Oracle Database 10g: Inside the ‘Self-managing’ Database

Oracle Database 10g: Inside the ‘Self-managing’ Database

racle has given its grid-ready, next-generation database the ability to manage itself and assume the burden of daily management tasks from the database administrator (DBA). During a session dedicated to the Database 10g product at the OracleWorld Conference in San Francisco this past week, Richard Sarwal, vice president of server performance and data server applied technology for Oracle, listed five main database manageability challenges that DBAs face today. He stressed that they all occur outside the database itself:

  • Application and SQL management
  • Backup and recovery
  • System requirements management
  • Storage management
  • Space management

Oracle has built functionality into its new database product that enables it to meet each of these challenges with only minimal intervention from administrators. “10g is pulling all those things that take place outside the database inside the database,” said Sarwal.

10g pulls management tasks that take place outside the database inside the database.
Based on customer feedback from the 2001 International Oracle Users Group (IOUG) User Survey, Oracle discovered that DBAs devote 55 percent of their time to ongoing systems management. A self-managing database, Oracle reasons, would return much of that time to DBAs, freeing them to concentrate on more strategic tasks.

How Database 10g Was Built
According to Sarwal, the first thing Oracle did was build a repository inside the database for raw system statistics and object data. The result was the Automatic Workload Repository (AWR). By default, 10g retrieves database system data and dumps it onto disk in the AWR every 30 minutes. Functioning as the database’s own data warehouse, this data collection informs Database 10g‘s self-management.

Next, Oracle had to ensure that the database would know how to use this data, which led to the creation of the Automatic Database Diagnostic Monitor (ADDM). Sarwal calls ADDM (pronounced “Adam”) the brains of the database. Its role is to automatically pore through all the infrastructure data in the AWR, detect patterns of use, and produce graphical diagnostics reports that include recommendations for improving performance. It can also prompt server-generated alerts when bottlenecks, “hot spots,” and other performance problems occur.

ADDM is a “performance expert in a box.”
ADDM is a “performance expert in a box” according to Sarwal. Using its Automatic Tuning Optimizer, for example, ADDM profiles SQL statements to identify bad (or high-load) SQL, automatically adapts to workload changes, and performs dynamic sampling and partial execution techniques.

“Let the Database Manage Storage”
Automatic Storage Management (ASM) is another new Database 10g feature that Oracle is touting. Andy Mendelsohn, Senior Vice President, Oracle Database and Application Server Technologies, describes it as a built-in, high-performance file system and disk manager. Mendelsohn says ASM is based on the principle that the database knows better how to manage storage than its administrators do. “Give the database the raw disk and let the database manage the storage for you,” he said.

The way it accomplishes this is by striping and mirroring data across modular storage arrays (i.e., low-cost storage-like disks), treating the database grid as a large pool of storage rather than separate volumes. ASM simultaneously performs automatic I/O tuning among disks and auto load-balancing based on user-defined requirements. The need for a volume manager and file systems is virtually eliminated, and ASM automatically remirrors when a disk fails.

Self-management, explained Executive Vice President for Server Technologies, Chuck Rozwat in a keynote address Wednesday, means that the database system is capable of taking a workload and figuring out where to run it in order to meet predefined requirements. For example, administrators can predefine a Quality of Service level or define a threshold for CPU usage and let Oracle 10g provision the workload to meet those needs.

Flashback Database: Single-command Recovery
Oracle has built Flashback features into 10g that will replace the old recovery method of restoring data from tape backups. A fundamental change in 10g is the shift in storage media from tape to disk. Mendelsohn explained that the use of tapes dates back decades to when tape was cheaper than disk. But now that the two media are nearly equal in cost, why not use disk, he argues. Tape recovery is linear, so administrators have to recover the database at a given point in time and roll the changes forward. Disk, on the other hand, enables random access recovery where only the needed data is restored.

Juan Loaiza, Oracle’s vice president of systems technologies, describes the Flashback Database feature as “a rewind button for the database.” A Flashback Log inside Database 10g acts as a continuous backup. It captures old versions of changed blocks, operating on only the changed data. An administrator can replay the log to restore a database and it restores only the changed blocks.

Database 10g offers these Flashback recoveries at all levels: database, table, and row.
Database 10g offers these recoveries at all levels: database (Flashback Database), table (Flashback Table), and row (Flashback Row). The result is incremental recovery: the ability to go back only as far as necessary and to conversely go forward if you went too far. Loaiza sums it up by saying, “if it took you five minutes to make an error, it shouldn’t take 50 hours to recover. It should take five minutes to fix.”

All this internal intelligence promises reduced failover times with faster recognition and restart. Loaiza said, “we’re trading disk space, which is cheaper these days, for downtime, which is expensive.”

The approximately 2,000 attendees of Wednesday’s keynote reacted enthusiastically to a demonstration of the new Flashback recovery features.

Oracle Database 10g is slated for release at the end of 2003.

Share the Post:
XDR solutions

The Benefits of Using XDR Solutions

Cybercriminals constantly adapt their strategies, developing newer, more powerful, and intelligent ways to attack your network. Since security professionals must innovate as well, more conventional endpoint detection solutions have evolved

AI is revolutionizing fraud detection

How AI is Revolutionizing Fraud Detection

Artificial intelligence – commonly known as AI – means a form of technology with multiple uses. As a result, it has become extremely valuable to a number of businesses across

AI innovation

Companies Leading AI Innovation in 2023

Artificial intelligence (AI) has been transforming industries and revolutionizing business operations. AI’s potential to enhance efficiency and productivity has become crucial to many businesses. As we move into 2023, several

data fivetran pricing

Fivetran Pricing Explained

One of the biggest trends of the 21st century is the massive surge in analytics. Analytics is the process of utilizing data to drive future decision-making. With so much of

kubernetes logging

Kubernetes Logging: What You Need to Know

Kubernetes from Google is one of the most popular open-source and free container management solutions made to make managing and deploying applications easier. It has a solid architecture that makes

ransomware cyber attack

Why Is Ransomware Such a Major Threat?

One of the most significant cyber threats faced by modern organizations is a ransomware attack. Ransomware attacks have grown in both sophistication and frequency over the past few years, forcing

data dictionary

Tools You Need to Make a Data Dictionary

Data dictionaries are crucial for organizations of all sizes that deal with large amounts of data. they are centralized repositories of all the data in organizations, including metadata such as