skip to main content
Using the Driver : Performance Considerations
  

Try DataDirect Drivers Now

Performance Considerations

The following connection options can enhance driver performance. You can also enhance performance through efficient application design. See "Designing ODBC Applications for Performance Optimization" for details.
Application Using Threads (ApplicationUsingThreads): The driver coordinates concurrent database operations (operations from different threads) by acquiring locks. Although locking prevents errors in the driver, it also decreases performance. If your application does not make ODBC calls from different threads, the driver has no reason to coordinate operations. In this case, the ApplicationUsingThreads attribute should be disabled (set to 0).
Note: If you are using a multi-threaded application, you must enable the Application Using Threads option.
Array Size (ArraySize): If this connection string attribute is set appropriately, the driver can improve performance of your application by reducing the number of round trips on the network. For example, if your application normally retrieves 200 rows, it is more efficient for the driver to retrieve 200 rows at one time over the network than to retrieve 50 rows at a time during four round trips over the network.
Cached Cursor Limit (CachedCursorLimit): To improve performance when your application executes concurrent Select statements, Cursor Identifiers can be cached. In this case, the Cursor Identifier is retrieved from a cache rather than being created for each connection. When an Identifier is needed, the driver takes one from its cache, if one is available, rather than creating a new one. Cached Cursor Identifiers are closed when the connection is closed. To cache Cursor Identifiers, the CachedCursorLimit attribute must be set to the appropriate number of concurrent open Select statements.
Cached Description Limit (CachedDescLimit): The driver can cache descriptions of Select statements and improve the performance of your ODBC application; therefore, if your application issues a fixed set of SQL queries throughout the life of the application, the description of the query should be cached. If a description is not cached, the description must be retrieved from the server, which reduces performance. The descriptions include the number of columns and the data type, length, and scale for each column. The matching is done by an exact-text match through the From clause. If the statement contains a Union or a subquery, the driver cannot cache the description.
Catalog Functions Include Synonyms (CatalogIncludesSynonyms): Standard ODBC behavior is to include synonyms in the result set of calls to the following catalog functions: SQLProcedures, SQLStatistics and SQLProcedureColumns. Retrieving this synonym information degrades performance. If your ODBC application does not need to return synonyms when using these catalog functions, the driver can improve performance if the CatalogIncludesSynonyms attribute is disabled (set to 0).
Catalog Options (CatalogOptions): If your application does not need to access the comments/remarks for database tables, performance of your application can be improved. In this case, the CatalogOptions attribute should be disabled (set to 0) because retrieving comments/remarks degrades performance. If this attribute is enabled (set to 1), result column REMARKS (for the catalog functions SQLTables and SQLColumns) and the result column COLUMN_DEF (for the catalog function SQLColumns) return actual values.
Client Information: The client information feature automatically adjusts server resources, such as CPU and memory, based on the service class associated with a workload. Therefore, an application’s performance is tied to the workload to which it is assigned and, ultimately, to the service class associated with that workload. The Oracle Wire Protocol driver allows your application to set client information in the Oracle database that can be used by the client information feature to classify work. If you know that your database environment can use client information, coordinate with your database administrator to determine how setting the following options affects performance.
*Accounting Info (AccountingInfo): Sets the CLIENT_INFO value of the V$SESSION table on the server.
*Action (Action): Sets ACTION column of the V$SESSION table on the server.
*Application Name (ApplicationName): Sets the dbms_session value in the database and the PROGRAM value of the V$SESSION table on the server.
*Client Host Name (ClientHostName): Sets the MACHINE value in the V$SESSION table on the server.
*Client ID (ClientID): Sets the CLIENT_IDENTIFIER value in the V$SESSION table on the server.
*Client User (ClientUser): Sets the OSUSER value in the V$SESSION table on the server.
*Module (Module): Sets the CLIENT_IDENTIFIER value in the V$SESSION table on the server.
*Program ID (ProgramID): Sets the PROCESS value in the V$SESSION table on the server.
Connection Pooling (Pooling): On Windows, UNIX, or Linux, if you enable the driver to use connection pooling, you can set additional options that affect performance:
*Load Balance Timeout (LoadBalanceTimeout): You can define how long to keep connections in the pool. The time that a connection was last used is compared to the current time and, if the timespan exceeds the value of the Load Balance Timeout option, the connection is destroyed. The Min Pool Size option can cause some connections to ignore this value.
*Connection Reset (ConnectionReset): Resetting a re-used connection to the initial configuration settings impacts performance negatively because the connection must issue additional commands to the server.
*Max Pool Size (MaxPoolSize): Setting the maximum number of connections that the pool can contain too low might cause delays while waiting for a connection to become available. Setting the number too high wastes resources.
*Min Pool Size (MinPoolSize): A connection pool is created when the first connection with a unique connection string connects to the database. The pool is populated with connections up to the minimum pool size, if one has been specified. The connection pool retains this number of connections, even when some connections exceed their Load Balance Timeout value.
Data Integrity Level (DataIntegrityLevel) and Data Integrity Types (DataIntegrityTypes): Checking data integrity may adversely reduce performance because of the additional overhead (mainly CPU usage) that is required to perform the check.
