Andreas Geppert

Dynamic workflow schema evolution based on workflow type versioning and workflow migration

An important yet open problem in workflow management is the evolution of workflow schemas, i.e., the creation, deletion and modification of workflow types in such a way that the schema remains correct. This problem is aggravated when instances of modified workflow types are active at the time of modification because any workflow instance has to conform to the definition of its type. The paper presents a framework for dynamic workflow schema evolution that is based on workflow type versioning and workflow migration. Workflow types can be versioned, and a new version can be derived from an existing one by applying modification operations. Workflow type versions allow us to handle active instances in an elegant way whenever a schema is modified. If possible, an affected workflow instance is migrated to the new version of its type. Otherwise, it continues to execute under its old type. We introduce correctness criteria that must be met by workflow schemas and workflow schema modification operations. We also define under which conditions the migration of workflow instances to new workflow type versions is allowed.

Event-based Distributed Workflow Execution with EVE

In event-driven workflow execution, events and event-condition-action rules are the fundamental metaphors for defining and enforcing workflow logic. Processing entities enact workflows by reacting to and generating new events. The foundation on events facilitates the integration of processing entities into coherent systems. In this paper, we present an event engine, called EVE, implementing event-driven execution of distributed workflows. Its functionality includes event registration, detection and management, as well as event notification to distributed, autonomous, reactive software components which represent workflow processing entities. We describe the distributed, multi-server, multi-client architecture of EVE and its use for workflow execution.

Resolving semantic heterogeneity in schema integration

Interoperability and integration of data sources are becoming evermore important issues as both, the amount of data and the number ofdata producers are growing. Interoperability not only has toresolve the differences in data structures, it also has to dealwith semantic heterogeneity. Semantics refer to the meaningof data in contrast to syntax, which only defines the structure ofthe schema items (e.g., classes and attributes). We focus on thepart of semantics related to the meanings of the terms used asidentifiers in schema definitions. This paper presents an approachto integrate schemas from different communities, where each suchcommunity is using its own ontology. The approach is based onmerging ontologies based on similarity relations among concepts ofdifferent ontologies. We present formal definitions of similarityrelations based on intensional definitions and conclude theextensional consequences. The process of merging ontologies basedon the detected similarity relations is discussed. The mergedontology is finally used to derive an integrated schema. Theresulting schema can be used as the global schema in a federateddatabase system.

Market-Based Workflow Management

This paper presents market-based workflow management, a novel approach to workflow specification and execution which regards activities contained in a workflow as goods traded on an electronic market. Information about expected cost and execution time is considered for activity specifications, and is used at runtime to execute workflows such that actual cost and execution times are balanced and optimized. To that end, task assignment uses a bidding protocol, in which each eligible processing entity specifies at which price and in which time interval he/she can execute the activity. The winner of a specific bidding process is requested execute the activity, and earns the amount specified in the corresponding bid. Market-based workflow management thus not only allows to optimize workflow executions with respect to execution time and overall cost; but the trading of activities represents an incentive for processing entities to engage in a workflow.

Defining the semantics of reactive components in event-driven workflow execution with event histories

The definition of exact semantics of workflows and involved processing entities is an open yet urgent problem for most systems. This paper considers the semantics and correctness of event-driven workflow execution by reactive components. The basis for the formalization in our approach is provided by an event his tory which records all the events generated during the execution of workflows by these components. Based on the formal notion of history, it is possible to determine the semantics of workflow systems as the set of event histories they can generate. It also allows to formalize the semantics of reactive components-and consequently, the semantics of workflows-as well as to check whether their observable behavior is correct. In addition to the precise understanding of event-driven workflow execution, the formalized notion of event history also serves as the specification of the workflow execution engine. Ultimately, the semantics of reactive components provides the basis for specifying transformation rules used to map these components to the execution level.

Logging and Post-Mortem Analysis of Workflow Executions Based on Event Histories

Workflow execution logging and post-mortem analysis of workflow executions are important tasks of a workflow management system. In this paper, we show how both tasks are addressed in EvE, a prototypical distributed system implementing event-driven workflow execution. Essentially, both tasks benefit from the concept of an event history as it is present in active database systems. By using event-based workflow execution, the work flow log actually corresponds to the event history, and post-mortem analysis is accom plished through querying the event history. In our approach, an analysis of the required information has led to the design and development of a workflow analysis query service.

Semantics of Reactive Components in Event-Driven Workflow Execution

The exact semantics of workflows and involved processing entities is an open yet urgent problem. This paper considers the semantics and correctness of event-driven workflow execution. The basis for the formalization in our approach is provided by an event history which records all events that have occurred during the execution of workflows. Workflows are executed by reactive components which operate on top of that history. Based on the history it is possible to determine the semantics of these reactive components (and consequently, the semantics of workflows) as well as to check whether their observable behavior is correct.

The SAMOS active DBMS prototype

We describe SAMOS, an active object-oriented database management system prototype. SAMOS offers a powerful rule definition language, including a small yet powerful set of event definition facilities. It is able to detect primitive and composite events automatically and efficiently. Upon event detection, SAMOS executes rules attached to the occurred events.

Resolution of Semantic Heterogeneity in Database Schema Integration Using Formal Ontologies

Abstract This paper addresses the problem of handling semantic heterogeneity during database schema integration. We focus on the semantics of terms used as identifiers in schema definitions. Our solution does not rely on the names of the schema elements or the structure of the schemas. Instead, we utilize formal ontologies consisting of intensional definitions of terms represented in a logical language. The approach is based on similarity relations between intensional definitions in different ontologies. We present the definitions of similarity relations based on intensional definitions in formal ontologies. The extensional consequences of intensional relations are addressed. The paper shows how similarity relations are discovered by a reasoning system using a higher-level ontology. These similarity relations are then used to derive an integrated schema in two steps. First, we show how to use similarity relations to generate the class hierarchy of the global schema. Second, we explain how to enhance the class definitions with attributes. This approach reduces the cost of generating or re-generating global schemas for tightly-coupled federated databases.

Performance evaluation of object-oriented active database systems using the BEAST benchmark

This paper uses the BEAST benchmark to present the first serious performance study of object-oriented active database management systems (ADBMS). BEAST stresses the performance-critical components of active systems: event detection, event composition, rule retrieval, and rule firing. Method invocation events and transactional events are taken into account. Four systems, namely ACOOD, NAOS, Ode, and SAMOS, have been tested with the benchmark tests of BEAST. The performance measurements demonstrate achievements in the area of active database technology, but also indicate tradeoffs (e.g. between performance and functionality). Finally, the benchmark identifies optimizations and provides hints to ADBMS desigers about producing systems with adequate performance and functionality --- as well as some open issues.