Network Setup Options for Open Data Services Server Applications

This chapter discusses options for making an Open Data Services server application available for use on the network. It describes configuration requirements for the Open Data Services server application, for SQL Server, and for DB-Library clients.

Your configuration requirements will depend on:

• The type of network you are using.
• Whether the Open Data Services server application is installed on the same computer as SQL Server or on a separate computer.
• Whether you want to provide clients with direct access or remote access to the Open Data Services server application.

Each software process on a network must have a unique set of identifiers that distinguishes it from all others. The connection information each computer requires depends on the type of network protocol that the Open Data Services server application uses to accept connections from other software processes on the network. In some cases, you can take advantage of SQL Server default values to eliminate some configuration steps.

Open Data Services for Windows NT can accept requests from clients and servers across multiple network protocols simultaneously. Open Data Services supports multiple network protocols though the use of loadable network drivers, known as Net-Libraries. Open Data Services supports the same server Net-Libraries as SQL Server. For specifics, see Microsoft SQL Server Setup.

This chapter discusses Net-Library configuration options.

Each Net-Library uses a two-part identifier to uniquely locate server processes in a SQL Server environment: a node identifier and a process identifier. The node identifier indicates the computer on which the software process resides. The process identifier locates a specific software process on that computer.

For networks that use named pipes, the node identifier is specified as a computer name, for example, \\svr_name, and the process identifier is specified as a pipe name, for example, \pipe\sql\query.

For networks that use TCP/IP socket addresses, the node identifier is specified as a numeric IP address, for example, 11.1.4.68, and the process identifier is specified as a socket number, for example, 1433.

When Open Data Services server applications and SQL Server are installed on separate computers on the network, they can use the same process identifiers. When an Open Data Services server application and SQL Server are installed together on a single computer, they share the same node identifier, requiring you to assign an alternate process identifier to one of them. In addition, any computer that requires direct access to that process must be configured to refer to it by its alternate process identifier.