SlideShare une entreprise Scribd logo
1  sur  42
Architecture of Flink's
Streaming Runtime
Robert Metzger
@rmetzger_
rmetzger@apache.org
What is stream processing
 Real-world data is unbounded and is
pushed to systems
 Right now: people are using the batch
paradigm for stream analysis (there was
no good stream processor available)
 New systems (Flink, Kafka) embrace
streaming nature of data
2
Web server Kafka topic
Stream processing
3
Flink is a stream processor with many faces
Streaming dataflow runtime
Flink's streaming runtime
4
Requirements for a stream processor
 Low latency
• Fast results (milliseconds)
 High throughput
• handle large data amounts (millions of events
per second)
 Exactly-once guarantees
• Correct results, also in failure cases
 Programmability
• Intuitive APIs
5
Pipelining
6
Basic building block to “keep the data moving”
• Low latency
• Operators push
data forward
• Data shipping as
buffers, not tuple-
wise
• Natural handling
of back-pressure
Fault Tolerance in streaming
 at least once: ensure all operators see all
events
• Storm: Replay stream in failure case
 Exactly once: Ensure that operators do
not perform duplicate updates to their
state
• Flink: Distributed Snapshots
• Spark: Micro-batches on batch runtime
7
Flink’s Distributed Snapshots
 Lightweight approach of storing the state
of all operators without pausing the
execution
 high throughput, low latency
 Implemented using barriers flowing
through the topology
8
Kafka
Consumer
offset = 162
Element
Counter
value = 152
Operator
stateData Stream
barrier
Before barrier =
part of the snapshot
After barrier =
Not in snapshot
(backup till next snapshot)
9
10
11
12
Best of all worlds for streaming
 Low latency
• Thanks to pipelined engine
 Exactly-once guarantees
• Distributed Snapshots
 High throughput
• Controllable checkpointing overhead
13
Throughput of distributed grep
14
Data
Generator
“grep”
operator
30 machines, 120 cores
0
20,000,000
40,000,000
60,000,000
80,000,000
100,000,000
120,000,000
140,000,000
160,000,000
180,000,000
200,000,000
Flink, no fault
tolerance
Flink, exactly
once (5s)
Storm, no
fault tolerance
Storm, micro-
batches
aggregate throughput
of 175 million
elements per second
aggregate throughput
of 9 million elements
per second
• Flink achieves 20x
higher throughput
• Flink throughput
almost the same
with and without
exactly-once
Aggregate throughput for stream record
grouping
15
0
10,000,000
20,000,000
30,000,000
40,000,000
50,000,000
60,000,000
70,000,000
80,000,000
90,000,000
100,000,000
Flink, no
fault
tolerance
Flink,
exactly
once
Storm, no
fault
tolerance
Storm, at
least once
aggregate throughput
of 83 million elements
per second
8,6 million elements/s
309k elements/s  Flink achieves 260x
higher throughput with
fault tolerance
30 machines,
120 cores Network
transfer
Latency in stream record grouping
16
Data
Generator
Receiver:
Throughput /
Latency measure
• Measure time for a record to
travel from source to sink
0.00
5.00
10.00
15.00
20.00
25.00
30.00
Flink, no
fault
tolerance
Flink, exactly
once
Storm, at
least once
Median latency
25 ms
1 ms
0.00
10.00
20.00
30.00
40.00
50.00
60.00
Flink, no
fault
tolerance
Flink,
exactly
once
Storm, at
least
once
99th percentile
latency
50 ms
17
Exactly-Once with YARN Chaos Monkey
 Validate exactly-once guarantees with
state-machine
18
“Faces” of Flink
19
Faces of a stream processor
20
Stream
processing
Batch
processing
Machine Learning at scale
Graph Analysis
Streaming dataflow runtime
The Flink Stack
21
Streaming dataflow runtime
Specialized
Abstractions
/ APIs
Core APIs
Flink Core
Runtime
Deployment
APIs for stream and batch
22
case class Word (word: String, frequency: Int)
val lines: DataStream[String] = env.fromSocketStream(...)
lines.flatMap {line => line.split(" ")
.map(word => Word(word,1))}
.window(Time.of(5,SECONDS)).every(Time.of(1,SECONDS))
.groupBy("word").sum("frequency")
.print()
val lines: DataSet[String] = env.readTextFile(...)
lines.flatMap {line => line.split(" ")
.map(word => Word(word,1))}
.groupBy("word").sum("frequency")
.print()
DataSet API (batch):
DataStream API (streaming):
The Flink Stack
23
Streaming dataflow runtime
DataSet (Java/Scala) DataStream (Java/Scala)
Experimental
Python API also
available
Data Source
orders.tbl
Filter
Map DataSource
lineitem.tbl
Join
Hybrid Hash
buildHT probe
hash-part [0] hash-part [0]
GroupRed
sort
forward
API independent Dataflow
Graph representation
Batch Optimizer Graph Builder
Batch is a special case of streaming
 Batch: run a bounded stream (data set) on
a stream processor
 Form a global window over the entire data
set for join or grouping operations
24
Batch-specific optimizations
 Managed memory on- and off-heap
