Training Cloudera

Training goals

code: CLOUDERA-ASAPT

This three-day hands-on training course delivers the key concepts and expertise developers need to improve the performance of their Apache Spark applications. During the course, participants will learn how to identify common sources of poor performance in Spark applications, techniques for avoiding or solving them, and best practices for Spark application monitoring.

Apache Spark Application Performance Tuning presents the architecture and concepts behind Apache Spark and underlying data platform, then builds on this foundational understanding by teaching students how to tune Spark application code. The course format emphasizes instructor-led demonstrations illustrate both performance issues and the techniques that address them, followed by hands-on exercises that give students an opportunity to practice what they've learned through an interactive notebook environment. The course applies to Spark 2.4, but also introduces the Spark 3.0 Adaptive Query Execution framework.

What You Will Learn

Students who successfully complete this course will be able to:

  • Understand Apache Spark's architecture, job execution, and how techniques such as lazy execution and pipelining can improve runtime performance
  • Evaluate the performance characteristics of core data structures such as RDD and DataFrames
  • Select the file formats that will provide the best performance for your application
  • Identify and resolve performance problems caused by data skew
  • Use partitioning, bucketing, and join optimizations to improve SparkSQL performance
  • Understand the performance overhead of Python-based RDDs, DataFrames, and user-defined functions
  • Take advantage of caching for better application performance
  • Understand how the Catalyst and Tungsten optimizers work
  • Understand how Workload XM can help troubleshoot and proactively monitor Spark applications performance
  • Learn about the new features in Spark 3.0 and specifically how the Adaptive Query Execution engine improves performance

What to Expect

This course is designed for software developers, engineers, and data scientists who have experience developing Spark applications and want to learn how to improve the performance of their code. This is not an introduction to Spark.

Spark examples and hands-on exercises are presented in Python and the ability to program in this language is required. Basic familiarity with the Linux command line is assumed. Basic knowledge of SQL is helpful.

Conspect Show list

  • Spark Architecture
    • RDDs
    • DataFrames and Datasets
    • Lazy Evaluation
    • Pipelining
  • Data Sources and Formats
    • Available Formats Overview
    • Impact on Performance
    • The Small Files Problem
  • Inferring Schemas
    • The Cost of Inference
    • Mitigating Tactics
  • Dealing With Skewed Data
    • Recognizing Skew
    • Mitigating Tactics
  • Catalyst and Tungsten Overview
    • Catalyst Overview
    • Tungsten Overview
  • Mitigating Spark Shuffles
    • Denormalization
    • Broadcast Joins
    • Map-Side Operations
    • Sort Merge Joins
  • Partitioned and Bucketed Tables
    • Partitioned Tables
    • Bucketed Tables
    • Impact on Performance
  • Improving Join Performance
    • Skewed Joins
    • Bucketed Joins
    • Incremental Joins
  • Pyspark Overhead and UDFs
    • Pyspark Overhead
    • Scalar UDFs
    • Vector UDFs using Apache Arrow
    • Scala UDFs
  • Caching Data for Reuse
    • Caching Options
    • Impact on Performance
    • Caching Pitfalls
  • Workload XM (WXM) Introduction
    • WXM Overview
    • WXM for Spark Developers
  • What's New in Spark 3.0?
    • Adaptive Number of Shuffle Partitions
    • Skew Joins
    • Convert Sort Merge Joins to Broadcast Joins
    • Dynamic Partition Pruning
    • Dynamic Coalesce Shuffle Partitions
Download conspect training as PDF

Additional information

Prerequisites

This course is designed for software developers, engineers, and data scientists who have experience developing Spark applications and want to learn how to improve the performance of their code. This is not an introduction to Spark.

Spark examples and hands-on exercises are presented in Python and the ability to program in this language is required. Basic familiarity with the Linux command line is assumed. Basic knowledge of SQL is helpful.
Difficulty level
Duration 3 days
Certificate

The participants will obtain certificates signed by Cloudera (course completion).

Upon completion of the course, attendees are encouraged to continue their study and register for the CDP Data Developer exam https://www.cloudera.com/about/training/certification/cdp-datadev-exam-cdp-3001.html

Certification is a great differentiator. It helps establish you as a leader in the field, providing employers and customers with tangible evidence of your skills and expertise.
Trainer

Certified Cloudera Instructor

Other training Cloudera | Cloudera Data Developer

Cloudera show more courses
Contact form

Please fill form below to obtain more info about this training.






* Fields marked with (*) are required !!!

Information on data processing by Compendium - Centrum Edukacyjne Spółka z o.o.

PRICE 2200 EUR

FORM OF TRAINING ?

 

TRAINING MATERIALS ?

 

SELECT TRAINING DATE

  • hybrid training: HYBRID
    • General information
    • Guaranteed dates
    • Last minute (-10%)
    • Language of the training
    • English
  • hybrid training: HYBRID
    • General information
    • Guaranteed dates
    • Last minute (-10%)
    • Language of the training
    • English
Book a training appointment
close

Traditional training

Sessions organised at Compendium CE are usually held in our locations in Kraków and Warsaw, but also in venues designated by the client. The group participating in training meets at a specific place and specific time with a coach and actively participates in laboratory sessions.

Dlearning training

You may participate from at any place in the world. It is sufficient to have a computer (or, actually a tablet, or smartphone) connected to the Internet. Compendium CE provides each Distance Learning training participant with adequate software enabling connection to the Data Center. For more information, please visit dlearning.eu site

close

Paper materials

Traditional materials: The price includes standard materials issued in the form of paper books, printed or other, depending on the arrangements with the manufacturer.

Electronic materials

Electronic materials: These are electronic training materials that are available to you based on your specific application: Skillpipe, eVantage, etc., or as PDF documents.

Ctab materials

Ctab materials: the price includes ctab tablet and electronic training materials or traditional training materials and supplies provided electronically according to manufacturer's specifications (in PDF or EPUB form). The materials provided are adapted for display on ctab tablets. For more information, check out the ctab website.

Upcoming Cloudera training

Training schedule Cloudera