Architecture

Analytics Hub

Timebase is an analytics hub that allows you to consume data from multiple sources, blend it, normalize it, perform calculations on it, and then store the results within time and model context. We call this the data transformation process, and Timebase is the platform that allows you to manage that transformation pipeline.

Timebase is the one place, a single-source-of-truth, where data is:

Unified
Accessible
Transformed
Trustworthy

As data streams into Timebase and is transformed by the pipeline, it immediately becomes available for presentation via reports and dashboards and publishing out to other systems requiring consolidation and calculation capability.

‍

Build a solid analytics foundation in 5 steps:

Model - structured, hierarchical representation of your assets, KPIs and more
Connect - consume data from multiple input streams (including realtime, timeseries, relational, and manual entry)
Transform - point normalization, period normalization, cleansing, classification, contextualizing, and calculation services
Visualize - browser based reporting, charting and dashboarding for description, diagnostic and predictive analytics
Bridge - analytics-ready information published to or pulled by your enterprise tools (ERP, ML, BI, messaging, etc.)

Edge to Enterprise

Scalability

When required, Timebase Systems can be distributed across multiple servers, virtual or physical. When installed on a server, the Timebase Bootstrap allows other components to be deployed to the server. In a sense, the Timebase Bootstrap is the communication bus between components running a Timebase System. This distributed architecture allows for large Timebase Systems to be scaled.

Edge

Running the Timebase Bootstrap on an edge device extends the Timebase System beyond its network boundary. As long as an outbound connection can be made from the edge device to the Timebase System, Data Sources, Data Consumers, and Message Services can be deployed as extensions to that system. For example, a Timebase System running in private cloud infrastructure would use this edge architecture to access on-premise Data Sources, Data Consumers and Message Services.

Tiering / Rollup

Timebase Systems can be tiered together to transfer information from lower level Timebase Systems to higher level Timebase Systems. A typical use-case is Timebase Systems at each production site feeding a single HQ Timebase System. Selected KPIs from Sites A, B, and C automatically propagate up to HQ as soon as they become available (typically a few minutes), presenting HQ-level analytics in near real-time!

Typical Architectures

Timebase is a modular system that can be deployed in a variety of architectures (as below) or in a combination of these.

1. Simple

Single on-premise Timebase Server including Microsoft SQL Server
Connected to multiple data sources including PLCs, Historians, and SQL databases
Workstations running Timebase Config and Visualization via browsers

‍

2. Deployed Data Sources

Connected to deployed data sources to ensure collection is local to source
Deployed Data Sources require the Timebase Bootstrap to be installed on the respective servers

‍

3. Distributed

Multiple Timebase Servers to distribute computing load
Dedicated Microsoft SQL Server

‍

4. Edge and Cloud

Timebase Server deployed to cloud provider (AWS, Azure, etc.)
Data Sources deployed to the edge (on-premise) using a single outbound connection

‍

5. Enterprise

Enterprise rollup from multiple Tier 1 (Site) Timebase Systems to one or more Tier 2 (Enterprise) Timebase Systems
Ideal for cross-site benchmarking, comparison, and planning
Central model template management

‍

Timebase is your Analytics Foundation