• Operators (join, sort, …) with out-of-core
support
• Optimized serialization stack for user-types
 Cost-based Optimizer
• Job execution depends on data size
25
The Flink Stack
26
Streaming dataflow runtime
Specialized
Abstractions
/ APIs
Core APIs
Flink Core
Runtime
Deployment
DataSet (Java/Scala) DataStream
FlinkML: Machine Learning
 API for ML pipelines inspired by scikit-learn
 Collection of packaged algorithms
• SVM, Multiple Linear Regression, Optimization, ALS, ...
27
val trainingData: DataSet[LabeledVector] = ...
val testingData: DataSet[Vector] = ...
val scaler = StandardScaler()
val polyFeatures = PolynomialFeatures().setDegree(3)
val mlr = MultipleLinearRegression()
val pipeline = scaler.chainTransformer(polyFeatures).chainPredictor(mlr)
pipeline.fit(trainingData)
val predictions: DataSet[LabeledVector] = pipeline.predict(testingData)
Gelly: Graph Processing
 Graph API and library
 Packaged algorithms
• PageRank, SSSP, Label Propagation, Community
Detection, Connected Components
28
ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
Graph<Long, Long, NullValue> graph = ...
DataSet<Vertex<Long, Long>> verticesWithCommunity = graph.run(
new LabelPropagation<Long>(30)).getVertices();
verticesWithCommunity.print();
env.execute();
Flink Stack += Gelly, ML
29
Gelly
ML
DataSet (Java/Scala) DataStream
Streaming dataflow runtime
Integration with other systems
30
SAMOA
DataSet DataStream
HadoopM/R
GoogleDataflow
Cascading
Storm
Zeppelin
• Use Hadoop Input/Output Formats
• Mapper / Reducer implementations
• Hadoop’s FileSystem implementations
• Run applications implemented against Google’s Data Flow API
on premise with Flink
• Run Cascading jobs on Flink, with almost no code change
• Benefit from Flink’s vastly better performance than
MapReduce
• Interactive, web-based data exploration
• Machine learning on data streams
• Compatibility layer for running Storm code
• FlinkTopologyBuilder: one line replacement for
existing jobs
• Wrappers for Storm Spouts and Bolts
• Coming soon: Exactly-once with Storm
Deployment options
Gelly
Table
ML
SAMOA
DataSet(Java/Scala)DataStream
Hadoop
LocalClusterYARNTezEmbedded
Dataflow
Dataflow
MRQL
Table
Cascading
Streamingdataflowruntime
Storm
Zeppelin
• Start Flink in your IDE / on your machine
• Local debugging / development using the
same code as on the cluster
• “bare metal” standalone installation of Flink
on a cluster
• Flink on Hadoop YARN (Hadoop 2.2.0+)
• Restarts failed containers
• Support for Kerberos-secured YARN/HDFS
setups
The full stack
32
Gelly
Table
ML
SAMOA
DataSet (Java/Scala) DataStream
HadoopM/R
Local Cluster Yarn Tez Embedded
Dataflow
Dataflow(WiP)
MRQL
Table
Cascading
Streaming dataflow runtime
Storm(WiP)
Zeppelin
Closing
33
tl;dr Summary
Flink is a software stack of
 Streaming runtime
• low latency
• high throughput
• fault tolerant, exactly-once data processing
 Rich APIs for batch and stream processing
• library ecosystem
• integration with many systems
 A great community of devs and users
 Used in production
34
What is currently happening?
 Features in progress:
• Master High Availability
• Vastly improved monitoring GUI
• Watermarks / Event time processing /
Windowing rework
• Graduate Streaming API out of Beta
 0.10.0-milestone-1 is currently voted
35
How do I get started?
36
Mailing Lists: (news | user | dev)@flink.apache.org
Twitter: @ApacheFlink
Blogs: flink.apache.org/blog, data-artisans.com/blog/
IRC channel: irc.freenode.net#flink
Start Flink on YARN in 4 commands:
# get the hadoop2 package from the Flink download page at
# http://flink.apache.org/downloads.html
wget <download url>
tar xvzf flink-0.9.1-bin-hadoop2.tgz
cd flink-0.9.1/
./bin/flink run -m yarn-cluster -yn 4 ./examples/flink-java-
examples-0.9.1-WordCount.jar
flink.apache.org 37
Flink Forward: 2 days conference with
free training in Berlin, Germany
• Schedule: http://flink-forward.org/?post_type=day
Appendix
38
Managed (off-heap) memory and out-of-
core support
39Memory runs out
Cost-based Optimizer
40
DataSource
orders.tbl
Filter
Map DataSource
lineitem.tbl
Join
Hybrid Hash
buildHT probe
broadcast forward
Combine
GroupRed
sort
DataSource
orders.tbl
Filter
Map DataSource
lineitem.tbl
Join
Hybrid Hash
buildHT probe
hash-part [0] hash-part [0]
hash-part [0,1]
GroupRed
sort
forward
Best plan
depends on
relative sizes
of input files
Iterative processing in Flink
Flink offers built-in iterations and delta
iterations to execute ML and graph
algorithms efficiently
41
map
join sum
ID1
ID2
ID3
Example: Matrix Factorization
42
Factorizing a matrix with
28 billion ratings for
recommendations
More at: http://data-artisans.com/computing-recommendations-with-flink.html

