SlideShare a Scribd company logo
Data Storage Tips for
Optimal Spark Performance
Vida Ha
Spark Summit West 2015
Today’s Talk
About Me
Vida Ha - Solutions Engineer at Databricks
Poor Data File Storage Choices Result in:
• Exceptions that are difficult to diagnose and fix.
• Slow Spark Jobs.
• Inaccurate Data Analysis.
2
Goal: Understand the Best Practices for Storing and
working with Data in Files with
Agenda
• Basic Data Storage Decisions
• File Sizes
• Compression Formats
• Data Format Best Practices
• Plain Text, Structured Text, Data Interchange, Columnar
• General Tips
3
4
Basic Data Storage Decisions
Choosing a File Size
• Too Small
• Lots of time opening file handles.
• Too Big
• Files need to be splittable.
• Optimize for your File System
• Good Rule of Thumb: 64 MB - 1 GB
5
Choosing a Compression Format
• The Obvious
• Minimize the Compressed Size of Files.
• Decoding characteristics.
• Pay attention to Splittable vs. NonSplittable.
• Common Compression Formats for Big Data:
• gzip, Snappy, bzip2, LZO, and LZ4.
• Columnar formats for Structured Data - Parquet.
6
7
Data Format Best Practices
Plain Text
• sc.textFile() splits the file into lines.
• So keep your lines a reasonable size.
• Or use a different method to read data in.
• Keep file sizes < 1 GB if compressed with a
non-splittable format.
8
Basic Structured Text Tiles
• Use Spark transformations to ETL the data.
• Optionally use Spark SQL for analyzing.
• Handle the inevitable malformed data with
Spark:
• Use a filter transformation to drop bad lines.
• Or use a map function to fix bad lines.
• Includes CSV, JSON, XML.
9
CSV
• Use the relatively new Spark-CSV package.
• Spark SQL Malformed Data Tip:
• Use a where clause and sanity check fields.
10
JSON
• Ideally have one JSON object per line.
• Otherwise, DIY parsing the JSON.
• Prefer specifying a schema over inferSchema.
• Watch out for arbitrary number of keys.
• Inferring Schema will result in an executor OOM error.
• Spark SQL Malformed Data Tip:
• Bad inputs have a column called _corrupt_record and other
columns will be null.
11
XML
• Not an ideal Big Data Format.
• Typically not one XML object per line.
• So you’ll need to DIY parse.
• Not a lot of built in library support for XML.
• Watch out for very large compressed files.
• May need to employ the General Tips to parse.
12
Data Interchange Formats with a Schema
• Good Practice to enforce an API with backward
compatibility.
• Avro, Parquet, and Thrift are common ones.
• Usually, good compression.
• Data format itself is not corrupt.
• But underlying records still be.
13
Avro
• Use DataFile Writer to write multiple objects.
• Use spark-avro package to read in.
• Don’t transfer “AvroKey” across driver and
workers.
• AvroKey is not serializable.
• Pull out fields of interest or convert to JSON.
14
Protocol Buffers
• Need to figure out how to encode multiple in a
file.
• Encode in Sequence Files or other similar file format.
• Prepend the number of bytes before reading the next
message.
• Base64 encode one per line.
• Currently, no built in Spark package to support.
• Opportunity to contribute to open source community.
• For now, convert to JSON and read into Spark SQL.
15
Columnar Formats: Parquet & ORD
• Great for use with Spark SQL.
• Parquet is actually best practice for Spark SQL.
• Makes it easy to pull only certain records at a
time.
• Worth time to encode if multiple passes on
data.
• Do not support appending one record at a time.
• Not good for collecting log records.
16
17
General Tips
Reuse Hadoop Libraries
• HadoopRDD & NewHadoopRDD
• Reuse Hadoop File format libraries.
• Hive SerDe’s are supported in Spark SQL
• Load your SerDe jar onto your Spark Cluster.
• Issue a SHOW CREATE TABLE command.
• Enter the create table command (with EXTERNAL)
18
Controlling the Output File Size
• Spark generally writes one file per partition.
• Use coalesce to write out less files.
• Use repartition to write out more files.
• This will cause a shuffle of the data.
• If repartition is too expensive:
• Call mapPartition to get an iterator and write out as many
(or few) files as you would like.
19
sc.wholeTextFiles()
• The whole file should fit into memory.
• Good for file formats that aren’t splittable by
line.
• Such as XML files.
• Will need to performance tune.
20
Use “filename” as the RDD element
• filenames = sc.parallelize([“s3:/file1”, “s3:/
file2”, …])
• Allows you to function on a single file.
• filenames.map(…call_function_on_filename)
• Use to decode non-standard compression
formats.
• Use to split up files that aren’t separable by
line.
21
Thank you.
Any questions?

More Related Content

What's hot

A Deep Dive into Query Execution Engine of Spark SQL
A Deep Dive into Query Execution Engine of Spark SQLA Deep Dive into Query Execution Engine of Spark SQL
A Deep Dive into Query Execution Engine of Spark SQL
Databricks
 
How We Optimize Spark SQL Jobs With parallel and sync IO
How We Optimize Spark SQL Jobs With parallel and sync IOHow We Optimize Spark SQL Jobs With parallel and sync IO
How We Optimize Spark SQL Jobs With parallel and sync IO
Databricks
 
Physical Plans in Spark SQL
Physical Plans in Spark SQLPhysical Plans in Spark SQL
Physical Plans in Spark SQL
Databricks
 
Memory Management in Apache Spark
Memory Management in Apache SparkMemory Management in Apache Spark
Memory Management in Apache Spark
Databricks
 
The Parquet Format and Performance Optimization Opportunities
The Parquet Format and Performance Optimization OpportunitiesThe Parquet Format and Performance Optimization Opportunities
The Parquet Format and Performance Optimization Opportunities
Databricks
 
