-
Solutions
-
VaultDB HUB
Train, Test and deploy and manage your models with integrated
Governance, Risk and Controls. Control data your models are training on for
Bias and compliance. -
VaultDB Compute
Train and Run your models across cloud providers on cheapest and
unlimited resources. -
VaultDB Actions (DevOps)
Train and Run your models across cloud providers on cheapest and
unlimited resources.
-
-
Developers
- About Us
- Contact Us
Train, Test and deploy and manage your models with
integrated Governance, Risk and Controls. Control data your
models are training on for Bias and compliance.
Train and Run your models across cloud providers on cheapest
and unlimited resources.
Train and Run your models across cloud providers on cheapest
and unlimited resources.
Distributed and AI-infused Database
The capability to “Commit, Push, and Merge” and AI capabilities
VaultDB collects data in many distributed devices and streams to the central cloud database through VaultDB HUB.
A centralized database server is a bottleneck in a distributed computing environment and does not scale. VaultDB removes this bottleneck by being highly distributed and optimizing it for AI and IoT devices. Feature Engineering and AI capabilities are bolted onto VaultDB.
IoT Database
Lite database for lite devices
Today’s business applications and Internet of Things (IoT) devices generate massive amounts of real-time data, which must be collected, processed, and stored using the limited hardware available in IoT devices.
Secure Database
Security and Data Privacy from the ground up
VaultDB is built with security and data privacy in mind. Its capability of GRC (governance-risk-compliance) ensures security, including authentication and authorization, data encryption, secure data transfer protocols, regular updates and patches, and strict data governance and access control policies.
Start With Us
How it Works
Applications running on IoTs can use the VaultDB database as a library. VaultDB collects data locally and then merges it into the central database running in the cloud. It natively supports GRC and feature engineering.
Database
GRC and Controls
Feature Engineering and AI
Frequently Asked Questions
What is VaultDB?
VaultDB is an in-process database that allows any process to utilize local computing capability for query and data processing instead of relying on a centralized database server. VaultDB and VaultDB HUB are optimized for distributed cloud processes and IoT devices.
Each process uses the VaultDB library (Python, NodeJS, ODBC, JDBC) for data access and processing and, once ready, pushes data to VaultDB HUB.
VaultDB has built-in GRC capability that allows users to secure data and access. In short, VaultDB provides an in-process database with a centralized VaultDB HUB that unlocks data management and AI capabilities in IoT devices.
VaultDB is an ideal tool for ETL (extract-transform-load). It supports different input formats like CSV, Parquet, and JSON. Vectorization is used to support faster data processing.
How much does VaultDB Cost?
VaultDB database software is free to use. You pay only for storage and computes to your cloud provider.
Why should I consider using VaultDB for my IoT
devices?
IoT devices generate vast amounts of real-time data, which is invaluable for organizations and individuals. This data is used for real-time decision-making and analytical and AI purposes.
VaultDB is perfect for this use case as it provides local writes when the device is disconnected and can push data to VaultDB HUB.
IoT devices can also use VaultDB to query merged local and centralized data with low latency.
VaultDB is highly cost-effective. You only have to bear storage costs, as the server does not run all the time.
Why should I consider using VaultDB for Model training testing?
Machine Learning Model training and testing requires lot of data preparation, data engineering and security.
VaultDB is built on DuckDB which is built for big Data analytics and data engineering.
VaultDB Python package is integrated with Pandas and Numpy, and this makes data processing multiple folds faster due to Vectorization.
VaultDB also supports PyArrow and other extensions supported by DuckDB.
VaultDB's GRC capability such as Fortress, Column level security, and Roles
help in preventing model bias and protect PI information.
VaultDB allows partitioning tables for faster processing of data reducing computing cost.
What is Data Inheritance?
Local VaultDB inheritence works like a database server of merged data from remote and local database instances. This helps in testing and other use-cases.
Why VaultDB?
Here's why you should use VaultDB:
1) Written in C++, resulting in high speeds
2) Built-in support to read and write Parquet, CSV, JSON files
3) Integration with Pandas, Numpy, PyArrow
4) Supports all DuckDB extensions
5) Works on Windows, Linux, and Mac
6) Can run on IOT/laptop/desktop and cloud for everything
7) Low maintenance costs
8) Allows fast development simulating Production environment