Contenu connexe

Tendances

Flink Streaming Hadoop Summit San Jose
Flink Streaming Hadoop Summit San JoseFlink Streaming Hadoop Summit San Jose
Flink Streaming Hadoop Summit San Jose
Kostas Tzoumas
 
Apache Flink@ Strata & Hadoop World London
Apache Flink@ Strata & Hadoop World LondonApache Flink@ Strata & Hadoop World London
Apache Flink@ Strata & Hadoop World London
Stephan Ewen
 
Till Rohrmann – Fault Tolerance and Job Recovery in Apache Flink
Till Rohrmann – Fault Tolerance and Job Recovery in Apache FlinkTill Rohrmann – Fault Tolerance and Job Recovery in Apache Flink
Till Rohrmann – Fault Tolerance and Job Recovery in Apache Flink
Flink Forward
 
Tran Nam-Luc – Stale Synchronous Parallel Iterations on Flink
Tran Nam-Luc – Stale Synchronous Parallel Iterations on FlinkTran Nam-Luc – Stale Synchronous Parallel Iterations on Flink
Tran Nam-Luc – Stale Synchronous Parallel Iterations on Flink
Flink Forward
 

Tendances (20)

Marton Balassi – Stateful Stream Processing
Marton Balassi – Stateful Stream ProcessingMarton Balassi – Stateful Stream Processing
Marton Balassi – Stateful Stream Processing
 
Flink Streaming Hadoop Summit San Jose
Flink Streaming Hadoop Summit San JoseFlink Streaming Hadoop Summit San Jose
Flink Streaming Hadoop Summit San Jose
 
Apache Flink@ Strata & Hadoop World London
Apache Flink@ Strata & Hadoop World LondonApache Flink@ Strata & Hadoop World London
Apache Flink@ Strata & Hadoop World London
 
Streaming Analytics & CEP - Two sides of the same coin?
Streaming Analytics & CEP - Two sides of the same coin?Streaming Analytics & CEP - Two sides of the same coin?
Streaming Analytics & CEP - Two sides of the same coin?
 
Till Rohrmann – Fault Tolerance and Job Recovery in Apache Flink
Till Rohrmann – Fault Tolerance and Job Recovery in Apache FlinkTill Rohrmann – Fault Tolerance and Job Recovery in Apache Flink
Till Rohrmann – Fault Tolerance and Job Recovery in Apache Flink
 
Flink internals web
Flink internals web Flink internals web
Flink internals web
 
Tran Nam-Luc – Stale Synchronous Parallel Iterations on Flink
Tran Nam-Luc – Stale Synchronous Parallel Iterations on FlinkTran Nam-Luc – Stale Synchronous Parallel Iterations on Flink
Tran Nam-Luc – Stale Synchronous Parallel Iterations on Flink
 
Streaming in the Wild with Apache Flink
Streaming in the Wild with Apache FlinkStreaming in the Wild with Apache Flink
Streaming in the Wild with Apache Flink
 
A look at Flink 1.2
A look at Flink 1.2A look at Flink 1.2
A look at Flink 1.2
 
Apache Flink: API, runtime, and project roadmap
Apache Flink: API, runtime, and project roadmapApache Flink: API, runtime, and project roadmap
Apache Flink: API, runtime, and project roadmap
 
Apache Flink at Strata San Jose 2016
Apache Flink at Strata San Jose 2016Apache Flink at Strata San Jose 2016
Apache Flink at Strata San Jose 2016
 
Continuous Processing with Apache Flink - Strata London 2016
Continuous Processing with Apache Flink - Strata London 2016Continuous Processing with Apache Flink - Strata London 2016
Continuous Processing with Apache Flink - Strata London 2016
 
Apache Flink Internals: Stream & Batch Processing in One System – Apache Flin...
Apache Flink Internals: Stream & Batch Processing in One System – Apache Flin...Apache Flink Internals: Stream & Batch Processing in One System – Apache Flin...
Apache Flink Internals: Stream & Batch Processing in One System – Apache Flin...
 
Taking a look under the hood of Apache Flink's relational APIs.
Taking a look under the hood of Apache Flink's relational APIs.Taking a look under the hood of Apache Flink's relational APIs.
Taking a look under the hood of Apache Flink's relational APIs.
 