Parquet performance tuning: the missing guide
Parquet performance tuning: the missing guideParquet performance tuning: the missing guide
Parquet performance tuning: the missing guide
Ryan Blue
 
Apache Spark Data Source V2 with Wenchen Fan and Gengliang Wang
Apache Spark Data Source V2 with Wenchen Fan and Gengliang WangApache Spark Data Source V2 with Wenchen Fan and Gengliang Wang
Apache Spark Data Source V2 with Wenchen Fan and Gengliang Wang
Databricks
 
Deep Dive: Memory Management in Apache Spark
Deep Dive: Memory Management in Apache SparkDeep Dive: Memory Management in Apache Spark
Deep Dive: Memory Management in Apache Spark
Databricks
 
Building a SIMD Supported Vectorized Native Engine for Spark SQL
Building a SIMD Supported Vectorized Native Engine for Spark SQLBuilding a SIMD Supported Vectorized Native Engine for Spark SQL
Building a SIMD Supported Vectorized Native Engine for Spark SQL
Databricks
 
Dynamic Partition Pruning in Apache Spark
Dynamic Partition Pruning in Apache SparkDynamic Partition Pruning in Apache Spark
Dynamic Partition Pruning in Apache Spark
Databricks
 
Introduction to Apache Spark
Introduction to Apache SparkIntroduction to Apache Spark
Introduction to Apache Spark
Rahul Jain
 
Understanding Query Plans and Spark UIs
Understanding Query Plans and Spark UIsUnderstanding Query Plans and Spark UIs
Understanding Query Plans and Spark UIs
Databricks
 
Enabling Vectorized Engine in Apache Spark
Enabling Vectorized Engine in Apache SparkEnabling Vectorized Engine in Apache Spark
Enabling Vectorized Engine in Apache Spark
Kazuaki Ishizaki
 
Near Real-Time Data Warehousing with Apache Spark and Delta Lake
Near Real-Time Data Warehousing with Apache Spark and Delta LakeNear Real-Time Data Warehousing with Apache Spark and Delta Lake
Near Real-Time Data Warehousing with Apache Spark and Delta Lake
Databricks
 
Hudi architecture, fundamentals and capabilities
Hudi architecture, fundamentals and capabilitiesHudi architecture, fundamentals and capabilities
Hudi architecture, fundamentals and capabilities
Nishith Agarwal
 
Advanced Apache Spark Meetup Project Tungsten Nov 12 2015
Advanced Apache Spark Meetup Project Tungsten Nov 12 2015Advanced Apache Spark Meetup Project Tungsten Nov 12 2015
Advanced Apache Spark Meetup Project Tungsten Nov 12 2015
Chris Fregly
 
Hoodie - DataEngConf 2017
Hoodie - DataEngConf 2017Hoodie - DataEngConf 2017
Hoodie - DataEngConf 2017
Vinoth Chandar
 
Processing Large Data with Apache Spark -- HasGeek
Processing Large Data with Apache Spark -- HasGeekProcessing Large Data with Apache Spark -- HasGeek
Processing Large Data with Apache Spark -- HasGeek
Venkata Naga Ravi
 
Designing Structured Streaming Pipelines—How to Architect Things Right
Designing Structured Streaming Pipelines—How to Architect Things RightDesigning Structured Streaming Pipelines—How to Architect Things Right
Designing Structured Streaming Pipelines—How to Architect Things Right
Databricks
 
Programming in Spark using PySpark
Programming in Spark using PySpark      Programming in Spark using PySpark
Programming in Spark using PySpark
Mostafa
 

What's hot (20)

A Deep Dive into Query Execution Engine of Spark SQL
A Deep Dive into Query Execution Engine of Spark SQLA Deep Dive into Query Execution Engine of Spark SQL
A Deep Dive into Query Execution Engine of Spark SQL
 
How We Optimize Spark SQL Jobs With parallel and sync IO
How We Optimize Spark SQL Jobs With parallel and sync IOHow We Optimize Spark SQL Jobs With parallel and sync IO
How We Optimize Spark SQL Jobs With parallel and sync IO
 
Physical Plans in Spark SQL
Physical Plans in Spark SQLPhysical Plans in Spark SQL
Physical Plans in Spark SQL
 
Memory Management in Apache Spark
Memory Management in Apache SparkMemory Management in Apache Spark
Memory Management in Apache Spark
 
The Parquet Format and Performance Optimization Opportunities
The Parquet Format and Performance Optimization OpportunitiesThe Parquet Format and Performance Optimization Opportunities
The Parquet Format and Performance Optimization Opportunities
 
Parquet performance tuning: the missing guide
Parquet performance tuning: the missing guideParquet performance tuning: the missing guide
Parquet performance tuning: the missing guide
 
Apache Spark Data Source V2 with Wenchen Fan and Gengliang Wang
Apache Spark Data Source V2 with Wenchen Fan and Gengliang WangApache Spark Data Source V2 with Wenchen Fan and Gengliang Wang
Apache Spark Data Source V2 with Wenchen Fan and Gengliang Wang
 
Deep Dive: Memory Management in Apache Spark
Deep Dive: Memory Management in Apache SparkDeep Dive: Memory Management in Apache Spark
Deep Dive: Memory Management in Apache Spark
 
Building a SIMD Supported Vectorized Native Engine for Spark SQL
Building a SIMD Supported Vectorized Native Engine for Spark SQLBuilding a SIMD Supported Vectorized Native Engine for Spark SQL
Building a SIMD Supported Vectorized Native Engine for Spark SQL
 
Dynamic Partition Pruning in Apache Spark
Dynamic Partition Pruning in Apache SparkDynamic Partition Pruning in Apache Spark
Dynamic Partition Pruning in Apache Spark
 
Introduction to Apache Spark
Introduction to Apache SparkIntroduction to Apache Spark
Introduction to Apache Spark
 
