Appian Architecture

Introduction

The Appian Business Process Management Suite (BPMS) is built using a multi-tier web application model. The client tier includes an entirely web browser-based interface, allowing for easy deployment of the application and portability to a wide array of client machine  configurations. The middle tier consists of one or more Java Platform Enterprise Edition (JEE) compliant application servers that manage the application control and interface with the back-end engines that manage the data specific to the Appian Suite.

Overview

v  The Appian Business Process Management Suite (BPMS) is built using a multi-tier web application model.

v  The client tier includes an entirely web browser-based interface, allowing for easy deployment of the application and portability to a wide array of client machine configurations.

v  The middle tier consists of one or more Java Platform Enterprise Edition (JEE) compliant application servers that manage the application control and interface with the back-end engines that manage the data specific to the Appian Suite.

Client Tier

ü  Appian is an entirely web-based application, unique in the business process management (BPM) industry.

ü  Users interact with the entire BPMS—including the portal, document management, analytics, Process Modeler, and task completion interfaces—through a web browser.

ü  Throughout the suite, extensive use of third generation AJAX  is used. This enables the application to provide rich, application-like interfaces without the use of plug-ins or ActiveX controls.

ü  In addition to web interfaces, users can also interact with Appian through other interfaces, including e-mail, for viewing reports and completing tasks.

ü  All client communication occurs through standard HTTPS (HyperText Transport Protocol Secure) protocols to the middle tier.

Middle Tier

ü  The middle tier in Appian Training consists of the control logic that drives the interaction with the client and accesses the information and logic stored in our back-end Appian Training engines.

ü  The middle tier is also responsible for integrating with other applications, external relational databases, and web services.

ü  The primary component of the middle tier is one or more Java application servers. In a typical production environment, the middle tier also consists of one or more web servers that handle requests from the client, manage static content, and forward dynamic requests to the application server(s).

ü  In addition, multiple web servers and application servers are recommended for load balancing and failover, to ensure high availability of the application.

ü  These application and web servers can further be split out across multiple physical machines for redundancy and load balancing.

ü  In a typical configuration, all client requests would be routed to a single traffic manager that routes requests to each of the running web servers. The web servers then provide static content and pass all dynamic requests directly to the Java application server(s).

ü  The application server is responsible for requesting information from, and storing information in, the Appian Training engines.

ü  A connection layer is used within the application server to connect to the Appian Training engines. This connection layer is a library that allows Java applications to connect transparently to the Engines.

ü  This layer manages the pool of connection and routes requests to the appropriate engine, to ensure high availability of the back end, in a similar fashion to JDBC.

ü  All integration with third-party applications, such as web services, relational databases, and enterprise applications, is handled through the application servers. Typically, this is handled through Smart Nodes that execute within the Java application servers.

Appian Engines

ü  Appian BPM Training is a comprehensive BPMS, consisting of advanced process, knowledge management, and integrated analytics. The BPMS has been designed from the ground up to be highly scalable and responsive.

ü  To achieve this, it uses a unique back-end architecture to manage data and perform real-time data analysis.

ü  The architecture consists of a set of Appian Training engines that manage tasks from storing and versioning process models; executing predefined business rules; user, group, and role management; document storage; discussion areas; and task execution.

ü  The engines can be separated onto different physical machines and replicated, typical for load balancing and fail-over arrangements in most relational databases. Each engine is responsible for providing a specific service within the overall application Suite.

Web Application

Appian is a J2EE application platform that creates business applications using executable process models, rules, groups, pages, and other objects such as documents. Applications designed on the Appian platform are portable from one Appian server to another.

Plug-ins

Appian Plug-ins simplify the creation and deployment of common Appian extensions like custom functions and smart services. Appian's Plug-in architecture is based on an industry-standard architecture called OSGI.

Appian Engines

Appian Engines are Real-Time In-Memory Databases (RIMDBs) that allow for very fast read and write access and extremely high scalability. The in-memory data is also persisted to disk, allowing you to use standard backup processes on all data. The RIMDB's real-time synchronous transaction logs capture all events that occur between in-memory checkpoints; allowing any recovery process to apply logged transactions on restart.

Relational Database

Relational databases store Appian data such as News and design object metadata as well as business data, which can be pre-existing or created by users of the applications deployed on the platform.

Search Server

The search server contains indices of design object metadata, runtime performance metrics, and business data for rapid search retrieval and analysis.

High Availability

Appian Cloud offers the ability to run in a high-availability (HA) configuration. An HA configuration protects your Appian instance from downtime in the event of a service failure (a software component of Appian fails), a server failure (the hardware that Appian runs on fails), or an AWS availability zone failure (the data center that Appian runs in fails).

This is achieved by replicating data across three AWS availability zones.

When running in a high availability configuration, an Appian Cloud site has a recovery point objective (RPO) of 15 minutes and a recovery time objective (RTO) of 15 minutes. This means that in the event of a system failure, your Appian instance will be back online and available to your users in less than 15 minutes, having lost no more than the last 15 minutes worth of data.

Because highly-available sites replicate data across three AWS availability zones it is only available in the US, Ireland, and Sydney AWS regions.

Download

S. No	Name	Size	Download
1	Appian Architecture	400 KB	Download