Don't Cross The Streams - Data Streaming And Apache Flink
Don't Cross The Streams  - Data Streaming And Apache FlinkDon't Cross The Streams  - Data Streaming And Apache Flink
Don't Cross The Streams - Data Streaming And Apache Flink
 
Big Data Warsaw
Big Data WarsawBig Data Warsaw
Big Data Warsaw
 
Aljoscha Krettek - The Future of Apache Flink
Aljoscha Krettek - The Future of Apache FlinkAljoscha Krettek - The Future of Apache Flink
Aljoscha Krettek - The Future of Apache Flink
 
Unified Stream and Batch Processing with Apache Flink
Unified Stream and Batch Processing with Apache FlinkUnified Stream and Batch Processing with Apache Flink
Unified Stream and Batch Processing with Apache Flink
 
Flink Forward Berlin 2017: Fabian Hueske - Using Stream and Batch Processing ...
Flink Forward Berlin 2017: Fabian Hueske - Using Stream and Batch Processing ...Flink Forward Berlin 2017: Fabian Hueske - Using Stream and Batch Processing ...
Flink Forward Berlin 2017: Fabian Hueske - Using Stream and Batch Processing ...
 
Machine Learning with Apache Flink at Stockholm Machine Learning Group
Machine Learning with Apache Flink at Stockholm Machine Learning GroupMachine Learning with Apache Flink at Stockholm Machine Learning Group
Machine Learning with Apache Flink at Stockholm Machine Learning Group
 

En vedette

En vedette (7)

Apache Kylin @ Big Data Europe 2015
Apache Kylin @ Big Data Europe 2015Apache Kylin @ Big Data Europe 2015
Apache Kylin @ Big Data Europe 2015
 
Apache Big Data EU 2015 - Phoenix
Apache Big Data EU 2015 - PhoenixApache Big Data EU 2015 - Phoenix
Apache Big Data EU 2015 - Phoenix
 
Being Ready for Apache Kafka - Apache: Big Data Europe 2015
Being Ready for Apache Kafka - Apache: Big Data Europe 2015Being Ready for Apache Kafka - Apache: Big Data Europe 2015
Being Ready for Apache Kafka - Apache: Big Data Europe 2015
 
Fraud Detection in Real-time @ Apache Big Data Con
Fraud Detection in Real-time @ Apache Big Data ConFraud Detection in Real-time @ Apache Big Data Con
Fraud Detection in Real-time @ Apache Big Data Con
 
Geospatial querying in Apache Marmotta - ApacheCon Big Data Europe 2015
Geospatial querying in Apache Marmotta - ApacheCon Big Data Europe 2015Geospatial querying in Apache Marmotta - ApacheCon Big Data Europe 2015
Geospatial querying in Apache Marmotta - ApacheCon Big Data Europe 2015
 
IBM Message Hub service in Bluemix - Apache Kafka in a public cloud
IBM Message Hub service in Bluemix - Apache Kafka in a public cloudIBM Message Hub service in Bluemix - Apache Kafka in a public cloud
IBM Message Hub service in Bluemix - Apache Kafka in a public cloud
 
Apache-Flink-What-How-Why-Who-Where-by-Slim-Baltagi
Apache-Flink-What-How-Why-Who-Where-by-Slim-BaltagiApache-Flink-What-How-Why-Who-Where-by-Slim-Baltagi
Apache-Flink-What-How-Why-Who-Where-by-Slim-Baltagi
 

Similaire à Architecture of Flink's Streaming Runtime @ ApacheCon EU 2015

Similaire à Architecture of Flink's Streaming Runtime @ ApacheCon EU 2015 (20)

Chicago Flink Meetup: Flink's streaming architecture
Chicago Flink Meetup: Flink's streaming architectureChicago Flink Meetup: Flink's streaming architecture
Chicago Flink Meetup: Flink's streaming architecture
 
Flexible and Real-Time Stream Processing with Apache Flink
Flexible and Real-Time Stream Processing with Apache FlinkFlexible and Real-Time Stream Processing with Apache Flink
Flexible and Real-Time Stream Processing with Apache Flink
 
Strata Singapore: Gearpump Real time DAG-Processing with Akka at Scale
Strata Singapore: GearpumpReal time DAG-Processing with Akka at ScaleStrata Singapore: GearpumpReal time DAG-Processing with Akka at Scale
Strata Singapore: Gearpump Real time DAG-Processing with Akka at Scale
 
Flink Streaming Berlin Meetup
Flink Streaming Berlin MeetupFlink Streaming Berlin Meetup
Flink Streaming Berlin Meetup
 
Apache Flink Meetup Munich (November 2015): Flink Overview, Architecture, Int...
Apache Flink Meetup Munich (November 2015): Flink Overview, Architecture, Int...Apache Flink Meetup Munich (November 2015): Flink Overview, Architecture, Int...
Apache Flink Meetup Munich (November 2015): Flink Overview, Architecture, Int...
 