Understanding Query Plans and Spark UIs
Understanding Query Plans and Spark UIsUnderstanding Query Plans and Spark UIs
Understanding Query Plans and Spark UIs
 
Enabling Vectorized Engine in Apache Spark
Enabling Vectorized Engine in Apache SparkEnabling Vectorized Engine in Apache Spark
Enabling Vectorized Engine in Apache Spark
 
Near Real-Time Data Warehousing with Apache Spark and Delta Lake
Near Real-Time Data Warehousing with Apache Spark and Delta LakeNear Real-Time Data Warehousing with Apache Spark and Delta Lake
Near Real-Time Data Warehousing with Apache Spark and Delta Lake
 
Hudi architecture, fundamentals and capabilities
Hudi architecture, fundamentals and capabilitiesHudi architecture, fundamentals and capabilities
Hudi architecture, fundamentals and capabilities
 
Advanced Apache Spark Meetup Project Tungsten Nov 12 2015
Advanced Apache Spark Meetup Project Tungsten Nov 12 2015Advanced Apache Spark Meetup Project Tungsten Nov 12 2015
Advanced Apache Spark Meetup Project Tungsten Nov 12 2015
 
Hoodie - DataEngConf 2017
Hoodie - DataEngConf 2017Hoodie - DataEngConf 2017
Hoodie - DataEngConf 2017
 
Processing Large Data with Apache Spark -- HasGeek
Processing Large Data with Apache Spark -- HasGeekProcessing Large Data with Apache Spark -- HasGeek
Processing Large Data with Apache Spark -- HasGeek
 
Designing Structured Streaming Pipelines—How to Architect Things Right
Designing Structured Streaming Pipelines—How to Architect Things RightDesigning Structured Streaming Pipelines—How to Architect Things Right
Designing Structured Streaming Pipelines—How to Architect Things Right
 
Programming in Spark using PySpark
Programming in Spark using PySpark      Programming in Spark using PySpark
Programming in Spark using PySpark
 

Viewers also liked

Solving Low Latency Query Over Big Data with Spark SQL-(Julien Pierre, Micros...
Solving Low Latency Query Over Big Data with Spark SQL-(Julien Pierre, Micros...Solving Low Latency Query Over Big Data with Spark SQL-(Julien Pierre, Micros...
Solving Low Latency Query Over Big Data with Spark SQL-(Julien Pierre, Micros...
Spark Summit
 
How to understand and analyze Apache Hive query execution plan for performanc...
How to understand and analyze Apache Hive query execution plan for performanc...How to understand and analyze Apache Hive query execution plan for performanc...
How to understand and analyze Apache Hive query execution plan for performanc...
DataWorks Summit/Hadoop Summit
 
SQL to Hive Cheat Sheet
SQL to Hive Cheat SheetSQL to Hive Cheat Sheet
SQL to Hive Cheat Sheet
Hortonworks
 
Apache Spark 2.0: Faster, Easier, and Smarter
Apache Spark 2.0: Faster, Easier, and SmarterApache Spark 2.0: Faster, Easier, and Smarter
Apache Spark 2.0: Faster, Easier, and Smarter
Databricks
 
Introduction to Spark Internals
Introduction to Spark InternalsIntroduction to Spark Internals
Introduction to Spark Internals
Pietro Michiardi
 
Apache Spark Architecture
Apache Spark ArchitectureApache Spark Architecture
Apache Spark Architecture
Alexey Grishchenko
 

Viewers also liked (6)

Solving Low Latency Query Over Big Data with Spark SQL-(Julien Pierre, Micros...
Solving Low Latency Query Over Big Data with Spark SQL-(Julien Pierre, Micros...Solving Low Latency Query Over Big Data with Spark SQL-(Julien Pierre, Micros...
Solving Low Latency Query Over Big Data with Spark SQL-(Julien Pierre, Micros...
 
How to understand and analyze Apache Hive query execution plan for performanc...
How to understand and analyze Apache Hive query execution plan for performanc...How to understand and analyze Apache Hive query execution plan for performanc...
How to understand and analyze Apache Hive query execution plan for performanc...
 
SQL to Hive Cheat Sheet
SQL to Hive Cheat SheetSQL to Hive Cheat Sheet
SQL to Hive Cheat Sheet
 
Apache Spark 2.0: Faster, Easier, and Smarter
Apache Spark 2.0: Faster, Easier, and SmarterApache Spark 2.0: Faster, Easier, and Smarter
Apache Spark 2.0: Faster, Easier, and Smarter
 
Introduction to Spark Internals
Introduction to Spark InternalsIntroduction to Spark Internals
Introduction to Spark Internals
 
Apache Spark Architecture
Apache Spark ArchitectureApache Spark Architecture
Apache Spark Architecture
 

Similar to Data Storage Tips for Optimal Spark Performance-(Vida Ha, Databricks)

Why you should care about data layout in the file system with Cheng Lian and ...
Why you should care about data layout in the file system with Cheng Lian and ...Why you should care about data layout in the file system with Cheng Lian and ...
Why you should care about data layout in the file system with Cheng Lian and ...
Databricks
 
Big Data (NJ SQL Server User Group)
Big Data (NJ SQL Server User Group)Big Data (NJ SQL Server User Group)
Big Data (NJ SQL Server User Group)Don Demcsak
 
Investigate TempDB Like Sherlock Holmes
Investigate TempDB Like Sherlock HolmesInvestigate TempDB Like Sherlock Holmes
Investigate TempDB Like Sherlock Holmes
Richard Douglas
 
Not Your Father's Database: How to Use Apache Spark Properly in Your Big Data...
Not Your Father's Database: How to Use Apache Spark Properly in Your Big Data...Not Your Father's Database: How to Use Apache Spark Properly in Your Big Data...
Not Your Father's Database: How to Use Apache Spark Properly in Your Big Data...
Databricks
 
