SlideShare ist ein Scribd-Unternehmen logo
1 von 26
Downloaden Sie, um offline zu lesen
Improving Spark’s
Reliability with
DataSourceV2
Ryan Blue
Spark Summit 2019
Data at Netflix
● YARN compute clusters are expendable
● Expendable clusters require architectural changes
○ GENIE is a job submission service that selects the cluster
○ METACAT is a cluster-independent metastore
○ S3 is the source of truth for data
Cloud-native data warehouse
● File list calls may be inaccurate
● Hive tables rely on accurate listing for correctness
● S3 queries may be incorrect, sometimes
S3 is eventually consistent
● File list calls may be inaccurate
● Hive tables rely on accurate listing for correctness
● S3 queries may be incorrect, sometimes
S3 is eventually consistent
At Netflix’s scale,
sometimes is every day.
● Requires consistent listing – S3MPER
● Requires in-place writes – BATCH PATTERN
● Requires atomic metastore changes – METACAT
A reliable S3 warehouse (in 2016)
Changes needed in Spark
● Integrate S3 batch pattern committers
● Spark versions
○ 1.6 – Hive path only
○ 2.0 – DataSource path for reads, not writes
○ 2.1+– Use DataSource path for reads and writes
Problems and Roadblocks
● Behavior is not defined
● What do save and saveAsTable do differently?
○ Create different logical plans . . .
that are converted to other logical plans
● When you use “overwrite” mode, what happens?
○ Depends on the data source
DataFrameWriter
● Delegates behavior to the source when tables don’t exist
● Overwrite might mean:
○ Replace table – data and metadata (Some code paths)
○ Replace all table data (Some code paths)
○ Replace static partitions (DataSource tables)
○ Replace dynamic partitions (Hive tables, SPARK-20236)
SaveMode
● What is “correct” for CTAS/overwrite when the table exists?
● PreprocessTableCreation vs PreprocessTableInsertion
○ Depends on the DataFrameWriter call
● Spark automatically inserts unsafe casts (e.g. string to int)
● Path tables have no schema validation on write
Validation
“[These] should do the same thing,
but as we've already published
these 2 interfaces and the
implementations may have different
logic, we have to keep these 2
different commands.”
“[These] should do the same thing,
but as we've already published
these 2 interfaces and the
implementations may have different
logic, we have to keep these 2
different commands.”
😕
● RunnableCommand
wraps a logical in a
pseudo-physical plan
● Commands created
inside run made it worse
Commands
● Substantial behavior changes for 2.0
○ Committed with no time to review
. . . to the 2.0 release branch
● Behavior not up for discussion
● Parts of PRs merged without attribution
Community Roadblocks
Iceberg and DataSourceV2
● Iceberg: tables without unpleasant surprises
● Fix tables, not the file system
● While fixing reliability and scale, fix usability:
○ Reliable schema evolution
○ Automatic partitioning
○ Configure tables, not jobs
A reliable S3 warehouse (in 2019)
● Need a way to plug in Iceberg cleanly
● Maintaining a separate write path takes time
● Spark’s write path had solidified
● DataSourceV2 was proposed . . .
Last year
● Isn’t v2 just an update to the read/write API?
● Existing design problems also affect v2
○ No write validation – yet another logical plan
○ SaveMode passed to sources
● Opportunity: avoid needing v3 to fix behavior
Why DataSourceV2?
● Define a set of common logical plans
○ CTAS, RTAS, Append, OverwriteByExpression, etc.
○ Document user expectations and behavior
○ Implement consistent behavior in Spark for all v2 sources
● SPIP: Standardize SQL logical plans
https://issues.apache.org/jira/browse/SPARK-23521
What’s different in DSv2
● Specialize physical plans, not logical plans
○ No more InsertIntoDataSourceTable and InsertIntoHiveTable
○ No forgetting to apply rules to a new logical plan
● Apply validation rules universally
○ Same rules for Append and Overwrite
● Avoid using RunnableCommand
Standard Logical Plans
● Create, alter, and drop tables in Spark, not sources
○ CTAS when table exists: fail the query in Spark
○ Requires a catalog plugin API
● SPIP: Spark API for Table Metadata
https://issues.apache.org/jira/browse/SPARK-27067
Consistent behavior
● Multi-catalog support
○ Create tables in the source of truth
○ Avoiding this caused strange Spark behavior
● SPIP: Identifiers for multi-catalog support
https://issues.apache.org/jira/browse/SPARK-27066
Catalog API
● Goal: working DSv2 in Spark 3.0
○ Independent of the v1 path
○ Default behavior to v1
● SPIPs have been adopted by community votes
● Append and overwrite plans are added and working
● Waiting on catalog API to add CTAS and DDL
Status
Thank you!
Questions?
Up next: Migrating to Spark at Netflix
At 11:50 today, in Room 2006

Weitere ähnliche Inhalte

Was ist angesagt?

