Protractor Corporate Information

Technical Information

Protractor software was written using Microsoft Visual Studio 6.0 Service Pack 4 in Visual Basic and C++ using pure object-oriented (OO) techniques.

We planned and built the software using a modern, multi-tier Component Object Model (COM+) architecture.

Protractor's User Interface (the client-side program) is a very robust, rich, 32-bit Windows Program (sometimes referred to as a 'fat' or 'thick' client).

Protractor's database backend is Microsoft SQL Server 2000. Multiple levels of data redundancy ensure that your company's data is absolutely private and secure. SQL Server 2000 is an enterprise-level, award winning, extremely fast, and reliable database software system.

All database access is via commands and parameters to optimized T-SQL Stored Procedures which are protected by SQL transactions. Much of the Business Logic is implemented in the Stored Procedures to optimize data access speed.

The diagram below illustrates the communication technique we use between your local Protractor Client and the Protractor Servers. The network may be your Wide Area Network, the public Internet, or your Local Area Network.

For a single computer installation, the Servers may run on the same computer as the Client. This allows the software to be scaled anywhere from a single computer installation to a world-wide enterprise environment with hundreds of clients.

When a Protractor Client makes a request for data, the request is translated to the standardized data language of the Internet (eXtendable Markup Language or 'XML') by the XML API Client. The XML data request is compressed (using Zip techniques by the ZLib Compression Library) for quicker transmission, then sent across the network using TCP/IP to the listening Protractor Server. The Client then waits and listens for the Server's reply.

When the Server gets a Client's request for data it is decompressed and placed in a request queue for the applicable Server - there could be hundreds of Clients sending simultaneous data requests. Protractor's queue handling algorithms use Windows 2000's low-level multi-tasking services, so requests may be happening simultaneously.

On reaching the top of the queue, the data request is made to the applicable Server (the Catalog Server, the Database Server, or the Supplier Connection Server). SQL Server 2000 uses the Operating System's advanced multi-tasking and multi-processor capabilities when available.

The applicable Server processes the data request, passes it back to the Protractor Server which translates it to XML, compresses it, and sends the data back to the waiting Client.

This communication technology optimizes the use of a SQL Server by allowing it to process all data requests locally on the server computer, thereby eliminating the network traffic created by locking, unlocking and transferring native database recordsets.

By using XML in all data requests, Protractor has readied itself for interaction with all future Internet Services (as per Microsoft's .NET initiative).