Not Your Father's Database by Vida Ha
Not Your Father's Database by Vida HaNot Your Father's Database by Vida Ha
Not Your Father's Database by Vida Ha
Spark Summit
 
Elasticsearch Arcihtecture & What's New in Version 5
Elasticsearch Arcihtecture & What's New in Version 5Elasticsearch Arcihtecture & What's New in Version 5
Elasticsearch Arcihtecture & What's New in Version 5
Burak TUNGUT
 
20120606 Lazy Programmers Write Self-Modifying Code /or/ Dealing with XML Ord...
20120606 Lazy Programmers Write Self-Modifying Code /or/ Dealing with XML Ord...20120606 Lazy Programmers Write Self-Modifying Code /or/ Dealing with XML Ord...
20120606 Lazy Programmers Write Self-Modifying Code /or/ Dealing with XML Ord...
David Horvath
 
Taming the resource tiger
Taming the resource tigerTaming the resource tiger
Taming the resource tiger
Elizabeth Smith
 
Column Stores and Google BigQuery
Column Stores and Google BigQueryColumn Stores and Google BigQuery
Column Stores and Google BigQuery
Csaba Toth
 
Storage talk
Storage talkStorage talk
Storage talkchristkv
 
Parquet and impala overview external
Parquet and impala overview externalParquet and impala overview external
Parquet and impala overview external
mattlieber
 
Apache Spark Fundamentals
Apache Spark FundamentalsApache Spark Fundamentals
Apache Spark Fundamentals
Zahra Eskandari
 
Taming the resource tiger
Taming the resource tigerTaming the resource tiger
Taming the resource tiger
Elizabeth Smith
 
Storage in hadoop
Storage in hadoopStorage in hadoop
Storage in hadoop
Puneet Tripathi
 
Build a modern data platform.pptx
Build a modern data platform.pptxBuild a modern data platform.pptx
Build a modern data platform.pptx
Ike Ellis
 
MongoDB Replication fundamentals - Desert Code Camp - October 2014
MongoDB Replication fundamentals - Desert Code Camp - October 2014MongoDB Replication fundamentals - Desert Code Camp - October 2014
MongoDB Replication fundamentals - Desert Code Camp - October 2014
clairvoyantllc
 
Optimizing Hive Queries
Optimizing Hive QueriesOptimizing Hive Queries
Optimizing Hive Queries
Owen O'Malley
 
Optimizing Hive Queries
Optimizing Hive QueriesOptimizing Hive Queries
Optimizing Hive Queries
DataWorks Summit
 
Best Practices for NoSQL Workloads on Amazon EC2 and Amazon EBS - February 20...
Best Practices for NoSQL Workloads on Amazon EC2 and Amazon EBS - February 20...Best Practices for NoSQL Workloads on Amazon EC2 and Amazon EBS - February 20...
Best Practices for NoSQL Workloads on Amazon EC2 and Amazon EBS - February 20...
Amazon Web Services
 
Scalability, Availability & Stability Patterns
Scalability, Availability & Stability PatternsScalability, Availability & Stability Patterns
Scalability, Availability & Stability Patterns
Jonas Bonér
 

Similar to Data Storage Tips for Optimal Spark Performance-(Vida Ha, Databricks) (20)

Why you should care about data layout in the file system with Cheng Lian and ...
Why you should care about data layout in the file system with Cheng Lian and ...Why you should care about data layout in the file system with Cheng Lian and ...
Why you should care about data layout in the file system with Cheng Lian and ...
 
Big Data (NJ SQL Server User Group)
Big Data (NJ SQL Server User Group)Big Data (NJ SQL Server User Group)
Big Data (NJ SQL Server User Group)
 
Investigate TempDB Like Sherlock Holmes
Investigate TempDB Like Sherlock HolmesInvestigate TempDB Like Sherlock Holmes
Investigate TempDB Like Sherlock Holmes
 
Not Your Father's Database: How to Use Apache Spark Properly in Your Big Data...
Not Your Father's Database: How to Use Apache Spark Properly in Your Big Data...Not Your Father's Database: How to Use Apache Spark Properly in Your Big Data...
Not Your Father's Database: How to Use Apache Spark Properly in Your Big Data...
 
Not Your Father's Database by Vida Ha
Not Your Father's Database by Vida HaNot Your Father's Database by Vida Ha
Not Your Father's Database by Vida Ha
 
Elasticsearch Arcihtecture & What's New in Version 5
Elasticsearch Arcihtecture & What's New in Version 5Elasticsearch Arcihtecture & What's New in Version 5
Elasticsearch Arcihtecture & What's New in Version 5
 
20120606 Lazy Programmers Write Self-Modifying Code /or/ Dealing with XML Ord...
20120606 Lazy Programmers Write Self-Modifying Code /or/ Dealing with XML Ord...20120606 Lazy Programmers Write Self-Modifying Code /or/ Dealing with XML Ord...
20120606 Lazy Programmers Write Self-Modifying Code /or/ Dealing with XML Ord...
 
Taming the resource tiger
Taming the resource tigerTaming the resource tiger
Taming the resource tiger
 
Column Stores and Google BigQuery
Column Stores and Google BigQueryColumn Stores and Google BigQuery
Column Stores and Google BigQuery
 
Storage talk
Storage talkStorage talk
Storage talk
 
Parquet and impala overview external
Parquet and impala overview externalParquet and impala overview external
Parquet and impala overview external
 
Apache Spark Fundamentals
Apache Spark FundamentalsApache Spark Fundamentals
Apache Spark Fundamentals
 
Taming the resource tiger
Taming the resource tigerTaming the resource tiger
Taming the resource tiger
 
Storage in hadoop
Storage in hadoopStorage in hadoop
Storage in hadoop
 
Build a modern data platform.pptx
Build a modern data platform.pptxBuild a modern data platform.pptx
Build a modern data platform.pptx
 
MongoDB Replication fundamentals - Desert Code Camp - October 2014
MongoDB Replication fundamentals - Desert Code Camp - October 2014MongoDB Replication fundamentals - Desert Code Camp - October 2014
MongoDB Replication fundamentals - Desert Code Camp - October 2014
 
Optimizing Hive Queries
Optimizing Hive QueriesOptimizing Hive Queries
Optimizing Hive Queries
 
Optimizing Hive Queries
Optimizing Hive QueriesOptimizing Hive Queries
Optimizing Hive Queries
 
Best Practices for NoSQL Workloads on Amazon EC2 and Amazon EBS - February 20...
Best Practices for NoSQL Workloads on Amazon EC2 and Amazon EBS - February 20...Best Practices for NoSQL Workloads on Amazon EC2 and Amazon EBS - February 20...
Best Practices for NoSQL Workloads on Amazon EC2 and Amazon EBS - February 20...
 