Spark SQL Deep Dive @ Melbourne Spark Meetup
Spark SQL Deep Dive @ Melbourne Spark MeetupSpark SQL Deep Dive @ Melbourne Spark Meetup
Spark SQL Deep Dive @ Melbourne Spark MeetupDatabricks
 
A Deep Dive into Spark SQL's Catalyst Optimizer with Yin Huai
A Deep Dive into Spark SQL's Catalyst Optimizer with Yin HuaiA Deep Dive into Spark SQL's Catalyst Optimizer with Yin Huai
A Deep Dive into Spark SQL's Catalyst Optimizer with Yin HuaiDatabricks
 
Parquet performance tuning: the missing guide
Parquet performance tuning: the missing guideParquet performance tuning: the missing guide
Parquet performance tuning: the missing guideRyan Blue
 
Deep Dive into the New Features of Apache Spark 3.0
Deep Dive into the New Features of Apache Spark 3.0Deep Dive into the New Features of Apache Spark 3.0
Deep Dive into the New Features of Apache Spark 3.0Databricks
 
Top 5 Mistakes When Writing Spark Applications
Top 5 Mistakes When Writing Spark ApplicationsTop 5 Mistakes When Writing Spark Applications
Top 5 Mistakes When Writing Spark ApplicationsSpark Summit
 
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 SQLDatabricks
 
From Query Plan to Query Performance: Supercharging your Apache Spark Queries...
From Query Plan to Query Performance: Supercharging your Apache Spark Queries...From Query Plan to Query Performance: Supercharging your Apache Spark Queries...
From Query Plan to Query Performance: Supercharging your Apache Spark Queries...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 SparkDatabricks
 
Spark Shuffle Deep Dive (Explained In Depth) - How Shuffle Works in Spark
Spark Shuffle Deep Dive (Explained In Depth) - How Shuffle Works in SparkSpark Shuffle Deep Dive (Explained In Depth) - How Shuffle Works in Spark
Spark Shuffle Deep Dive (Explained In Depth) - How Shuffle Works in SparkBo Yang
 
How to Extend Apache Spark with Customized Optimizations
How to Extend Apache Spark with Customized OptimizationsHow to Extend Apache Spark with Customized Optimizations
How to Extend Apache Spark with Customized OptimizationsDatabricks
 
Hive Bucketing in Apache Spark with Tejas Patil
Hive Bucketing in Apache Spark with Tejas PatilHive Bucketing in Apache Spark with Tejas Patil
Hive Bucketing in Apache Spark with Tejas PatilDatabricks
 
Spark shuffle introduction
Spark shuffle introductionSpark shuffle introduction
Spark shuffle introductioncolorant
 
Accelerating Apache Spark Shuffle for Data Analytics on the Cloud with Remote...
Accelerating Apache Spark Shuffle for Data Analytics on the Cloud with Remote...Accelerating Apache Spark Shuffle for Data Analytics on the Cloud with Remote...
Accelerating Apache Spark Shuffle for Data Analytics on the Cloud with Remote...Databricks
 
Apache Iceberg: An Architectural Look Under the Covers
Apache Iceberg: An Architectural Look Under the CoversApache Iceberg: An Architectural Look Under the Covers
Apache Iceberg: An Architectural Look Under the CoversScyllaDB
 
Adaptive Query Execution: Speeding Up Spark SQL at Runtime
Adaptive Query Execution: Speeding Up Spark SQL at RuntimeAdaptive Query Execution: Speeding Up Spark SQL at Runtime
Adaptive Query Execution: Speeding Up Spark SQL at RuntimeDatabricks
 
Optimizing Apache Spark SQL Joins
Optimizing Apache Spark SQL JoinsOptimizing Apache Spark SQL Joins
Optimizing Apache Spark SQL JoinsDatabricks
 
Best Practice of Compression/Decompression Codes in Apache Spark with Sophia...
 Best Practice of Compression/Decompression Codes in Apache Spark with Sophia... Best Practice of Compression/Decompression Codes in Apache Spark with Sophia...
Best Practice of Compression/Decompression Codes in Apache Spark with Sophia...Databricks
 
Tame the small files problem and optimize data layout for streaming ingestion...
Tame the small files problem and optimize data layout for streaming ingestion...Tame the small files problem and optimize data layout for streaming ingestion...
Tame the small files problem and optimize data layout for streaming ingestion...Flink Forward
 
Apache Arrow Flight: A New Gold Standard for Data Transport
Apache Arrow Flight: A New Gold Standard for Data TransportApache Arrow Flight: A New Gold Standard for Data Transport
Apache Arrow Flight: A New Gold Standard for Data TransportWes McKinney
 

Was ist angesagt? (20)

Spark SQL Deep Dive @ Melbourne Spark Meetup
Spark SQL Deep Dive @ Melbourne Spark MeetupSpark SQL Deep Dive @ Melbourne Spark Meetup
Spark SQL Deep Dive @ Melbourne Spark Meetup
 
A Deep Dive into Spark SQL's Catalyst Optimizer with Yin Huai
A Deep Dive into Spark SQL's Catalyst Optimizer with Yin HuaiA Deep Dive into Spark SQL's Catalyst Optimizer with Yin Huai
A Deep Dive into Spark SQL's Catalyst Optimizer with Yin Huai
 
Parquet performance tuning: the missing guide
Parquet performance tuning: the missing guideParquet performance tuning: the missing guide
Parquet performance tuning: the missing guide
 
Deep Dive into the New Features of Apache Spark 3.0
Deep Dive into the New Features of Apache Spark 3.0Deep Dive into the New Features of Apache Spark 3.0
Deep Dive into the New Features of Apache Spark 3.0
 
Top 5 Mistakes When Writing Spark Applications
Top 5 Mistakes When Writing Spark ApplicationsTop 5 Mistakes When Writing Spark Applications
Top 5 Mistakes When Writing Spark Applications
 
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
 
From Query Plan to Query Performance: Supercharging your Apache Spark Queries...
From Query Plan to Query Performance: Supercharging your Apache Spark Queries...From Query Plan to Query Performance: Supercharging your Apache Spark Queries...
From Query Plan to Query Performance: Supercharging your Apache Spark Queries...
 