Tale of two streaming frameworks- Apace Storm & Apache Flink
Tale of two streaming frameworks- Apace Storm & Apache FlinkTale of two streaming frameworks- Apace Storm & Apache Flink
Tale of two streaming frameworks- Apace Storm & Apache Flink
 
Tale of two streaming frameworks (Karthik D - Walmart)
Tale of two streaming frameworks (Karthik D - Walmart)Tale of two streaming frameworks (Karthik D - Walmart)
Tale of two streaming frameworks (Karthik D - Walmart)
 
K. Tzoumas & S. Ewen – Flink Forward Keynote
K. Tzoumas & S. Ewen – Flink Forward KeynoteK. Tzoumas & S. Ewen – Flink Forward Keynote
K. Tzoumas & S. Ewen – Flink Forward Keynote
 
Apache Flink(tm) - A Next-Generation Stream Processor
Apache Flink(tm) - A Next-Generation Stream ProcessorApache Flink(tm) - A Next-Generation Stream Processor
Apache Flink(tm) - A Next-Generation Stream Processor
 
QCon London - Stream Processing with Apache Flink
QCon London - Stream Processing with Apache FlinkQCon London - Stream Processing with Apache Flink
QCon London - Stream Processing with Apache Flink
 
GOTO Night Amsterdam - Stream processing with Apache Flink
GOTO Night Amsterdam - Stream processing with Apache FlinkGOTO Night Amsterdam - Stream processing with Apache Flink
GOTO Night Amsterdam - Stream processing with Apache Flink
 
Apache Flink Overview at SF Spark and Friends
Apache Flink Overview at SF Spark and FriendsApache Flink Overview at SF Spark and Friends
Apache Flink Overview at SF Spark and Friends
 
Stephan Ewen - Experiences running Flink at Very Large Scale
Stephan Ewen -  Experiences running Flink at Very Large ScaleStephan Ewen -  Experiences running Flink at Very Large Scale
Stephan Ewen - Experiences running Flink at Very Large Scale
 
Counting Elements in Streams
Counting Elements in StreamsCounting Elements in Streams
Counting Elements in Streams
 
Flink forward-2017-netflix keystones-paas
Flink forward-2017-netflix keystones-paasFlink forward-2017-netflix keystones-paas
Flink forward-2017-netflix keystones-paas
 
Devoxx university - Kafka de haut en bas
Devoxx university - Kafka de haut en basDevoxx university - Kafka de haut en bas
Devoxx university - Kafka de haut en bas
 
Ceilosca
CeiloscaCeilosca
Ceilosca
 
Ingestion and Dimensions Compute and Enrich using Apache Apex
Ingestion and Dimensions Compute and Enrich using Apache ApexIngestion and Dimensions Compute and Enrich using Apache Apex
Ingestion and Dimensions Compute and Enrich using Apache Apex
 
Debunking Common Myths in Stream Processing
Debunking Common Myths in Stream ProcessingDebunking Common Myths in Stream Processing
Debunking Common Myths in Stream Processing
 
Intro to Apache Apex - Next Gen Platform for Ingest and Transform
Intro to Apache Apex - Next Gen Platform for Ingest and TransformIntro to Apache Apex - Next Gen Platform for Ingest and Transform
Intro to Apache Apex - Next Gen Platform for Ingest and Transform
 

Plus de Robert Metzger

Flink Cummunity Update July (Berlin Meetup)
Flink Cummunity Update July (Berlin Meetup)Flink Cummunity Update July (Berlin Meetup)
Flink Cummunity Update July (Berlin Meetup)
Robert Metzger
 
Flink Community Update February 2015
Flink Community Update February 2015Flink Community Update February 2015
Flink Community Update February 2015
Robert Metzger
 

Plus de Robert Metzger (20)

How to Contribute to Apache Flink (and Flink at the Apache Software Foundation)
How to Contribute to Apache Flink (and Flink at the Apache Software Foundation)How to Contribute to Apache Flink (and Flink at the Apache Software Foundation)
How to Contribute to Apache Flink (and Flink at the Apache Software Foundation)
 
dA Platform Overview
dA Platform OverviewdA Platform Overview
dA Platform Overview
 
Apache Flink @ Tel Aviv / Herzliya Meetup
Apache Flink @ Tel Aviv / Herzliya MeetupApache Flink @ Tel Aviv / Herzliya Meetup
Apache Flink @ Tel Aviv / Herzliya Meetup
 
Apache Flink Community Updates November 2016 @ Berlin Meetup
Apache Flink Community Updates November 2016 @ Berlin MeetupApache Flink Community Updates November 2016 @ Berlin Meetup
Apache Flink Community Updates November 2016 @ Berlin Meetup
 
