DataWorks Plus logo

Fusion-im

Information Management

 

Fusion-im logo

Access all of your agency's data in one place

 

Technical Information

Data from the data source is passed through an adapter to a messaging server and then transformed using business logic objects to the destination workstation. Data can be from one or from multiple data sources. The system can be used by a single agency for applications involving warrants, RMS data, JMS data or other agency information. The system is scalable to be used across multiple agencies to share data among disparate systems.

 

Done securely with enterprise level scalability, open architectures allow pervasive interaction between applications in a way that gains the greatest cost effectiveness within an enterprise. In keeping with this philosophy, DataWorks Plus ascribes to the Java J2EE components that apply to the needs of enterprise customers today. Additionally, the open source Apache project has brought a great deal of benefits to the Java community, especially with the Tomcat (Java-based Servlet architecture), Cocoon (XML and XSLT tools), and Axis (SOAP Framework) projects. DataWorks Plus has incorporated these technologies into the core server solution that we provide. We have followed with great interest the advent of Java Messaging Services (JMS) and the use of such an architecture to connect to diverse systems. By incorporating a middleware solution into the architecture, a secure connection can be achieved quickly and easily, and abstraction of the location of information and applications is achieved. Publishing large quantities of information to multiple recipients also becomes much simpler from an architectural and administrative standpoint as well as from a distribution standpoint.

 

Past Mistakes

Because of these and other features of the JAVA Messaging Service architecture, DataWorks Plus has incorporated a JAVA Messaging Service solution at the core of its infrastructure, routing all communications between disparate or disconnected systems be done through this environment.

 

JAVA Messaging Service allows us to bring key scalability features to the design as well, with messaging servers that can be clustered and replicated for redundancy and scalability, as well as allowing local messaging ontologies to be routed to external servers with little administration effort. Messages, when connecting to internal or external data sources, are guaranteed to be delivered from point to point, regardless of whether systems are connected on a continual basis. This enterprise level of information exchange is required – and will be demanded by enterprise interaction over the public network.

 

Connectivity to the JAVA Messaging Service servers is facilitated through a small client application running as near to the actual data source as possible. Where practical, this client should actually operate on the same machine as the source application or data. Realizing that this is not always feasible or recommended, DataWorks Plus allows the components to be run anywhere within the enterprise that allows physically secure connectivity to the source data.

 

The adapter can proxy through any port that is available on the firewall (with HTTP, HTTPS or SSL and SMTP being the most common ports), making implementation simpler that other integration architectures. Security is also a concern both internally and routing information externally. Aware that this is of paramount concern in exchanging enterprise data, DataWorks Plus has built into the core architecture the ability for each piece of information that is exchanged (internally or externally) to be encrypted at two levels. Each socket connection to the server can be facilitated through an SSL link to the server, with the contents of the documents also encrypted (payload encryption) using a public/private key architecture and standard encryption algorithms.

 

Delivery of information can be very complicated. It’s simple enough when you just have two systems that want to pass a transaction. It becomes much more complicated when you want an external business partner to have access to real time inventory quantities. You don’t want an external source knowing what IP addresses are used within your network. Better yet, you want to be able to change internet service providers without having to notify all those partners of that change. Or, you may want to expand operations to allow multiple or clustered servers to respond to requests. The messaging architecture and queuing ontology that DataWorks Plus uses allows that change to be as simple as restarting the agent on a new machine and connecting to the server again.

 

Sources of information within an enterprise may vary greatly. Everything from a relational database, to legacy applications, to Excel Spreadsheets, to flat files exported from a batch process, to EDI, to robust APIs may be available. We believe that to be truly effective and pervasive within the enterprise architecture, we should be as non-invasive as possible. That means connecting to whatever is there, however we can. In order to do this, we must have an architecture that minimizes custom programming, and facilitates connectivity as quickly as possible to the core architecture. It also means being able to take the source data and transform and translate the data from and to any format that might be needed.

 

Many times no customization is needed when connecting….

In order to do this, we have created an object-oriented approach that allows 80-90% of the connectivity components to be reused, only requiring 10-20% to be written for a custom connection. Many times, no customization is required when connecting to a standard data connection such as a database or a known API. We call these connection points adapters, and each one can be a subclass of the core architecture that allows the connections to the server, to the backend data, or to the partner system. The adapter framework is a key component that sets us apart from our competitors, allowing us to quickly adopt new data points and interaction, and making that data available to those in the enterprise you want to have access to it. Secure access to connections is controlled at the point of data origin, allowing very strict, or global access to one or more pieces of data that are defined in the adapter connection and mapped through the server-based process.

 

Adapters are used for connecting to all types of data, and are used for reading and writing data to the data sources. Scalability of connections to data sources that were traditionally not able to add many users can be facilitated through this approach as well, since one adapter connection to the data source can manage many client connections.

 

Similar to the adapter framework, we also have incorporated some key agents in the architecture that perform specific functions to the overall community of components used by any node or set of nodes in the network.

 

Translation of data from one format to another is one example of an agent that is built into the DataWorks Plus framework. This agent receives data in whatever format it is sourced from, along with routing information for where the data needs to be sent. Once translated into its destination format, it is routed to the destination, and so on, until communication between the endpoints is complete. Using this methodology, one engine can be used to incorporate translations between many varied clients and applications quickly and efficiently.

 

Application agents that facilitate interaction between a web page and the architecture combine multiple adapters and user interface components to allow users to view and update information from diverse data sources through a common web-based design. Many of our customers have used this agent to facilitate additional functionality or review of information for approval before routing to the eventual endpoint. Providing a web-based design allows users to access information anywhere through a common interface.

The DataWorks Plus architecture supports open standards and connectivity to all types and sources of data within the enterprise and with business partners. Connectivity is facilitated quickly and easily through a robust object-oriented design, and is transformed and delivered securely and reliably to any destination anywhere. Application functionality is built into the system quickly and easily through agent technology, expanding functionality without modifying existing applications. Fusionim is your total connectivity solution.