Scalable Real-time Analytics
Bolt’s scalable high performance analytical engine provides a powerful set of deep learning and machine learning models, allowing our customers to analyze data in innovative ways to significantly reduce time to identify and resolve issues.
Univariate Anomaly Detection
Multivariate Anomaly Detection
The open, pluggable architecture supports integration of third party and in-house developed models to allow customers to create flexible solutions. The containerized architecture consisting of models and model orchestration allows for flexible deployment in the cloud or on-premises.
To connect data to the models, intelligent connectors enable quick and easy addition of data sources.
Bolt Intelligent Ingestion Architecture
Scalable: Tested to ingest up to 6 TB of data daily
Extensible: New data sources for ingestion can be added in hours
Pipelined: Allows data to be manipulated and cleansed for analysis
Bolt Analytics Architecture
Flexible: Use CPU’s or GPU’s for training
Manage: Train, Infer, and manage multiple models in parallel
Unique: Time series models that detect and predict on single or multiple variables