Join processing is optimized to reduce the number of rows built. This can result in a major performance improvement for large tables. Also the query can contain a hint clause to tell the OpenAccess SDK SQL engine the order in which to perform the join. Other advanced join related features include Join Pushdown and Block Join.
• Join Pushdown is ideal for hierarchical or object-oriented systems.
• Block Join is ideal when there is a large overhead in accessing the inner rows for each outer row.
Advanced optimization
This feature builds on basic query optimization to allow your IP to get the complete WHERE clause expression in an AND or OR form. In addition, the Table Expression Pushdown feature enables your IP to have full access to the filter conditions on each table. The parse tree API can be used to examine the complete query.
Authentication support
To authenticate access to the server, you can use:
• The client’s TCP/IP address
• Data source specific user name and password
• Operating system login
• Kerberos
Please refer to the Authentication section in the OpenAccess SDK Administrator’s Guide for information on authentication.
Binary schema
You can compile your schema into a binary format for quicker access during execution. The supplied schema compiler converts schema stored in Microsoft Access format to OpenAccess SDK binary format. You can modify the compiler to read the schema from your existing data dictionary.
Cross-platform ODBC
ODBC is supported on non-Windows platforms to allow you to develop portable applications.
Cursor-based return
Execution of a query can be done in multiple steps with each step returning some data to the client. This allows the client to start seeing data before the entire data set is processed.
Customization of driver information and data types for Java and .NET IPs
The Java and .NET IPs can obtain information about the data source such as the SQL capabilities, limits on object names, and other information.
Disk caching
Intermediate results are saved to disk instead of using memory.
Support for very large tables
OpenAccess SDK can handle very large tables that contain more than two billion rows. This means that the cardinality can be of more than two billion rows. Therefore, UPDATE, SELECT, and DELETE statements can process rows up to 9,223,372,036,854,775,807.
Internationalization
Unicode is supported to allow data in any language to be accessed. Error messages, dialog boxes, and other items that contain text can be localized. OpenAccess SDK allows the use of Unicode characters in SQL queries, identifiers, and in columns.
Join pushdown
In many cases, the backend for which the interface is being developed is capable of optimized join processing between certain tables and should be given responsibility in the IP for the execution of these joins.
License management
Use the OpenAccess SDK license management schema to ship limited-time demonstration copies of the ODBC driver, the ADO data provider, or the JDBC driver to your customers. The customer can then purchase a full version by contacting your organization for the new key.
Load balancing
OpenAccess SDK Servers can be run on multiple machines and exposed to the client as one logical data source. The client can be configured to access this server using random or round-robin selection methods.
Multiple-platform support
The same IP source code can be used to build servers on any of the supported Windows, Linux, and UNIX platforms. Directories and MAKE files are organized to allow single-source code to be used for all platform builds.
For a complete list of supported platforms, visit:
Allows the IP to support the execution of native commands. This allows the driver developer to support a syntax that is specialized for the data source.
Redundancy
OpenAccess SDK Servers can be run on multiple machines and exposed to the client as one logical data source. This allows OpenAccess SDK to be configured to support failover.
SQL pass-through mode
Use the OpenAccess SDK SQL engine to do the SQL parsing and then access the parse tree to build up a command for your data source to execute. In this mode of operation, OpenAccess SDK provides the entire client API-related processing, parsing, and if required, the client/server processing. Your IP can handle the command execution.
SSL server certificate support
SSL is supported for encryption of network traffic between the client and the server.
User-defined data types
Columns with user-defined data types, such as special date formats ('now', 'now – 2'), are handled by allowing the IP to perform the expression processing.
User-defined scalar functions
The IP can register with the OpenAccess SDK SQL engine and implement scalar functions that take a fixed or variable number of arguments.
User-defined views
Full support for user-defined views on top of your schema whether you are implementing static or dynamic schema.
User rights
Full support for data access rights by allowing the IP to validate access to all objects.