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
 
Spark Shuffle Deep Dive (Explained In Depth) - How Shuffle Works in Spark
Spark Shuffle Deep Dive (Explained In Depth) - How Shuffle Works in SparkSpark Shuffle Deep Dive (Explained In Depth) - How Shuffle Works in Spark
Spark Shuffle Deep Dive (Explained In Depth) - How Shuffle Works in Spark
 
How to Extend Apache Spark with Customized Optimizations
How to Extend Apache Spark with Customized OptimizationsHow to Extend Apache Spark with Customized Optimizations
How to Extend Apache Spark with Customized Optimizations
 
Hive Bucketing in Apache Spark with Tejas Patil
Hive Bucketing in Apache Spark with Tejas PatilHive Bucketing in Apache Spark with Tejas Patil
Hive Bucketing in Apache Spark with Tejas Patil
 
Spark shuffle introduction
Spark shuffle introductionSpark shuffle introduction
Spark shuffle introduction
 
Accelerating Apache Spark Shuffle for Data Analytics on the Cloud with Remote...
Accelerating Apache Spark Shuffle for Data Analytics on the Cloud with Remote...Accelerating Apache Spark Shuffle for Data Analytics on the Cloud with Remote...
Accelerating Apache Spark Shuffle for Data Analytics on the Cloud with Remote...
 
Spark SQL
Spark SQLSpark SQL
Spark SQL
 
Apache Iceberg: An Architectural Look Under the Covers
Apache Iceberg: An Architectural Look Under the CoversApache Iceberg: An Architectural Look Under the Covers
Apache Iceberg: An Architectural Look Under the Covers
 
Adaptive Query Execution: Speeding Up Spark SQL at Runtime
Adaptive Query Execution: Speeding Up Spark SQL at RuntimeAdaptive Query Execution: Speeding Up Spark SQL at Runtime
Adaptive Query Execution: Speeding Up Spark SQL at Runtime
 
Optimizing Apache Spark SQL Joins
Optimizing Apache Spark SQL JoinsOptimizing Apache Spark SQL Joins
Optimizing Apache Spark SQL Joins
 
Best Practice of Compression/Decompression Codes in Apache Spark with Sophia...
 Best Practice of Compression/Decompression Codes in Apache Spark with Sophia... Best Practice of Compression/Decompression Codes in Apache Spark with Sophia...
Best Practice of Compression/Decompression Codes in Apache Spark with Sophia...
 
Tame the small files problem and optimize data layout for streaming ingestion...
Tame the small files problem and optimize data layout for streaming ingestion...Tame the small files problem and optimize data layout for streaming ingestion...
Tame the small files problem and optimize data layout for streaming ingestion...
 
Apache Arrow Flight: A New Gold Standard for Data Transport
Apache Arrow Flight: A New Gold Standard for Data TransportApache Arrow Flight: A New Gold Standard for Data Transport
Apache Arrow Flight: A New Gold Standard for Data Transport
 

Ähnlich wie Improving Apache Spark's Reliability with DataSourceV2

Migrating to Spark 2.0 - Part 2
Migrating to Spark 2.0 - Part 2Migrating to Spark 2.0 - Part 2
Migrating to Spark 2.0 - Part 2datamantra
 
Deep Dive into Spark SQL with Advanced Performance Tuning with Xiao Li & Wenc...
Deep Dive into Spark SQL with Advanced Performance Tuning with Xiao Li & Wenc...Deep Dive into Spark SQL with Advanced Performance Tuning with Xiao Li & Wenc...
Deep Dive into Spark SQL with Advanced Performance Tuning with Xiao Li & Wenc...Databricks
 
NoSQL Database- cassandra column Base DB
NoSQL Database- cassandra column Base DBNoSQL Database- cassandra column Base DB
NoSQL Database- cassandra column Base DBsadegh salehi
 
Introduction to spark 2.0
Introduction to spark 2.0Introduction to spark 2.0
Introduction to spark 2.0datamantra
 
Spark Concepts - Spark SQL, Graphx, Streaming
Spark Concepts - Spark SQL, Graphx, StreamingSpark Concepts - Spark SQL, Graphx, Streaming
Spark Concepts - Spark SQL, Graphx, StreamingPetr Zapletal
 
Lambda architecture @ Indix
Lambda architecture @ IndixLambda architecture @ Indix
Lambda architecture @ IndixRajesh Muppalla
 
Dynamic DDL: Adding Structure to Streaming Data on the Fly with David Winters...
Dynamic DDL: Adding Structure to Streaming Data on the Fly with David Winters...Dynamic DDL: Adding Structure to Streaming Data on the Fly with David Winters...
Dynamic DDL: Adding Structure to Streaming Data on the Fly with David Winters...Databricks
 
Fast federated SQL with Apache Calcite
Fast federated SQL with Apache CalciteFast federated SQL with Apache Calcite
Fast federated SQL with Apache CalciteChris Baynes
 
Putting the Spark into Functional Fashion Tech Analystics
Putting the Spark into Functional Fashion Tech AnalysticsPutting the Spark into Functional Fashion Tech Analystics
Putting the Spark into Functional Fashion Tech AnalysticsGareth Rogers
 
