Exadata Cell Offload Smart Scan Features Boosting Database Performance with Oracle Exadata Storage Server

YouTube video


In this episode of the Accelerator Tech Dark Series, we will explore the powerful features of Exadata Cell Offload and Smart Scan in the Oracle Exadata storage server. These features play a crucial role in improving the performance of Oracle Exadata, and understanding their functionality is essential for maximizing the capabilities of this powerful system. So, let’s dive into the details and discover how Exadata Cell Offload and Smart Scan can supercharge your database performance.

Understanding Exadata Cell Offload

Exadata Cell Offload, in simple terms, refers to offloading resource-intensive workloads from the database server to the storage server. In a traditional architecture, when a database call is made, the data passes through a narrow CPU tunnel, which can lead to high CPU wait times and negatively impact application performance. However, Oracle Exadata solves this problem by moving the storage components exclusively into the storage server.

With Exadata Cell Offload, the database server communicates directly with the storage server through the IDB protocol. The storage server, being aware of the type of data stored and its characteristics (such as minimum and maximum values), can efficiently retrieve the required data blocks for a database transaction. As a result, the CPU waiting time on the database server is significantly reduced, leading to improved database performance.

Exploring Exadata Smart Scan

Exadata Smart Scan is a feature that further enhances database performance by offloading SQL processing from the database server to the storage servers. When a select statement is executed on a table, Exadata Smart Scan directs the statement to the storage server via the IDB protocol. The storage server, equipped with Smart Scan capabilities, identifies the required data blocks for the select statement and retrieves only the necessary data for the database server.

Exadata Smart Scan operates using three main functionalities:

  1. Predicate Filtering: This feature allows the storage server to filter out subsets of data based on specified conditions. For example, if an employee table contains five million records, but only one million records meet a certain department criteria, the storage server can retrieve only the required one million records, significantly reducing the amount of data transferred and processed by the database server.

  2. Column Projection: With Smart Scan, the database server can request specific columns from a table, and the storage server will retrieve and transmit only those columns back to the database server. This eliminates the need for the database server to process unnecessary columns, resulting in improved performance and reduced resource consumption.

  3. Join Filtering: In a data warehouse environment where complex table joints and filtering operations are common, Exadata Smart Scan leverages the bloom filtering functionality within the storage server. It allows the storage server to handle the filtering and join operations internally, reducing the amount of data passed through the CPU tunnel and optimizing performance.

The powerful combination of Exadata Cell Offload and Smart Scan substantially improves database performance compared to other appliances in the market. By minimizing database I/O operations and reducing the data passing through the CPU tunnel, Exadata ensures efficient and lightning-fast query processing.

Prerequisites and Identifying Smart Scan Usage

To leverage the benefits of Exadata Smart Scan, certain prerequisites must be met:

  1. The query must undergo a full table scan since Smart Scan is enabled for table access storage full.
  2. The underscore serial direct read parameter must be set to “always” to enable the database to read directly from the storage server.
  3. The cell offload processing parameter must be set to “true” to enable offloading of database operations to the storage servers.

To verify whether a query is utilizing Smart Scan, you can examine the execution plan of the query. If you see the term “table access storage full” in the plan, it indicates that Smart Scan is being employed. Additionally, you can use the v$session_stat view to check the total eligible bytes for cell offload and the physical I/O interconnect bytes written by the Smart Scan for a specific query.

By optimizing query execution with Exadata Cell Offload and Smart Scan, organizations can achieve significant reductions in database I/O, CPU wait times, and resource consumption, thus delivering exceptional database performance and improved user experience.


Exadata Cell Offload and Smart Scan are integral features of the Oracle Exadata storage server that enable powerful database performance enhancements. By offloading resource-intensive workloads and intelligently scanning and retrieving only the necessary data, Exadata maximizes CPU capacity, minimizes I/O operations, and boosts query processing speed. As a result, organizations can achieve unparalleled performance gains and provide a seamless user experience to their valuable customers.

So, if you are looking to supercharge your database performance, harness the power of Exadata Cell Offload and Smart Scan with Oracle Exadata Storage Server!

Remember to subscribe to our channel, Tech TV, for more informative videos on Oracle Exadata and related topics. Your support and engagement are greatly appreciated as we strive to deliver valuable content to our audience. Don’t hesitate to share this video on social media and visit our blog for more in-depth insights into Oracle Exadata. If you have any queries or questions, feel free to reach out to us. Until next time, happy optimizing!

Keyword: Exadata Storage Server