Default Buffer Size for Long/LOB Columns (DefaultLongDataBuffLen): To improve performance when your application fetches images, pictures, or long text or binary data, a buffer size can be set to accommodate the maximum size of the data. The buffer size should only be large enough to accommodate the maximum amount of data retrieved; otherwise, performance is reduced by transferring large amounts of data into an oversized buffer. If your application retrieves more than 1 MB of data, the buffer size should be increased accordingly.
Describe At Prepare (DescribeAtPrepare): When enabled, this option requires extra network traffic. If your application does not require result set information at prepare time (for instance, you request information about the result set using SQLColAttribute(s), SQLDescribeCol, SQLNumResultCols, and so forth, before calling SQLExecute on a prepared statement), you can increase performance by disabling this option.
Enable Bulk Load (EnableBulkLoad): If your application performs bulk loading of data, you can improve performance by configuring the driver to use the database system's bulk load functionality instead of database array binding. The trade-off to consider for improved performance is that using the bulk load functionality can bypass data integrity constraints.
EnableServerResultCache (EnableServerResultCache): If your application connects to Oracle 11g and executes the same query multiple times, you can improve performance by using the Oracle feature server-side resultset caching. When enabled, Oracle stores the result set in database memory. On subsequent executions of the same query, the result set is returned from database memory if the underlying tables have not been modified. Without result set caching, the server would process the query and formulate a new result set.
Enable Scrollable Cursors (EnableScrollableCursors) and Enable Static Cursors for Long Data (EnableStaticCursorsForLongData): When your application uses Static or Keyset (Scrollable) cursors, the EnableScrollableCursors attribute must be enabled (set to 1). Also, if your application retrieves images, pictures, long text or binary data while using Static cursors, the EnableStaticCursorsForLongData attribute must be enabled (set to 1). However, this can degrade performance when retrieving long data with Static cursors as the entire result set is stored on the client. To improve performance, you might consider designing your application to retrieve long data through forward-only cursors.
Encryption Method (EncryptionMethod), Encryption Level (EncryptionLevel), and Encryption Types (EncryptionTypes): Data encryption may adversely affect performance because of the additional overhead (mainly CPU usage) required to encrypt and decrypt data. Using data encryption can degrade performance more than performing data integrity checks.
Failover Mode (FailoverMode): Although high availability that replays queries after a failure provides increased levels of protection, it can adversely affect performance because of increased overhead.
LOB Prefetch Size (LOBPrefetchSize): You can improve performance when fetching LOBs by enabling the LOB Prefetch Size option (set to value equal to or greater than 0). With LOB prefetching enabled, the driver can return LOB meta-data and the beginning of LOB data along with the LOB locator during a fetch operation, therefore reducing the number of round trips and improving performance. For significant gains, specify a value that is large enough to entirely prefetch LOB values. This allows data to be available without having to go through LOB protocol, which can be expensive.
Lock Timeout (LockTimeOut): Sometimes users attempt to select data that is locked by another user. Oracle provides three options when accessing locked data with SELECT … FOR UPDATE statements:
*Wait indefinitely for the lock to be released (-1)
*Return an error immediately (0)
*Return an error if the lock has not been released within a specific number of seconds (n seconds)
Some applications may benefit by not waiting indefinitely and continuing execution; this keeps the application from hanging. The application, however, needs to handle lock timeouts properly with an appropriate timeout value; otherwise, processing time could be wasted handling lock timeouts, and deadlocks could go undetected.
To improve performance, either enter a number of seconds or enter 0 as the value for this option.
Procedure Returns Results (ProcedureRetResults): The driver can be tuned for improved performance if your application's stored procedures do not return results. In this case, the ProcedureRetResults attribute should be disabled (set to 0).
SDU Size (SDUSize): Set this option based on the size of result sets returned by your application. If your application returns large result sets, set this option to the maximum SDU size configured on the database server. This reduces the total number of packets required to return data to the client, thus improving performance. If your application returns small result sets, set this option to a size smaller than the maximum to avoid burdening your network with unnecessarily large packets.
Server Process Type (ServerType): When using a dedicated server connection, a server process on UNIX (a thread on Windows) is created to serve only your application connection. When you disconnect, the process goes away. The socket connection is made directly between your application and this dedicated server process. This can provide tremendous performance improvements, but will use significantly more resources on UNIX servers. Because this is a thread on Oracle servers running on Windows platforms, the additional resource usage on the server is significantly less. This option should be set to 2 (dedicated) when you have a batch environment with lower numbers of connections, your Oracle server has excess processing capacity and memory available when at maximum load, or if you have a performance-sensitive application that would be degraded by sharing Oracle resources with other applications.
Use Current Schema for Catalog Functions (UseCurrentSchema): If your application needs to access database objects owned only by the current user, then performance can be improved. In this case, the Use Current Schema for Catalog Functions option must be enabled. When this option is enabled, the driver returns only database objects owned by the current user when executing catalog functions. Calls to catalog functions are optimized by grouping queries. Enabling this option is equivalent to passing the Logon ID used on the connection as the SchemaName argument to the catalog functions.