Skip to main content

2016 | Buch

Expert Oracle Indexing and Access Paths

Maximum Performance for Your Database

verfasst von: Darl Kuhn, Sam R Alapati, Bill Padfield

Verlag: Apress

insite
SUCHEN

Über dieses Buch

Speed up the execution of important database queries by making good choices about which indexes to create. Choose correct index types for different scenarios. Avoid indexing pitfalls that can actually have indexes hurting performance rather than helping. Maintain indexes so as to provide consistent and predictable query response over the lifetime of an application.

Expert Oracle Indexing and Access Paths is about the one database structure at the heart of almost all performance concerns: the index. Database system performance is one of the top concerns in information technology today. Administrators struggle to keep up with the explosion of access and activity driven by the proliferation of computing into everything from phones to tablets to PCs in our increasingly connected world. At the heart of any good-performing database lies a sound indexing strategy that makes appropriate use of indexing, and especially of the vendor-specific indexing features on offer.

Few databases fully exploit the wealth of data access mechanisms provided by Oracle. Expert Oracle Indexing and Access Paths helps by bringing together information on indexing and how to use it into one blissfully short volume that you can read quickly and have at your fingertips for reference. Learn the different types of indexes available and when each is best applied. Recognize when queries aren’t using indexes as you intend. Manage your indexing for maximum performance. Confidently use the In Memory column store feature as an alternate access path to improve performance. Let Expert Indexing in Oracle Database 12c be your guide to deep mastery of the most fundamental performance optimization structure in Oracle Database.

Explains how indexes help performance, and sometimes hinder it tooDemystifies the various index choices so that you can chose rightlyDescribes the database administration chores associated with indexesDemonstrates the use of the In Memory column store as an alternate access path to the data

What You Will Learn

Create an overall indexing strategy to guide your decisions

Choose the correct indexing mechanisms for your applications

Manage and maintain indices to avoid degradation and preserve efficiency

Take better advantage of underused index types such as index-organized tables

Choose the appropriate columns to index, with confidence

Blend partitioning and materialized views into your indexing strategy

Who This Book Is For

Expert Oracle Indexing and Access Paths is for all levels of database administrators and application developers who are struggling with the database performance and scalability challenge. Any database administrator involved with indexing, which is any database administrator period, will appreciate the wealth of advice packed into this gem of a book.

Inhaltsverzeichnis

Frontmatter
Chapter 1. Introduction to Oracle Indexes
Abstract
Similar to the way you use indexes in everyday life, a database index primarily exists to enable fast access to one or more rows of information stored in a database.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Chapter 2. B-tree Indexes
Abstract
The B-tree index is the default index type in Oracle.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Chapter 3. Bitmap Indexes
Abstract
Bitmap indexes are best suited for data warehouse or decision support systems (DSS). Common to the data warehouse is the star schema, in which a central fact table contains all the detailed information for a particular subject, such as customer revenue, and the number of related dimension tables containing associated reference type data on a particular dimension, such as time or geography. In a star schema, the dimension tables are the parent and the central fact table is the child table.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Chapter 4. Index-Organized Tables
Abstract
The simplest explanation of an index-organized table is that it is accessed like any other Oracle table (typically a heap-organized table) but is physically stored like an Oracle B-tree index. Index-organized tables are typically created on “thin” tables (tables without too many columns). Typically, multiple columns of the table make up the primary key of the index-organized table. The non-key columns can also be stored as part of the B-tree index. The proper configuration and use of index-organized tables is fairly specific and does not meet all application needs.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Chapter 5. Specialized Indexes
Abstract
Thus far, you’ve learned how to create normal Oracle indexes, which include both B-tree indexes as well as bitmap indexes. You also learned how to create and manage index organized tables. This chapter explains several specialized indexes that you can create for serving various needs. Most of these specialized indexes are actually B-tree indexes, so it’s not the organization of the index that makes them special. For example, an invisible index is an index that’s not automatically available to the cost-based optimizer, unless you make it available by setting a specific initialization parameter. You use invisible indexes mainly for testing the use of an index and to make sure that an index is redundant before you drop it.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Chapter 6. Partitioned Indexes
Abstract
If you use partitioned tables as part of your application, it is likely, if not imperative, that you use partitioned indexes to complement the advantages gained by using table partitioning. Usually, having partitioned table and indexes go hand in hand—when there’s one, there’s usually both. This is common, but not essential. It is possible to have partitioned tables without partitioned indexes, and it is possible to have a non-partitioned table with partitioned indexes. There are several factors that affect the design of the database tables and indexes, including the following.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Chapter 7. Tuning Index Usage
Abstract
To tune index usage, it’s important to understand the various types of index access paths available to the cost optimizer. This chapter summarizes the most important index access paths available to the optimizer. Often, Oracle database administrators are bewildered when the optimizer chooses not to use what they believe is a very useful index, but does a full table scan instead. This chapter devotes attention to exploring some reasons why the optimizer might prefer a full table scan to using an index. You’ll also learn how to force the optimizer to use (or not to use) an index. Oracle provides a large number of index-related hints to enable you to control the behavior of the optimizer regarding index usage. You’ll find descriptions of the key index-related hints in this chapter.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Chapter 8. Maintaining Indexes
Abstract
Maintaining indexes is a big part of an Oracle DBA’s workload. There are many aspects to maintaining indexes, and often there are multiple ways to achieve the same goals. This chapter explains several key aspects of index maintenance to help you improve the performance of your indexes and to efficiently manage index space usage.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Chapter 9. SQL Tuning Advisor
Abstract
To fully understand how the SQL Tuning Advisor tool works, we need to lay some groundwork and explain a few terms. Firstly, the Oracle query optimizer operates in two different modes: normal and tuning. When a SQL statement executes, the optimizer operates in normal mode and quickly identifies a reasonable execution plan. In this mode, the optimizer spends only a fraction of a second to determine the optimal plan.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Chapter 10. In-Memory Column Store
Abstract
When issuing a query to retrieve data from the database, the Oracle query optimizer quickly determines the most efficient way to return the requested information. The “path” to the data consists of accessing the table, or using an index, or a combination of tables and indexes. In this mode, the path entails Oracle reading table blocks from disk into memory in order to access the data. Within the blocks, rows of table data are stored. This method is tried and true and works well.
Darl Kuhn, Sam R. Alapati, Bill Padfield
Backmatter
Metadaten
Titel
Expert Oracle Indexing and Access Paths
verfasst von
Darl Kuhn
Sam R Alapati
Bill Padfield
Copyright-Jahr
2016
Verlag
Apress
Electronic ISBN
978-1-4842-1984-3
Print ISBN
978-1-4842-1983-6
DOI
https://doi.org/10.1007/978-1-4842-1984-3

Premium Partner