Scalability, Availability & Stability Patterns
Scalability, Availability & Stability PatternsScalability, Availability & Stability Patterns
Scalability, Availability & Stability Patterns
 

More from Spark Summit

FPGA-Based Acceleration Architecture for Spark SQL Qi Xie and Quanfu Wang
FPGA-Based Acceleration Architecture for Spark SQL Qi Xie and Quanfu Wang FPGA-Based Acceleration Architecture for Spark SQL Qi Xie and Quanfu Wang
FPGA-Based Acceleration Architecture for Spark SQL Qi Xie and Quanfu Wang
Spark Summit
 
VEGAS: The Missing Matplotlib for Scala/Apache Spark with DB Tsai and Roger M...
VEGAS: The Missing Matplotlib for Scala/Apache Spark with DB Tsai and Roger M...VEGAS: The Missing Matplotlib for Scala/Apache Spark with DB Tsai and Roger M...
VEGAS: The Missing Matplotlib for Scala/Apache Spark with DB Tsai and Roger M...
Spark Summit
 
Apache Spark Structured Streaming Helps Smart Manufacturing with Xiaochang Wu
Apache Spark Structured Streaming Helps Smart Manufacturing with  Xiaochang WuApache Spark Structured Streaming Helps Smart Manufacturing with  Xiaochang Wu
Apache Spark Structured Streaming Helps Smart Manufacturing with Xiaochang Wu
Spark Summit
 
Improving Traffic Prediction Using Weather Data with Ramya Raghavendra
Improving Traffic Prediction Using Weather Data  with Ramya RaghavendraImproving Traffic Prediction Using Weather Data  with Ramya Raghavendra
Improving Traffic Prediction Using Weather Data with Ramya Raghavendra
Spark Summit
 
A Tale of Two Graph Frameworks on Spark: GraphFrames and Tinkerpop OLAP Artem...
A Tale of Two Graph Frameworks on Spark: GraphFrames and Tinkerpop OLAP Artem...A Tale of Two Graph Frameworks on Spark: GraphFrames and Tinkerpop OLAP Artem...
A Tale of Two Graph Frameworks on Spark: GraphFrames and Tinkerpop OLAP Artem...
Spark Summit
 
No More Cumbersomeness: Automatic Predictive Modeling on Apache Spark Marcin ...
No More Cumbersomeness: Automatic Predictive Modeling on Apache Spark Marcin ...No More Cumbersomeness: Automatic Predictive Modeling on Apache Spark Marcin ...
No More Cumbersomeness: Automatic Predictive Modeling on Apache Spark Marcin ...
Spark Summit
 
Apache Spark and Tensorflow as a Service with Jim Dowling
Apache Spark and Tensorflow as a Service with Jim DowlingApache Spark and Tensorflow as a Service with Jim Dowling
Apache Spark and Tensorflow as a Service with Jim Dowling
Spark Summit
 
Apache Spark and Tensorflow as a Service with Jim Dowling
Apache Spark and Tensorflow as a Service with Jim DowlingApache Spark and Tensorflow as a Service with Jim Dowling
Apache Spark and Tensorflow as a Service with Jim Dowling
Spark Summit
 
MMLSpark: Lessons from Building a SparkML-Compatible Machine Learning Library...
MMLSpark: Lessons from Building a SparkML-Compatible Machine Learning Library...MMLSpark: Lessons from Building a SparkML-Compatible Machine Learning Library...
MMLSpark: Lessons from Building a SparkML-Compatible Machine Learning Library...
Spark Summit
 
Next CERN Accelerator Logging Service with Jakub Wozniak
Next CERN Accelerator Logging Service with Jakub WozniakNext CERN Accelerator Logging Service with Jakub Wozniak
Next CERN Accelerator Logging Service with Jakub Wozniak
Spark Summit
 
Powering a Startup with Apache Spark with Kevin Kim
Powering a Startup with Apache Spark with Kevin KimPowering a Startup with Apache Spark with Kevin Kim
Powering a Startup with Apache Spark with Kevin Kim
Spark Summit
 
Improving Traffic Prediction Using Weather Datawith Ramya Raghavendra
Improving Traffic Prediction Using Weather Datawith Ramya RaghavendraImproving Traffic Prediction Using Weather Datawith Ramya Raghavendra
Improving Traffic Prediction Using Weather Datawith Ramya Raghavendra
Spark Summit
 
Hiding Apache Spark Complexity for Fast Prototyping of Big Data Applications—...
Hiding Apache Spark Complexity for Fast Prototyping of Big Data Applications—...Hiding Apache Spark Complexity for Fast Prototyping of Big Data Applications—...
Hiding Apache Spark Complexity for Fast Prototyping of Big Data Applications—...
Spark Summit
 
How Nielsen Utilized Databricks for Large-Scale Research and Development with...
How Nielsen Utilized Databricks for Large-Scale Research and Development with...How Nielsen Utilized Databricks for Large-Scale Research and Development with...
How Nielsen Utilized Databricks for Large-Scale Research and Development with...
Spark Summit
 