Productionalizing a spark application
Productionalizing a spark applicationProductionalizing a spark application
Productionalizing a spark applicationdatamantra
 
Dynamic DDL: Adding structure to streaming IoT data on the fly
Dynamic DDL: Adding structure to streaming IoT data on the flyDynamic DDL: Adding structure to streaming IoT data on the fly
Dynamic DDL: Adding structure to streaming IoT data on the flyDataWorks Summit
 
The evolution of Netflix's S3 data warehouse (Strata NY 2018)
The evolution of Netflix's S3 data warehouse (Strata NY 2018)The evolution of Netflix's S3 data warehouse (Strata NY 2018)
The evolution of Netflix's S3 data warehouse (Strata NY 2018)Ryan Blue
 
CDC patterns in Apache Kafka®
CDC patterns in Apache Kafka®CDC patterns in Apache Kafka®
CDC patterns in Apache Kafka®confluent
 
Introduction to Structured Data Processing with Spark SQL
Introduction to Structured Data Processing with Spark SQLIntroduction to Structured Data Processing with Spark SQL
Introduction to Structured Data Processing with Spark SQLdatamantra
 
Putting the Spark into Functional Fashion Tech Analystics
Putting the Spark into Functional Fashion Tech AnalysticsPutting the Spark into Functional Fashion Tech Analystics
Putting the Spark into Functional Fashion Tech AnalysticsGareth Rogers
 
Migrating to spark 2.0
Migrating to spark 2.0Migrating to spark 2.0
Migrating to spark 2.0datamantra
 
Large-Scale Automated Storage on Kubernetes - Matt Schallert OSCON 2019
Large-Scale Automated Storage on Kubernetes - Matt Schallert OSCON 2019Large-Scale Automated Storage on Kubernetes - Matt Schallert OSCON 2019
Large-Scale Automated Storage on Kubernetes - Matt Schallert OSCON 2019Matt Schallert
 
Scala Days Highlights | BoldRadius
Scala Days Highlights | BoldRadiusScala Days Highlights | BoldRadius
Scala Days Highlights | BoldRadiusBoldRadius Solutions
 

Ähnlich wie Improving Apache Spark's Reliability with DataSourceV2 (20)

Migrating to Spark 2.0 - Part 2
Migrating to Spark 2.0 - Part 2Migrating to Spark 2.0 - Part 2
Migrating to Spark 2.0 - Part 2
 
Deep Dive into Spark SQL with Advanced Performance Tuning with Xiao Li & Wenc...
Deep Dive into Spark SQL with Advanced Performance Tuning with Xiao Li & Wenc...Deep Dive into Spark SQL with Advanced Performance Tuning with Xiao Li & Wenc...
Deep Dive into Spark SQL with Advanced Performance Tuning with Xiao Li & Wenc...
 
NoSQL Database- cassandra column Base DB
NoSQL Database- cassandra column Base DBNoSQL Database- cassandra column Base DB
NoSQL Database- cassandra column Base DB
 
Introduction to spark 2.0
Introduction to spark 2.0Introduction to spark 2.0
Introduction to spark 2.0
 
Spark Concepts - Spark SQL, Graphx, Streaming
Spark Concepts - Spark SQL, Graphx, StreamingSpark Concepts - Spark SQL, Graphx, Streaming
Spark Concepts - Spark SQL, Graphx, Streaming
 
Presto
PrestoPresto
Presto
 
Catalyst optimizer
Catalyst optimizerCatalyst optimizer
Catalyst optimizer
 
Lambda architecture @ Indix
Lambda architecture @ IndixLambda architecture @ Indix
Lambda architecture @ Indix
 
Dynamic DDL: Adding Structure to Streaming Data on the Fly with David Winters...
Dynamic DDL: Adding Structure to Streaming Data on the Fly with David Winters...Dynamic DDL: Adding Structure to Streaming Data on the Fly with David Winters...
Dynamic DDL: Adding Structure to Streaming Data on the Fly with David Winters...
 
Fast federated SQL with Apache Calcite
Fast federated SQL with Apache CalciteFast federated SQL with Apache Calcite
Fast federated SQL with Apache Calcite
 
Putting the Spark into Functional Fashion Tech Analystics
Putting the Spark into Functional Fashion Tech AnalysticsPutting the Spark into Functional Fashion Tech Analystics
Putting the Spark into Functional Fashion Tech Analystics
 
Productionalizing a spark application
Productionalizing a spark applicationProductionalizing a spark application
Productionalizing a spark application
 
Dynamic DDL: Adding structure to streaming IoT data on the fly
Dynamic DDL: Adding structure to streaming IoT data on the flyDynamic DDL: Adding structure to streaming IoT data on the fly
Dynamic DDL: Adding structure to streaming IoT data on the fly
 
The evolution of Netflix's S3 data warehouse (Strata NY 2018)
The evolution of Netflix's S3 data warehouse (Strata NY 2018)The evolution of Netflix's S3 data warehouse (Strata NY 2018)
The evolution of Netflix's S3 data warehouse (Strata NY 2018)
 
CDC patterns in Apache Kafka®
CDC patterns in Apache Kafka®CDC patterns in Apache Kafka®
CDC patterns in Apache Kafka®
 
Introduction to Structured Data Processing with Spark SQL
Introduction to Structured Data Processing with Spark SQLIntroduction to Structured Data Processing with Spark SQL
Introduction to Structured Data Processing with Spark SQL
 