A Data Streaming Architecture with Apache Flink (berlin Buzzwords 2016)
A Data Streaming Architecture with Apache Flink (berlin Buzzwords 2016)A Data Streaming Architecture with Apache Flink (berlin Buzzwords 2016)
A Data Streaming Architecture with Apache Flink (berlin Buzzwords 2016)
 
Community Update May 2016 (January - May) | Berlin Apache Flink Meetup
Community Update May 2016 (January - May) | Berlin Apache Flink MeetupCommunity Update May 2016 (January - May) | Berlin Apache Flink Meetup
Community Update May 2016 (January - May) | Berlin Apache Flink Meetup
 
January 2016 Flink Community Update & Roadmap 2016
January 2016 Flink Community Update & Roadmap 2016January 2016 Flink Community Update & Roadmap 2016
January 2016 Flink Community Update & Roadmap 2016
 
Flink Community Update December 2015: Year in Review
Flink Community Update December 2015: Year in ReviewFlink Community Update December 2015: Year in Review
Flink Community Update December 2015: Year in Review
 
Flink September 2015 Community Update
Flink September 2015 Community UpdateFlink September 2015 Community Update
Flink September 2015 Community Update
 
Click-Through Example for Flink’s KafkaConsumer Checkpointing
Click-Through Example for Flink’s KafkaConsumer CheckpointingClick-Through Example for Flink’s KafkaConsumer Checkpointing
Click-Through Example for Flink’s KafkaConsumer Checkpointing
 
August Flink Community Update
August Flink Community UpdateAugust Flink Community Update
August Flink Community Update
 
Flink Cummunity Update July (Berlin Meetup)
Flink Cummunity Update July (Berlin Meetup)Flink Cummunity Update July (Berlin Meetup)
Flink Cummunity Update July (Berlin Meetup)
 
Apache Flink First Half of 2015 Community Update
Apache Flink First Half of 2015 Community UpdateApache Flink First Half of 2015 Community Update
Apache Flink First Half of 2015 Community Update
 
Apache Flink Deep-Dive @ Hadoop Summit 2015 in San Jose, CA
Apache Flink Deep-Dive @ Hadoop Summit 2015 in San Jose, CAApache Flink Deep-Dive @ Hadoop Summit 2015 in San Jose, CA
Apache Flink Deep-Dive @ Hadoop Summit 2015 in San Jose, CA
 
Apache Flink Hands On
Apache Flink Hands OnApache Flink Hands On
Apache Flink Hands On
 
Berlin Apache Flink Meetup May 2015, Community Update
Berlin Apache Flink Meetup May 2015, Community UpdateBerlin Apache Flink Meetup May 2015, Community Update
Berlin Apache Flink Meetup May 2015, Community Update
 
Unified batch and stream processing with Flink @ Big Data Beers Berlin May 2015
Unified batch and stream processing with Flink @ Big Data Beers Berlin May 2015Unified batch and stream processing with Flink @ Big Data Beers Berlin May 2015
Unified batch and stream processing with Flink @ Big Data Beers Berlin May 2015
 
Flink Community Update April 2015
Flink Community Update April 2015Flink Community Update April 2015
Flink Community Update April 2015
 
Apache Flink Community Update March 2015
Apache Flink Community Update March 2015Apache Flink Community Update March 2015
Apache Flink Community Update March 2015
 
Flink Community Update February 2015
Flink Community Update February 2015Flink Community Update February 2015
Flink Community Update February 2015
 

Dernier

+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
?#DUbAI#??##{{(☎️+971_581248768%)**%*]'#abortion pills for sale in dubai@
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
WSO2
 

Dernier (20)

DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
 
Six Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal OntologySix Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal Ontology
 
WSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering DevelopersWSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering Developers
 
Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
 
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot ModelMcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptx
 