Spline: Apache Spark Lineage not Only for the Banking Industry with Marek Nov...
Spline: Apache Spark Lineage not Only for the Banking Industry with Marek Nov...Spline: Apache Spark Lineage not Only for the Banking Industry with Marek Nov...
Spline: Apache Spark Lineage not Only for the Banking Industry with Marek Nov...
Spark Summit
 
Goal Based Data Production with Sim Simeonov
Goal Based Data Production with Sim SimeonovGoal Based Data Production with Sim Simeonov
Goal Based Data Production with Sim Simeonov
Spark Summit
 
Preventing Revenue Leakage and Monitoring Distributed Systems with Machine Le...
Preventing Revenue Leakage and Monitoring Distributed Systems with Machine Le...Preventing Revenue Leakage and Monitoring Distributed Systems with Machine Le...
Preventing Revenue Leakage and Monitoring Distributed Systems with Machine Le...
Spark Summit
 
Getting Ready to Use Redis with Apache Spark with Dvir Volk
Getting Ready to Use Redis with Apache Spark with Dvir VolkGetting Ready to Use Redis with Apache Spark with Dvir Volk
Getting Ready to Use Redis with Apache Spark with Dvir Volk
Spark Summit
 
Deduplication and Author-Disambiguation of Streaming Records via Supervised M...
Deduplication and Author-Disambiguation of Streaming Records via Supervised M...Deduplication and Author-Disambiguation of Streaming Records via Supervised M...
Deduplication and Author-Disambiguation of Streaming Records via Supervised M...
Spark Summit
 
MatFast: In-Memory Distributed Matrix Computation Processing and Optimization...
MatFast: In-Memory Distributed Matrix Computation Processing and Optimization...MatFast: In-Memory Distributed Matrix Computation Processing and Optimization...
MatFast: In-Memory Distributed Matrix Computation Processing and Optimization...
Spark Summit
 

More from Spark Summit (20)

FPGA-Based Acceleration Architecture for Spark SQL Qi Xie and Quanfu Wang
FPGA-Based Acceleration Architecture for Spark SQL Qi Xie and Quanfu Wang FPGA-Based Acceleration Architecture for Spark SQL Qi Xie and Quanfu Wang
FPGA-Based Acceleration Architecture for Spark SQL Qi Xie and Quanfu Wang
 
VEGAS: The Missing Matplotlib for Scala/Apache Spark with DB Tsai and Roger M...
VEGAS: The Missing Matplotlib for Scala/Apache Spark with DB Tsai and Roger M...VEGAS: The Missing Matplotlib for Scala/Apache Spark with DB Tsai and Roger M...
VEGAS: The Missing Matplotlib for Scala/Apache Spark with DB Tsai and Roger M...
 
Apache Spark Structured Streaming Helps Smart Manufacturing with Xiaochang Wu
Apache Spark Structured Streaming Helps Smart Manufacturing with  Xiaochang WuApache Spark Structured Streaming Helps Smart Manufacturing with  Xiaochang Wu
Apache Spark Structured Streaming Helps Smart Manufacturing with Xiaochang Wu
 
Improving Traffic Prediction Using Weather Data with Ramya Raghavendra
Improving Traffic Prediction Using Weather Data  with Ramya RaghavendraImproving Traffic Prediction Using Weather Data  with Ramya Raghavendra
Improving Traffic Prediction Using Weather Data with Ramya Raghavendra
 
A Tale of Two Graph Frameworks on Spark: GraphFrames and Tinkerpop OLAP Artem...
A Tale of Two Graph Frameworks on Spark: GraphFrames and Tinkerpop OLAP Artem...A Tale of Two Graph Frameworks on Spark: GraphFrames and Tinkerpop OLAP Artem...
A Tale of Two Graph Frameworks on Spark: GraphFrames and Tinkerpop OLAP Artem...
 
No More Cumbersomeness: Automatic Predictive Modeling on Apache Spark Marcin ...
No More Cumbersomeness: Automatic Predictive Modeling on Apache Spark Marcin ...No More Cumbersomeness: Automatic Predictive Modeling on Apache Spark Marcin ...
No More Cumbersomeness: Automatic Predictive Modeling on Apache Spark Marcin ...
 
Apache Spark and Tensorflow as a Service with Jim Dowling
Apache Spark and Tensorflow as a Service with Jim DowlingApache Spark and Tensorflow as a Service with Jim Dowling
Apache Spark and Tensorflow as a Service with Jim Dowling
 
Apache Spark and Tensorflow as a Service with Jim Dowling
Apache Spark and Tensorflow as a Service with Jim DowlingApache Spark and Tensorflow as a Service with Jim Dowling
Apache Spark and Tensorflow as a Service with Jim Dowling
 
MMLSpark: Lessons from Building a SparkML-Compatible Machine Learning Library...
MMLSpark: Lessons from Building a SparkML-Compatible Machine Learning Library...MMLSpark: Lessons from Building a SparkML-Compatible Machine Learning Library...
MMLSpark: Lessons from Building a SparkML-Compatible Machine Learning Library...
 
Next CERN Accelerator Logging Service with Jakub Wozniak
Next CERN Accelerator Logging Service with Jakub WozniakNext CERN Accelerator Logging Service with Jakub Wozniak
Next CERN Accelerator Logging Service with Jakub Wozniak
 
Powering a Startup with Apache Spark with Kevin Kim
Powering a Startup with Apache Spark with Kevin KimPowering a Startup with Apache Spark with Kevin Kim
Powering a Startup with Apache Spark with Kevin Kim
 
Improving Traffic Prediction Using Weather Datawith Ramya Raghavendra
Improving Traffic Prediction Using Weather Datawith Ramya RaghavendraImproving Traffic Prediction Using Weather Datawith Ramya Raghavendra
Improving Traffic Prediction Using Weather Datawith Ramya Raghavendra
 
Hiding Apache Spark Complexity for Fast Prototyping of Big Data Applications—...
Hiding Apache Spark Complexity for Fast Prototyping of Big Data Applications—...Hiding Apache Spark Complexity for Fast Prototyping of Big Data Applications—...
Hiding Apache Spark Complexity for Fast Prototyping of Big Data Applications—...
 