Putting the Spark into Functional Fashion Tech Analystics
Putting the Spark into Functional Fashion Tech AnalysticsPutting the Spark into Functional Fashion Tech Analystics
Putting the Spark into Functional Fashion Tech Analystics
 
Migrating to spark 2.0
Migrating to spark 2.0Migrating to spark 2.0
Migrating to spark 2.0
 
Large-Scale Automated Storage on Kubernetes - Matt Schallert OSCON 2019
Large-Scale Automated Storage on Kubernetes - Matt Schallert OSCON 2019Large-Scale Automated Storage on Kubernetes - Matt Schallert OSCON 2019
Large-Scale Automated Storage on Kubernetes - Matt Schallert OSCON 2019
 
Scala Days Highlights | BoldRadius
Scala Days Highlights | BoldRadiusScala Days Highlights | BoldRadius
Scala Days Highlights | BoldRadius
 

Mehr von Databricks

DW Migration Webinar-March 2022.pptx
DW Migration Webinar-March 2022.pptxDW Migration Webinar-March 2022.pptx
DW Migration Webinar-March 2022.pptxDatabricks
 
Data Lakehouse Symposium | Day 1 | Part 1
Data Lakehouse Symposium | Day 1 | Part 1Data Lakehouse Symposium | Day 1 | Part 1
Data Lakehouse Symposium | Day 1 | Part 1Databricks
 
Data Lakehouse Symposium | Day 1 | Part 2
Data Lakehouse Symposium | Day 1 | Part 2Data Lakehouse Symposium | Day 1 | Part 2
Data Lakehouse Symposium | Day 1 | Part 2Databricks
 
Data Lakehouse Symposium | Day 2
Data Lakehouse Symposium | Day 2Data Lakehouse Symposium | Day 2
Data Lakehouse Symposium | Day 2Databricks
 
Data Lakehouse Symposium | Day 4
Data Lakehouse Symposium | Day 4Data Lakehouse Symposium | Day 4
Data Lakehouse Symposium | Day 4Databricks
 
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
5 Critical Steps to Clean Your Data Swamp When Migrating Off of HadoopDatabricks
 
Democratizing Data Quality Through a Centralized Platform
Democratizing Data Quality Through a Centralized PlatformDemocratizing Data Quality Through a Centralized Platform
Democratizing Data Quality Through a Centralized PlatformDatabricks
 
Learn to Use Databricks for Data Science
Learn to Use Databricks for Data ScienceLearn to Use Databricks for Data Science
Learn to Use Databricks for Data ScienceDatabricks
 
Why APM Is Not the Same As ML Monitoring
Why APM Is Not the Same As ML MonitoringWhy APM Is Not the Same As ML Monitoring
Why APM Is Not the Same As ML MonitoringDatabricks
 
The Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
The Function, the Context, and the Data—Enabling ML Ops at Stitch FixThe Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
The Function, the Context, and the Data—Enabling ML Ops at Stitch FixDatabricks
 
Stage Level Scheduling Improving Big Data and AI Integration
Stage Level Scheduling Improving Big Data and AI IntegrationStage Level Scheduling Improving Big Data and AI Integration
Stage Level Scheduling Improving Big Data and AI IntegrationDatabricks
 
Simplify Data Conversion from Spark to TensorFlow and PyTorch
Simplify Data Conversion from Spark to TensorFlow and PyTorchSimplify Data Conversion from Spark to TensorFlow and PyTorch
Simplify Data Conversion from Spark to TensorFlow and PyTorchDatabricks
 
Scaling your Data Pipelines with Apache Spark on Kubernetes
Scaling your Data Pipelines with Apache Spark on KubernetesScaling your Data Pipelines with Apache Spark on Kubernetes
Scaling your Data Pipelines with Apache Spark on KubernetesDatabricks
 
Scaling and Unifying SciKit Learn and Apache Spark Pipelines
Scaling and Unifying SciKit Learn and Apache Spark PipelinesScaling and Unifying SciKit Learn and Apache Spark Pipelines
Scaling and Unifying SciKit Learn and Apache Spark PipelinesDatabricks
 
Sawtooth Windows for Feature Aggregations
Sawtooth Windows for Feature AggregationsSawtooth Windows for Feature Aggregations
Sawtooth Windows for Feature AggregationsDatabricks
 
Redis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
Redis + Apache Spark = Swiss Army Knife Meets Kitchen SinkRedis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
Redis + Apache Spark = Swiss Army Knife Meets Kitchen SinkDatabricks
 
Re-imagine Data Monitoring with whylogs and Spark
Re-imagine Data Monitoring with whylogs and SparkRe-imagine Data Monitoring with whylogs and Spark
Re-imagine Data Monitoring with whylogs and SparkDatabricks
 
Raven: End-to-end Optimization of ML Prediction Queries
Raven: End-to-end Optimization of ML Prediction QueriesRaven: End-to-end Optimization of ML Prediction Queries
Raven: End-to-end Optimization of ML Prediction QueriesDatabricks
 
Processing Large Datasets for ADAS Applications using Apache Spark
Processing Large Datasets for ADAS Applications using Apache SparkProcessing Large Datasets for ADAS Applications using Apache Spark
Processing Large Datasets for ADAS Applications using Apache SparkDatabricks
 
Massive Data Processing in Adobe Using Delta Lake
Massive Data Processing in Adobe Using Delta LakeMassive Data Processing in Adobe Using Delta Lake
Massive Data Processing in Adobe Using Delta LakeDatabricks
 