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 

Architecture of Flink's Streaming Runtime @ ApacheCon EU 2015

  • 1. Architecture of Flink's Streaming Runtime Robert Metzger @rmetzger_ rmetzger@apache.org
  • 2. What is stream processing  Real-world data is unbounded and is pushed to systems  Right now: people are using the batch paradigm for stream analysis (there was no good stream processor available)  New systems (Flink, Kafka) embrace streaming nature of data 2 Web server Kafka topic Stream processing
  • 3. 3 Flink is a stream processor with many faces Streaming dataflow runtime
  • 5. Requirements for a stream processor  Low latency • Fast results (milliseconds)  High throughput • handle large data amounts (millions of events per second)  Exactly-once guarantees • Correct results, also in failure cases  Programmability • Intuitive APIs 5
  • 6. Pipelining 6 Basic building block to “keep the data moving” • Low latency • Operators push data forward • Data shipping as buffers, not tuple- wise • Natural handling of back-pressure
  • 7. Fault Tolerance in streaming  at least once: ensure all operators see all events • Storm: Replay stream in failure case  Exactly once: Ensure that operators do not perform duplicate updates to their state • Flink: Distributed Snapshots • Spark: Micro-batches on batch runtime 7
  • 8. Flink’s Distributed Snapshots  Lightweight approach of storing the state of all operators without pausing the execution  high throughput, low latency  Implemented using barriers flowing through the topology 8 Kafka Consumer offset = 162 Element Counter value = 152 Operator stateData Stream barrier Before barrier = part of the snapshot After barrier = Not in snapshot (backup till next snapshot)
  • 9. 9
  • 10. 10
  • 11. 11
  • 12. 12
  • 13. Best of all worlds for streaming  Low latency • Thanks to pipelined engine  Exactly-once guarantees • Distributed Snapshots  High throughput • Controllable checkpointing overhead 13
  • 14. Throughput of distributed grep 14 Data Generator “grep” operator 30 machines, 120 cores 0 20,000,000 40,000,000 60,000,000 80,000,000 100,000,000 120,000,000 140,000,000 160,000,000 180,000,000 200,000,000 Flink, no fault tolerance Flink, exactly once (5s) Storm, no fault tolerance Storm, micro- batches aggregate throughput of 175 million elements per second aggregate throughput of 9 million elements per second • Flink achieves 20x higher throughput • Flink throughput almost the same with and without exactly-once
  • 15. Aggregate throughput for stream record grouping 15 0 10,000,000 20,000,000 30,000,000 40,000,000 50,000,000 60,000,000 70,000,000 80,000,000 90,000,000 100,000,000 Flink, no fault tolerance Flink, exactly once Storm, no fault tolerance Storm, at least once aggregate throughput of 83 million elements per second 8,6 million elements/s 309k elements/s  Flink achieves 260x higher throughput with fault tolerance 30 machines, 120 cores Network transfer
  • 16. Latency in stream record grouping 16 Data Generator Receiver: Throughput / Latency measure • Measure time for a record to travel from source to sink 0.00 5.00 10.00 15.00 20.00 25.00 30.00 Flink, no fault tolerance Flink, exactly once Storm, at least once Median latency 25 ms 1 ms 0.00 10.00 20.00 30.00 40.00 50.00 60.00 Flink, no fault tolerance Flink, exactly once Storm, at least once 99th percentile latency 50 ms
  • 17. 17
  • 18. Exactly-Once with YARN Chaos Monkey  Validate exactly-once guarantees with state-machine 18
  • 20. Faces of a stream processor 20 Stream processing Batch processing Machine Learning at scale Graph Analysis Streaming dataflow runtime
  • 21. The Flink Stack 21 Streaming dataflow runtime Specialized Abstractions / APIs Core APIs Flink Core Runtime Deployment
  • 22. APIs for stream and batch 22 case class Word (word: String, frequency: Int) val lines: DataStream[String] = env.fromSocketStream(...) lines.flatMap {line => line.split(" ") .map(word => Word(word,1))} .window(Time.of(5,SECONDS)).every(Time.of(1,SECONDS)) .groupBy("word").sum("frequency") .print() val lines: DataSet[String] = env.readTextFile(...) lines.flatMap {line => line.split(" ") .map(word => Word(word,1))} .groupBy("word").sum("frequency") .print() DataSet API (batch): DataStream API (streaming):
  • 23. The Flink Stack 23 Streaming dataflow runtime DataSet (Java/Scala) DataStream (Java/Scala) Experimental Python API also available Data Source orders.tbl Filter Map DataSource lineitem.tbl Join Hybrid Hash buildHT probe hash-part [0] hash-part [0] GroupRed sort forward API independent Dataflow Graph representation Batch Optimizer Graph Builder
  • 24. Batch is a special case of streaming  Batch: run a bounded stream (data set) on a stream processor  Form a global window over the entire data set for join or grouping operations 24
  • 25. Batch-specific optimizations  Managed memory on- and off-heap • Operators (join, sort, …) with out-of-core support • Optimized serialization stack for user-types  Cost-based Optimizer • Job execution depends on data size 25
  • 26. The Flink Stack 26 Streaming dataflow runtime Specialized Abstractions / APIs Core APIs Flink Core Runtime Deployment DataSet (Java/Scala) DataStream