How Nielsen Utilized Databricks for Large-Scale Research and Development with...
How Nielsen Utilized Databricks for Large-Scale Research and Development with...How Nielsen Utilized Databricks for Large-Scale Research and Development with...
How Nielsen Utilized Databricks for Large-Scale Research and Development with...
 
Spline: Apache Spark Lineage not Only for the Banking Industry with Marek Nov...
Spline: Apache Spark Lineage not Only for the Banking Industry with Marek Nov...Spline: Apache Spark Lineage not Only for the Banking Industry with Marek Nov...
Spline: Apache Spark Lineage not Only for the Banking Industry with Marek Nov...
 
Goal Based Data Production with Sim Simeonov
Goal Based Data Production with Sim SimeonovGoal Based Data Production with Sim Simeonov
Goal Based Data Production with Sim Simeonov
 
Preventing Revenue Leakage and Monitoring Distributed Systems with Machine Le...
Preventing Revenue Leakage and Monitoring Distributed Systems with Machine Le...Preventing Revenue Leakage and Monitoring Distributed Systems with Machine Le...
Preventing Revenue Leakage and Monitoring Distributed Systems with Machine Le...
 
Getting Ready to Use Redis with Apache Spark with Dvir Volk
Getting Ready to Use Redis with Apache Spark with Dvir VolkGetting Ready to Use Redis with Apache Spark with Dvir Volk
Getting Ready to Use Redis with Apache Spark with Dvir Volk
 
Deduplication and Author-Disambiguation of Streaming Records via Supervised M...
Deduplication and Author-Disambiguation of Streaming Records via Supervised M...Deduplication and Author-Disambiguation of Streaming Records via Supervised M...
Deduplication and Author-Disambiguation of Streaming Records via Supervised M...
 
MatFast: In-Memory Distributed Matrix Computation Processing and Optimization...
MatFast: In-Memory Distributed Matrix Computation Processing and Optimization...MatFast: In-Memory Distributed Matrix Computation Processing and Optimization...
MatFast: In-Memory Distributed Matrix Computation Processing and Optimization...
 

Recently uploaded

原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
u86oixdj
 
一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单
enxupq
 
Criminal IP - Threat Hunting Webinar.pdf
Criminal IP - Threat Hunting Webinar.pdfCriminal IP - Threat Hunting Webinar.pdf
Criminal IP - Threat Hunting Webinar.pdf
Criminal IP
 
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
pchutichetpong
 
The affect of service quality and online reviews on customer loyalty in the E...
The affect of service quality and online reviews on customer loyalty in the E...The affect of service quality and online reviews on customer loyalty in the E...
The affect of service quality and online reviews on customer loyalty in the E...
jerlynmaetalle
 
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
Tiktokethiodaily
 
一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单
ocavb
 
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
nscud
 
Q1’2024 Update: MYCI’s Leap Year Rebound
Q1’2024 Update: MYCI’s Leap Year ReboundQ1’2024 Update: MYCI’s Leap Year Rebound
Q1’2024 Update: MYCI’s Leap Year Rebound
Oppotus
 
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
vcaxypu
 
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
nscud
 
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
axoqas
 
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
ukgaet
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
ewymefz
 
一比一原版(UofS毕业证书)萨省大学毕业证如何办理
一比一原版(UofS毕业证书)萨省大学毕业证如何办理一比一原版(UofS毕业证书)萨省大学毕业证如何办理
一比一原版(UofS毕业证书)萨省大学毕业证如何办理
v3tuleee
 
一比一原版(QU毕业证)皇后大学毕业证成绩单
一比一原版(QU毕业证)皇后大学毕业证成绩单一比一原版(QU毕业证)皇后大学毕业证成绩单
一比一原版(QU毕业证)皇后大学毕业证成绩单
enxupq
 
Adjusting primitives for graph : SHORT REPORT / NOTES
Adjusting primitives for graph : SHORT REPORT / NOTESAdjusting primitives for graph : SHORT REPORT / NOTES
Adjusting primitives for graph : SHORT REPORT / NOTES
Subhajit Sahu
 
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdfSample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Linda486226
 
Empowering Data Analytics Ecosystem.pptx
Empowering Data Analytics Ecosystem.pptxEmpowering Data Analytics Ecosystem.pptx
Empowering Data Analytics Ecosystem.pptx
benishzehra469
 
一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单
ewymefz
 

Recently uploaded (20)

原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
原版制作(Deakin毕业证书)迪肯大学毕业证学位证一模一样
 
一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单
 
Criminal IP - Threat Hunting Webinar.pdf
Criminal IP - Threat Hunting Webinar.pdfCriminal IP - Threat Hunting Webinar.pdf
Criminal IP - Threat Hunting Webinar.pdf
 
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...
 
The affect of service quality and online reviews on customer loyalty in the E...
The affect of service quality and online reviews on customer loyalty in the E...The affect of service quality and online reviews on customer loyalty in the E...
The affect of service quality and online reviews on customer loyalty in the E...
 
