This page describes the high-level architecture of Appian Process Mining including how Process Mining fits into Appian Cloud and how your data flows through Process Mining components.

Cloud architecture

The following diagram describes a Process Mining installation in Appian Cloud.

Security

Appian deploys your Process Mining installations as a single tenant, so your cloud data remains independent from other customers. As shown in the preceding diagram, Appian Process Mining isolates customers in their own namespace, following Kubernetes best practices.

Your data is kept secure using the HTTPS protocol with TLS encryption for all communication between Appian Process Mining and the browser. All traffic is also routed through a web application firewall (WAF).

Appian Process Mining uses encrypted storage volumes to encrypt data at rest.

An instance of Appian manages the Kubernetes sites in Appian Cloud within a virtual private cloud (VPC) which includes access control to nodes and clusters.

URL naming convention

Appian Cloud Process Mining installations use the following naming conventions for site URLs:

• Mining Prep: <sitename>.appianmining.com/mining-prep. For example, appian.appianmining.com/mining-prep.
• Process Mining: <sitename>.appianmining.com. For example, appian.appianmining.com.

Platform architecture

The following diagram describes the high-level architecture of Process Mining modules for both self-managed and Appian Cloud installations.

Access

You sign into Mining Prep and Process Mining from a web browser securely with HTTPS. Login credentials are shared between both modules. The Process Mining module handles identity and access management (IAM) for both Mining Prep and Process Mining modules in conformance with the OpenID Connection (OIDC) specification.

The ingress server supports SSL and routes external traffic to the different Process Mining services.

To learn more about Process Mining security, see Security Considerations.

Data sources

Appian Process Mining supports connections to a variety of data sources. You can upload a CSV file or connect to another data source like a database or Enterprise Resource Planning (ERP) application. Source systems are not available publicly for Appian Cloud installations of Mining Prep but are supported for self-managed installations.

Standard connectors for the following systems come with the platform:

• Records
• PostgreSQL
• MariaDB
• MSSQL
• Oracle

Access to external data sources uses Open Database Connectivity (ODBC).

Meta-data for Process Mining and Mining Prep is stored separately from table, event, and user data.

• In Mining Prep, meta-data refers to information about tables like transformation steps.
• In Process Mining, meta-data refers to information about event logs, connected dashboards, models, filters, settings, and the connected user groups.

Additionally, Process Mining meta-data can be modified directly from the API management service. Other requests are routed through the message queue to the Process Mining engine.

Engines

Appian's custom process mining engines handle most of the tasks and data coordination behind the scenes.

• The Mining Prep engine handles API requests, data transformations, data extraction, and scheduling operations.
• The Process Mining engine handles model generation, conformance checking, automated root cause analysis (ARCA), and data aggregation operations.

Connectivity between modules

Once you've transformed data sets into event logs in Mining Prep, these logs are loaded into Process Mining over HTTPS. The API and resource management middleware routes logs you upload from Mining Prep or logs you upload directly into the Process Mining module.

To learn more about connecting Mining Prep and Process Mining modules, see Connect Mining Prep and Process Mining.