HomeSolutionsProcess Governance & Visibility

Process Governance and Operational Visibility

 

Process Governance and Operational Visibility are complementary disciplines of IT governance, whose goal is to provide organizations with a consistent and reliable view of their data processing resources.  Operational Visibility tools allow employees responsible for system operations to manage and monitor the software components and in-flight data that comprises business processes.  Process Governance technologies allow system architects to specify rules for process dependencies, error detection and reaction to exception conditions.  

Management and Governance   

When it comes to business process automation and service-oriented systems (SOA), conventional governance and management have proven to be inadequate. The loosely coupled nature of service interactions in a business process often requires policy enforcement and meta-data administration that conventional management tools do not provide.  Service oriented applications are typically distributed across different organizations and may include components that are beyond control of the IT staff.

Operations teams that manage service-oriented systems are tasked with understanding and controlling an active, constantly evolving network of interdependent services that may span multiple business units. 

Recovering from exceptions and determining the root cause of errors is usually a collaborative process involving multiple stake holders and technologies.  While certain aspects of failure recovery may be automated, many situations require coordination between teams and a significant amount of operator interaction.

In-flight Process Analytics

Unlike many conventional systems that use a system-of-record database to drive their logic, business process applications rely on transient, process-local data that provides information about process state or versions of data as it moves thru the system.

Such data tends to be available during the life span of a process and is usually discarded when a process completes.  However, in-flight data often contains critical information that may be used for problem resolution or state-full process inspection.  As organizations continue to automate their business processes, transient data is becoming more valuable because it often represents a more accurate state of the system.

The Service Application Engine™ provides a rich set of facilities for real-time analytics and governance of in-flight data. The application fabric's Dataspace technology offers model driven data persistence that allows users to store process data into queues, tables or map collections and archive such data after process completion.  In-flight data may be queried across multiple process instances using ANSI compliant SQL queries or by user defined query syntax.   Collections may be accessed through a variety of API and protocols including JDBC and Restful HTTP requests.

Root Cause Analysis & Exception Handling

One of the key benefits of implementing a Service Oriented Architecture is a reduction of operational inefficiencies. Automating a business process by implementing loosely coupled services often allows a company to reduce operational cost or make changes to an existing process faster.

However, experience has shown that a high percentage of inefficiency is simply attributable to human error, such as a lack of situational awareness or critical thinking that fails to consider all aspects of a given problem. This is not surprising given the breadth and scope of modern business processes. With a large number of moving parts it is often difficult to track process activity, analyze the root cause of a problem and account for the global impact of a specific failure. As such, response to failure becomes a slow and iterative process.

The application fabric provides tools for visualizing business processes and identifying critical exceptions, allowing operators to pinpoint the exact location of a problem across process types.  Using the fabric’s Event Definition Language developers may define actions, audit or exception triggers on service calls or data collection events. Definitions may specify which events or data content are considered errors, warnings or conditions that halt further processing. This approach allows for granular error definition and active response to failure, simplifying root cause analysis and accelerating problem resolution.

Collaborative Governance

Proper governance of de-centralized, mission-critical systems is essential to an organization’s success and requires a specialized tool set that facilitates collaborative management and governance. In many cases, governance does not imply detecting and reacting to failure, but encompasses the process of failure recovery and problem resolution, for example the execution of compensating transactions or undo operations.  As such a tool set must provide operators with a correct administrative perspective of system components, allow them to visualize the flow of information through the system and give them the capability to use such in-flight data for decision making and problem resolution.

Collaborative governance encompasses people, processes, and the technology required to manage a consistent flow of data across the enterprise. It is a process that naturally fits within the Collaborative Computing paradigm. If enterprise data assets are considered product, then collaborative governance tools provide the necessary command and control facilities that allow IT staff to keep the factory operational. The application fabric provides critical tools and features required of a collaborative governance system allowing groups of operators to cooperatively manage business processes and SOA infrastructures.  

Key Platform Features

  • Graphical process visualization using the TruView Operations Console™ allows developers and operations staff to collaboratively manage business processes and data flow via a browser based application.
  • Provides facilities for real-time process tracking and notifications, step-by-step auditing, process history and archiving.
  • Process Queues allow users to design state-full business processes where state and process data may be queried and archived. Processes may be stopped or suspended due to exceptions or other error conditions, allowing business users to inspect and repair process data.
  • Supports visualization and management of multiple process models such as Pipeline, Offer/Take and Multi-party Acknowledgement for data distribution.
  • Provides configurable retry, timeout, exception handling intervals and parallelism settings, allowing users to tune for performance, for instance in case of a time sensitive batch window. 
  • Comprehensive Web Interface for Auditing and Reporting of critical metrics and process audit information such as Service Requests, SQL Auditing, results of data transformation operations and more.
  • Automatic API generation for Restful HTTP clients using XML or JSON data formats, allows developers to extend the portals functionality to include other applications and mobile devices.
  • Extensible customization framework for adding rich graphical components and Action Windows to the console environment.
  • Support for Roles and Organizations allows the Operations Console to present Administrative Perspectives relative to the needs of the user, only showing processes and alerts pertinent to their organization.
  • Integrated security allows developers to lock down processes and API calls across the process space and validate against external security mechanisms such as LDAP.
  • Process queues and other data collections allow developers to perform complex query and aggregation of in-flight process data using standard SQL or user defined syntax.
  • No need for additional technologies such as database or messaging systems.  The Service Application Engine provides native messaging and data management facilities that are tightly coupled and transactional. Your administrative view is never out of sync with the actual process data.
  • Collaborate with team members using Real-Time Messaging and Notifications over HTTP, native messaging or popular Instant Messaging clients (ie. Trillium, ICQ or Pandion).