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
1.Seydhcuxhxyxhccuuxuxyxyxmisolids 2019.pptx
 
一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单
 
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
 
Q1’2024 Update: MYCI’s Leap Year Rebound
Q1’2024 Update: MYCI’s Leap Year ReboundQ1’2024 Update: MYCI’s Leap Year Rebound
Q1’2024 Update: MYCI’s Leap Year Rebound
 
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
 
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
 
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
哪里卖(usq毕业证书)南昆士兰大学毕业证研究生文凭证书托福证书原版一模一样
 
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
 
一比一原版(UofS毕业证书)萨省大学毕业证如何办理
一比一原版(UofS毕业证书)萨省大学毕业证如何办理一比一原版(UofS毕业证书)萨省大学毕业证如何办理
一比一原版(UofS毕业证书)萨省大学毕业证如何办理
 
一比一原版(QU毕业证)皇后大学毕业证成绩单
一比一原版(QU毕业证)皇后大学毕业证成绩单一比一原版(QU毕业证)皇后大学毕业证成绩单
一比一原版(QU毕业证)皇后大学毕业证成绩单
 
Adjusting primitives for graph : SHORT REPORT / NOTES
Adjusting primitives for graph : SHORT REPORT / NOTESAdjusting primitives for graph : SHORT REPORT / NOTES
Adjusting primitives for graph : SHORT REPORT / NOTES
 
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdfSample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
Sample_Global Non-invasive Prenatal Testing (NIPT) Market, 2019-2030.pdf
 
Empowering Data Analytics Ecosystem.pptx
Empowering Data Analytics Ecosystem.pptxEmpowering Data Analytics Ecosystem.pptx
Empowering Data Analytics Ecosystem.pptx
 
一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单
 

Data Storage Tips for Optimal Spark Performance-(Vida Ha, Databricks)

  • 1. Data Storage Tips for Optimal Spark Performance Vida Ha Spark Summit West 2015
  • 2. Today’s Talk About Me Vida Ha - Solutions Engineer at Databricks Poor Data File Storage Choices Result in: • Exceptions that are difficult to diagnose and fix. • Slow Spark Jobs. • Inaccurate Data Analysis. 2 Goal: Understand the Best Practices for Storing and working with Data in Files with
  • 3. Agenda • Basic Data Storage Decisions • File Sizes • Compression Formats • Data Format Best Practices • Plain Text, Structured Text, Data Interchange, Columnar • General Tips 3
  • 5. Choosing a File Size • Too Small • Lots of time opening file handles. • Too Big • Files need to be splittable. • Optimize for your File System • Good Rule of Thumb: 64 MB - 1 GB 5
  • 6. Choosing a Compression Format • The Obvious • Minimize the Compressed Size of Files. • Decoding characteristics. • Pay attention to Splittable vs. NonSplittable. • Common Compression Formats for Big Data: • gzip, Snappy, bzip2, LZO, and LZ4. • Columnar formats for Structured Data - Parquet. 6
  • 7. 7 Data Format Best Practices
  • 8. Plain Text • sc.textFile() splits the file into lines. • So keep your lines a reasonable size. • Or use a different method to read data in. • Keep file sizes < 1 GB if compressed with a non-splittable format. 8
  • 9. Basic Structured Text Tiles • Use Spark transformations to ETL the data. • Optionally use Spark SQL for analyzing. • Handle the inevitable malformed data with Spark: • Use a filter transformation to drop bad lines. • Or use a map function to fix bad lines. • Includes CSV, JSON, XML. 9
  • 10. CSV • Use the relatively new Spark-CSV package. • Spark SQL Malformed Data Tip: • Use a where clause and sanity check fields. 10
  • 11. JSON • Ideally have one JSON object per line. • Otherwise, DIY parsing the JSON. • Prefer specifying a schema over inferSchema. • Watch out for arbitrary number of keys. • Inferring Schema will result in an executor OOM error. • Spark SQL Malformed Data Tip: • Bad inputs have a column called _corrupt_record and other columns will be null. 11
  • 12. XML • Not an ideal Big Data Format. • Typically not one XML object per line. • So you’ll need to DIY parse. • Not a lot of built in library support for XML. • Watch out for very large compressed files. • May need to employ the General Tips to parse. 12
  • 13. Data Interchange Formats with a Schema • Good Practice to enforce an API with backward compatibility. • Avro, Parquet, and Thrift are common ones. • Usually, good compression. • Data format itself is not corrupt. • But underlying records still be. 13
  • 14. Avro • Use DataFile Writer to write multiple objects. • Use spark-avro package to read in. • Don’t transfer “AvroKey” across driver and workers. • AvroKey is not serializable. • Pull out fields of interest or convert to JSON. 14
  • 15. Protocol Buffers • Need to figure out how to encode multiple in a file. • Encode in Sequence Files or other similar file format. • Prepend the number of bytes before reading the next message. • Base64 encode one per line. • Currently, no built in Spark package to support. • Opportunity to contribute to open source community. • For now, convert to JSON and read into Spark SQL. 15
  • 16. Columnar Formats: Parquet & ORD • Great for use with Spark SQL. • Parquet is actually best practice for Spark SQL. • Makes it easy to pull only certain records at a time. • Worth time to encode if multiple passes on data. • Do not support appending one record at a time. • Not good for collecting log records. 16
  • 18. Reuse Hadoop Libraries • HadoopRDD & NewHadoopRDD • Reuse Hadoop File format libraries. • Hive SerDe’s are supported in Spark SQL • Load your SerDe jar onto your Spark Cluster. • Issue a SHOW CREATE TABLE command. • Enter the create table command (with EXTERNAL) 18
  • 19. Controlling the Output File Size • Spark generally writes one file per partition. • Use coalesce to write out less files. • Use repartition to write out more files. • This will cause a shuffle of the data. • If repartition is too expensive: • Call mapPartition to get an iterator and write out as many (or few) files as you would like. 19
  • 20. sc.wholeTextFiles() • The whole file should fit into memory. • Good for file formats that aren’t splittable by line. • Such as XML files. • Will need to performance tune. 20
  • 21. Use “filename” as the RDD element • filenames = sc.parallelize([“s3:/file1”, “s3:/ file2”, …]) • Allows you to function on a single file. • filenames.map(…call_function_on_filename) • Use to decode non-standard compression formats. • Use to split up files that aren’t separable by line. 21