Exadata For Developers (ENG)

Why attend this workshop?

The Oracle Exadata Database Machine comes with a set of unique features. A thorough understanding of these features is a prerequisite for making the most out of the platform. Application database developers need to be aware of these features, how they work and under what circumstances they might not work as expected. Quite often developers are faced with the task of migrating existing applications to the Exadata platform - the workshop also covers how to analyse an existing application regarding the usage of Exadata features - including to what degree the application should be able to benefit from Exadata without changes, and what parts of the application could be optimized to maximize the usage of Exadata features.

Objectives

- Get a deep and thorough understanding of the unique features of Oracle's Exadata Database Machine

- Learn how to measure and analyse the usage of Exadata features

- Understand the new Exadata features added since 12c/18c/19c releases

Required skills

Thorough understanding of general Oracle Architecture, SQL, SQL*Plus scripting

Target audience

Database application developers and DBAs

Course description

The Oracle Exadata Database Machine comes with a set of unique features. Starting with explaining the Exadata architecture we will then look then in detail at the Smart Scan feature, what to consider regarding execution plans, what the Smart Scan prerequisites are, how to control the usage, and in particular how to measure in detail the effectiveness of using Smart Scans. We'll then cover in detail what different means are available to monitor the Exadata feature usage on different levels. Storage Indexes will also be covered in detail along with the new Attribute Clustering and Zone Maps features added in Oracle 12c. Next will be a detailed look at the Exadata Flash Cache, continuing with the Hybrid Columnar Compression features. Finally we'll look at Parallel Execution in the context of Exadata and what you should consider to make the most out of the platform. 

For each feature you'll learn and understand what data and activity patterns might lead to different behaviour and level of efficiency of the feature, which also allows you as a developer to judge whether feature will be helpful to your application or not, respectively what you might have to change to make better usage of the feature.

All topics covered come with extensive demo scripts that allow reproducing the feature and effects explained.

 Topics covered

- Introduction to Oracle's Exadata Database Machine

- Offloading - Smart Scans

- Smart Scans and execution plans

- Smart Scans deep dive: Prerequisites, (serial) direct path reads vs. buffered reads, disablers, predicate offloading

- Monitor and measure Exadata feature usage including ready to use scripts

- New wait events, Exadata related statistics, Cell Offload Efficiency, (G)V$SQL* views, AWR, Real-Time SQL Monitoring

- Storage Indexes, Attribute Clustering, Zone Maps

- Why not every Smart Scan is efficient

- Exadata Flash Cache details

- Hybrid Columnar Compression deep dive: Internal format, compression levels, DML, locking patterns, single row access, schema evolution

- Exadata and Parallel Execution: Hash Join Buffered, Physical design, partition wise operations

- Analyse existing applications for Exadata feature usage