Mehr von Databricks (20)

DW Migration Webinar-March 2022.pptx
DW Migration Webinar-March 2022.pptxDW Migration Webinar-March 2022.pptx
DW Migration Webinar-March 2022.pptx
 
Data Lakehouse Symposium | Day 1 | Part 1
Data Lakehouse Symposium | Day 1 | Part 1Data Lakehouse Symposium | Day 1 | Part 1
Data Lakehouse Symposium | Day 1 | Part 1
 
Data Lakehouse Symposium | Day 1 | Part 2
Data Lakehouse Symposium | Day 1 | Part 2Data Lakehouse Symposium | Day 1 | Part 2
Data Lakehouse Symposium | Day 1 | Part 2
 
Data Lakehouse Symposium | Day 2
Data Lakehouse Symposium | Day 2Data Lakehouse Symposium | Day 2
Data Lakehouse Symposium | Day 2
 
Data Lakehouse Symposium | Day 4
Data Lakehouse Symposium | Day 4Data Lakehouse Symposium | Day 4
Data Lakehouse Symposium | Day 4
 
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
 
Democratizing Data Quality Through a Centralized Platform
Democratizing Data Quality Through a Centralized PlatformDemocratizing Data Quality Through a Centralized Platform
Democratizing Data Quality Through a Centralized Platform
 
Learn to Use Databricks for Data Science
Learn to Use Databricks for Data ScienceLearn to Use Databricks for Data Science
Learn to Use Databricks for Data Science
 
Why APM Is Not the Same As ML Monitoring
Why APM Is Not the Same As ML MonitoringWhy APM Is Not the Same As ML Monitoring
Why APM Is Not the Same As ML Monitoring
 
The Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
The Function, the Context, and the Data—Enabling ML Ops at Stitch FixThe Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
The Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
 
Stage Level Scheduling Improving Big Data and AI Integration
Stage Level Scheduling Improving Big Data and AI IntegrationStage Level Scheduling Improving Big Data and AI Integration
Stage Level Scheduling Improving Big Data and AI Integration
 
Simplify Data Conversion from Spark to TensorFlow and PyTorch
Simplify Data Conversion from Spark to TensorFlow and PyTorchSimplify Data Conversion from Spark to TensorFlow and PyTorch
Simplify Data Conversion from Spark to TensorFlow and PyTorch
 
Scaling your Data Pipelines with Apache Spark on Kubernetes
Scaling your Data Pipelines with Apache Spark on KubernetesScaling your Data Pipelines with Apache Spark on Kubernetes
Scaling your Data Pipelines with Apache Spark on Kubernetes
 
Scaling and Unifying SciKit Learn and Apache Spark Pipelines
Scaling and Unifying SciKit Learn and Apache Spark PipelinesScaling and Unifying SciKit Learn and Apache Spark Pipelines
Scaling and Unifying SciKit Learn and Apache Spark Pipelines
 
Sawtooth Windows for Feature Aggregations
Sawtooth Windows for Feature AggregationsSawtooth Windows for Feature Aggregations
Sawtooth Windows for Feature Aggregations
 
Redis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
Redis + Apache Spark = Swiss Army Knife Meets Kitchen SinkRedis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
Redis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
 
Re-imagine Data Monitoring with whylogs and Spark
Re-imagine Data Monitoring with whylogs and SparkRe-imagine Data Monitoring with whylogs and Spark
Re-imagine Data Monitoring with whylogs and Spark
 
Raven: End-to-end Optimization of ML Prediction Queries
Raven: End-to-end Optimization of ML Prediction QueriesRaven: End-to-end Optimization of ML Prediction Queries
Raven: End-to-end Optimization of ML Prediction Queries
 
Processing Large Datasets for ADAS Applications using Apache Spark
Processing Large Datasets for ADAS Applications using Apache SparkProcessing Large Datasets for ADAS Applications using Apache Spark
Processing Large Datasets for ADAS Applications using Apache Spark
 
Massive Data Processing in Adobe Using Delta Lake
Massive Data Processing in Adobe Using Delta LakeMassive Data Processing in Adobe Using Delta Lake
Massive Data Processing in Adobe Using Delta Lake
 

Kürzlich hochgeladen

Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...amitlee9823
 
Call Girls Bommasandra Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
Call Girls Bommasandra Just Call 👗 7737669865 👗 Top Class Call Girl Service B...Call Girls Bommasandra Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
Call Girls Bommasandra Just Call 👗 7737669865 👗 Top Class Call Girl Service B...amitlee9823
 
Call Girls In Hsr Layout ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Hsr Layout ☎ 7737669865 🥵 Book Your One night StandCall Girls In Hsr Layout ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Hsr Layout ☎ 7737669865 🥵 Book Your One night Standamitlee9823
 
BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service
BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort ServiceBDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service
BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort ServiceDelhi Call girls
 
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...ZurliaSoop
 
hybrid Seed Production In Chilli & Capsicum.pptx
hybrid Seed Production In Chilli & Capsicum.pptxhybrid Seed Production In Chilli & Capsicum.pptx
hybrid Seed Production In Chilli & Capsicum.pptx9to5mart
 
5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed
5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed
5CL-ADBA,5cladba, Chinese supplier, safety is guaranteedamy56318795
 
Probability Grade 10 Third Quarter Lessons
Probability Grade 10 Third Quarter LessonsProbability Grade 10 Third Quarter Lessons
Probability Grade 10 Third Quarter LessonsJoseMangaJr1
 