  • 27. FlinkML: Machine Learning  API for ML pipelines inspired by scikit-learn  Collection of packaged algorithms • SVM, Multiple Linear Regression, Optimization, ALS, ... 27 val trainingData: DataSet[LabeledVector] = ... val testingData: DataSet[Vector] = ... val scaler = StandardScaler() val polyFeatures = PolynomialFeatures().setDegree(3) val mlr = MultipleLinearRegression() val pipeline = scaler.chainTransformer(polyFeatures).chainPredictor(mlr) pipeline.fit(trainingData) val predictions: DataSet[LabeledVector] = pipeline.predict(testingData)
  • 28. Gelly: Graph Processing  Graph API and library  Packaged algorithms • PageRank, SSSP, Label Propagation, Community Detection, Connected Components 28 ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment(); Graph<Long, Long, NullValue> graph = ... DataSet<Vertex<Long, Long>> verticesWithCommunity = graph.run( new LabelPropagation<Long>(30)).getVertices(); verticesWithCommunity.print(); env.execute();
  • 29. Flink Stack += Gelly, ML 29 Gelly ML DataSet (Java/Scala) DataStream Streaming dataflow runtime
  • 30. Integration with other systems 30 SAMOA DataSet DataStream HadoopM/R GoogleDataflow Cascading Storm Zeppelin • Use Hadoop Input/Output Formats • Mapper / Reducer implementations • Hadoop’s FileSystem implementations • Run applications implemented against Google’s Data Flow API on premise with Flink • Run Cascading jobs on Flink, with almost no code change • Benefit from Flink’s vastly better performance than MapReduce • Interactive, web-based data exploration • Machine learning on data streams • Compatibility layer for running Storm code • FlinkTopologyBuilder: one line replacement for existing jobs • Wrappers for Storm Spouts and Bolts • Coming soon: Exactly-once with Storm
  • 31. Deployment options Gelly Table ML SAMOA DataSet(Java/Scala)DataStream Hadoop LocalClusterYARNTezEmbedded Dataflow Dataflow MRQL Table Cascading Streamingdataflowruntime Storm Zeppelin • Start Flink in your IDE / on your machine • Local debugging / development using the same code as on the cluster • “bare metal” standalone installation of Flink on a cluster • Flink on Hadoop YARN (Hadoop 2.2.0+) • Restarts failed containers • Support for Kerberos-secured YARN/HDFS setups
  • 32. The full stack 32 Gelly Table ML SAMOA DataSet (Java/Scala) DataStream HadoopM/R Local Cluster Yarn Tez Embedded Dataflow Dataflow(WiP) MRQL Table Cascading Streaming dataflow runtime Storm(WiP) Zeppelin
  • 34. tl;dr Summary Flink is a software stack of  Streaming runtime • low latency • high throughput • fault tolerant, exactly-once data processing  Rich APIs for batch and stream processing • library ecosystem • integration with many systems  A great community of devs and users  Used in production 34
  • 35. What is currently happening?  Features in progress: • Master High Availability • Vastly improved monitoring GUI • Watermarks / Event time processing / Windowing rework • Graduate Streaming API out of Beta  0.10.0-milestone-1 is currently voted 35
  • 36. How do I get started? 36 Mailing Lists: (news | user | dev)@flink.apache.org Twitter: @ApacheFlink Blogs: flink.apache.org/blog, data-artisans.com/blog/ IRC channel: irc.freenode.net#flink Start Flink on YARN in 4 commands: # get the hadoop2 package from the Flink download page at # http://flink.apache.org/downloads.html wget <download url> tar xvzf flink-0.9.1-bin-hadoop2.tgz cd flink-0.9.1/ ./bin/flink run -m yarn-cluster -yn 4 ./examples/flink-java- examples-0.9.1-WordCount.jar
  • 37. flink.apache.org 37 Flink Forward: 2 days conference with free training in Berlin, Germany • Schedule: http://flink-forward.org/?post_type=day
  • 39. Managed (off-heap) memory and out-of- core support 39Memory runs out
  • 40. Cost-based Optimizer 40 DataSource orders.tbl Filter Map DataSource lineitem.tbl Join Hybrid Hash buildHT probe broadcast forward Combine GroupRed sort DataSource orders.tbl Filter Map DataSource lineitem.tbl Join Hybrid Hash buildHT probe hash-part [0] hash-part [0] hash-part [0,1] GroupRed sort forward Best plan depends on relative sizes of input files
  • 41. Iterative processing in Flink Flink offers built-in iterations and delta iterations to execute ML and graph algorithms efficiently 41 map join sum ID1 ID2 ID3
  • 42. Example: Matrix Factorization 42 Factorizing a matrix with 28 billion ratings for recommendations More at: http://data-artisans.com/computing-recommendations-with-flink.html

Notes de l'éditeur

  1. 10:30 – 11:30: (in the event page its until 11:20) – 40 minutes --> 20 slides
  2. GCE 30 instances with 4 cores and 15 GB of memory each. Flink master from July, 24th, Storm 0.9.3. All the code used for the evaluation can be found here. Flink 1.5 million elements per second per core Aggregate Throughput in cluster 182 million elements per second. Storm 82,000 elements per second per core Aggregate 0.57 million elements per second Storm with Acknowledge 4,700 elements per second per core, Latency 30-120 milliseconds Trident: 75,000 elements per second per core
  3. Flink 720,000 events per second per core 690,000 with checkpointing activated Storm With at-least-once: 2,600 events per second per core
  4. Flink 720,000 events per second per core 690,000 with checkpointing activated Storm With at-least-once: 2,600 events per second per core
  5. Flink 0 Buffer timeout: latency median 0 msec, 99 %tile 20 msec 24,500 events per second per core