through the ReductStore CLI, making the management of data retention and replication straightforward and effective.
Yes, ReductStore ensures compatibility with a variety of programming languages via its HTTP(S) API. Being a blob storage system, it supports the storage of all types of unstructured data in byte form, making it straightforward to incorporate into your existing setups and manage a wide range of data.
The Fastest Time Series Object Store for AI Infrastructure
Discover why ReductStore is the top choice for AI and edge computing applications.
Download our White Paper