Call Girls In Attibele ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Attibele ☎ 7737669865 🥵 Book Your One night StandCall Girls In Attibele ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Attibele ☎ 7737669865 🥵 Book Your One night Standamitlee9823
 
Discover Why Less is More in B2B Research
Discover Why Less is More in B2B ResearchDiscover Why Less is More in B2B Research
Discover Why Less is More in B2B Researchmichael115558
 
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...amitlee9823
 
👉 Amritsar Call Girl 👉📞 6367187148 👉📞 Just📲 Call Ruhi Call Girl Phone No Amri...
👉 Amritsar Call Girl 👉📞 6367187148 👉📞 Just📲 Call Ruhi Call Girl Phone No Amri...👉 Amritsar Call Girl 👉📞 6367187148 👉📞 Just📲 Call Ruhi Call Girl Phone No Amri...
👉 Amritsar Call Girl 👉📞 6367187148 👉📞 Just📲 Call Ruhi Call Girl Phone No Amri...karishmasinghjnh
 
Capstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics ProgramCapstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics ProgramMoniSankarHazra
 
Midocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxMidocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxolyaivanovalion
 
Vip Mumbai Call Girls Thane West Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Thane West Call On 9920725232 With Body to body massage...Vip Mumbai Call Girls Thane West Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Thane West Call On 9920725232 With Body to body massage...amitlee9823
 
➥🔝 7737669865 🔝▻ mahisagar Call-girls in Women Seeking Men 🔝mahisagar🔝 Esc...
➥🔝 7737669865 🔝▻ mahisagar Call-girls in Women Seeking Men  🔝mahisagar🔝   Esc...➥🔝 7737669865 🔝▻ mahisagar Call-girls in Women Seeking Men  🔝mahisagar🔝   Esc...
➥🔝 7737669865 🔝▻ mahisagar Call-girls in Women Seeking Men 🔝mahisagar🔝 Esc...amitlee9823
 

Kürzlich hochgeladen (20)

Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
 
(NEHA) Call Girls Katra Call Now 8617697112 Katra Escorts 24x7
(NEHA) Call Girls Katra Call Now 8617697112 Katra Escorts 24x7(NEHA) Call Girls Katra Call Now 8617697112 Katra Escorts 24x7
(NEHA) Call Girls Katra Call Now 8617697112 Katra Escorts 24x7
 
Call Girls Bommasandra Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
Call Girls Bommasandra Just Call 👗 7737669865 👗 Top Class Call Girl Service B...Call Girls Bommasandra Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
Call Girls Bommasandra Just Call 👗 7737669865 👗 Top Class Call Girl Service B...
 
Call Girls In Hsr Layout ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Hsr Layout ☎ 7737669865 🥵 Book Your One night StandCall Girls In Hsr Layout ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Hsr Layout ☎ 7737669865 🥵 Book Your One night Stand
 
BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service
BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort ServiceBDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service
BDSM⚡Call Girls in Mandawali Delhi >༒8448380779 Escort Service
 
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
 
CHEAP Call Girls in Rabindra Nagar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Rabindra Nagar  (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICECHEAP Call Girls in Rabindra Nagar  (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Rabindra Nagar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
 
hybrid Seed Production In Chilli & Capsicum.pptx
hybrid Seed Production In Chilli & Capsicum.pptxhybrid Seed Production In Chilli & Capsicum.pptx
hybrid Seed Production In Chilli & Capsicum.pptx
 
5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed
5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed
5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed
 
Probability Grade 10 Third Quarter Lessons
Probability Grade 10 Third Quarter LessonsProbability Grade 10 Third Quarter Lessons
Probability Grade 10 Third Quarter Lessons
 
Call Girls In Attibele ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Attibele ☎ 7737669865 🥵 Book Your One night StandCall Girls In Attibele ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Attibele ☎ 7737669865 🥵 Book Your One night Stand
 
Discover Why Less is More in B2B Research
Discover Why Less is More in B2B ResearchDiscover Why Less is More in B2B Research
Discover Why Less is More in B2B Research
 
Predicting Loan Approval: A Data Science Project
Predicting Loan Approval: A Data Science ProjectPredicting Loan Approval: A Data Science Project
Predicting Loan Approval: A Data Science Project
 
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
Chintamani Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore ...
 
👉 Amritsar Call Girl 👉📞 6367187148 👉📞 Just📲 Call Ruhi Call Girl Phone No Amri...
👉 Amritsar Call Girl 👉📞 6367187148 👉📞 Just📲 Call Ruhi Call Girl Phone No Amri...👉 Amritsar Call Girl 👉📞 6367187148 👉📞 Just📲 Call Ruhi Call Girl Phone No Amri...
👉 Amritsar Call Girl 👉📞 6367187148 👉📞 Just📲 Call Ruhi Call Girl Phone No Amri...
 
Capstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics ProgramCapstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics Program
 
Abortion pills in Jeddah | +966572737505 | Get Cytotec
Abortion pills in Jeddah | +966572737505 | Get CytotecAbortion pills in Jeddah | +966572737505 | Get Cytotec
Abortion pills in Jeddah | +966572737505 | Get Cytotec
 
Midocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxMidocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFx
 
Vip Mumbai Call Girls Thane West Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Thane West Call On 9920725232 With Body to body massage...Vip Mumbai Call Girls Thane West Call On 9920725232 With Body to body massage...
Vip Mumbai Call Girls Thane West Call On 9920725232 With Body to body massage...
 
➥🔝 7737669865 🔝▻ mahisagar Call-girls in Women Seeking Men 🔝mahisagar🔝 Esc...
➥🔝 7737669865 🔝▻ mahisagar Call-girls in Women Seeking Men  🔝mahisagar🔝   Esc...➥🔝 7737669865 🔝▻ mahisagar Call-girls in Women Seeking Men  🔝mahisagar🔝   Esc...
➥🔝 7737669865 🔝▻ mahisagar Call-girls in Women Seeking Men 🔝mahisagar🔝 Esc...
 

Improving Apache Spark's Reliability with DataSourceV2

  • 3. ● YARN compute clusters are expendable ● Expendable clusters require architectural changes ○ GENIE is a job submission service that selects the cluster ○ METACAT is a cluster-independent metastore ○ S3 is the source of truth for data Cloud-native data warehouse
  • 4. ● File list calls may be inaccurate ● Hive tables rely on accurate listing for correctness ● S3 queries may be incorrect, sometimes S3 is eventually consistent
  • 5. ● File list calls may be inaccurate ● Hive tables rely on accurate listing for correctness ● S3 queries may be incorrect, sometimes S3 is eventually consistent
  • 7. ● Requires consistent listing – S3MPER ● Requires in-place writes – BATCH PATTERN ● Requires atomic metastore changes – METACAT A reliable S3 warehouse (in 2016)
  • 8. Changes needed in Spark ● Integrate S3 batch pattern committers ● Spark versions ○ 1.6 – Hive path only ○ 2.0 – DataSource path for reads, not writes ○ 2.1+– Use DataSource path for reads and writes
  • 10. ● Behavior is not defined ● What do save and saveAsTable do differently? ○ Create different logical plans . . . that are converted to other logical plans ● When you use “overwrite” mode, what happens? ○ Depends on the data source DataFrameWriter
  • 11. ● Delegates behavior to the source when tables don’t exist ● Overwrite might mean: ○ Replace table – data and metadata (Some code paths) ○ Replace all table data (Some code paths) ○ Replace static partitions (DataSource tables) ○ Replace dynamic partitions (Hive tables, SPARK-20236) SaveMode
  • 12. ● What is “correct” for CTAS/overwrite when the table exists? ● PreprocessTableCreation vs PreprocessTableInsertion ○ Depends on the DataFrameWriter call ● Spark automatically inserts unsafe casts (e.g. string to int) ● Path tables have no schema validation on write Validation
  • 13. “[These] should do the same thing, but as we've already published these 2 interfaces and the implementations may have different logic, we have to keep these 2 different commands.”
  • 14. “[These] should do the same thing, but as we've already published these 2 interfaces and the implementations may have different logic, we have to keep these 2 different commands.” 😕
  • 15. ● RunnableCommand wraps a logical in a pseudo-physical plan ● Commands created inside run made it worse Commands
  • 16. ● Substantial behavior changes for 2.0 ○ Committed with no time to review . . . to the 2.0 release branch ● Behavior not up for discussion ● Parts of PRs merged without attribution Community Roadblocks
  • 18. ● Iceberg: tables without unpleasant surprises ● Fix tables, not the file system ● While fixing reliability and scale, fix usability: ○ Reliable schema evolution ○ Automatic partitioning ○ Configure tables, not jobs A reliable S3 warehouse (in 2019)
  • 19. ● Need a way to plug in Iceberg cleanly ● Maintaining a separate write path takes time ● Spark’s write path had solidified ● DataSourceV2 was proposed . . . Last year
  • 20. ● Isn’t v2 just an update to the read/write API? ● Existing design problems also affect v2 ○ No write validation – yet another logical plan ○ SaveMode passed to sources ● Opportunity: avoid needing v3 to fix behavior Why DataSourceV2?
  • 21. ● Define a set of common logical plans ○ CTAS, RTAS, Append, OverwriteByExpression, etc. ○ Document user expectations and behavior ○ Implement consistent behavior in Spark for all v2 sources ● SPIP: Standardize SQL logical plans https://issues.apache.org/jira/browse/SPARK-23521 What’s different in DSv2
  • 22. ● Specialize physical plans, not logical plans ○ No more InsertIntoDataSourceTable and InsertIntoHiveTable ○ No forgetting to apply rules to a new logical plan ● Apply validation rules universally ○ Same rules for Append and Overwrite ● Avoid using RunnableCommand Standard Logical Plans
  • 23. ● Create, alter, and drop tables in Spark, not sources ○ CTAS when table exists: fail the query in Spark ○ Requires a catalog plugin API ● SPIP: Spark API for Table Metadata https://issues.apache.org/jira/browse/SPARK-27067 Consistent behavior
  • 24. ● Multi-catalog support ○ Create tables in the source of truth ○ Avoiding this caused strange Spark behavior ● SPIP: Identifiers for multi-catalog support https://issues.apache.org/jira/browse/SPARK-27066 Catalog API
  • 25. ● Goal: working DSv2 in Spark 3.0 ○ Independent of the v1 path ○ Default behavior to v1 ● SPIPs have been adopted by community votes ● Append and overwrite plans are added and working ● Waiting on catalog API to add CTAS and DDL Status
  • 26. Thank you! Questions? Up next: Migrating to Spark at Netflix At 11:50 today, in Room 2006