SlideShare a Scribd company logo
Company 
LOGO 
An Introduction of JStorm 
LongdaFeng(zhongyan.feng@alibaba-inc.com)
Longda Feng 
Alibaba 
Agenda 
Background 
Basic Concept & Scenarios 
Why start JStorm? 
JStorm vs Storm 
Question and Answer.
Who are we? 
JStorm Team was among one of the 
earliest that uses Storm in China. 
 Storm 0.5.1/0.5.4/0.6.0/0.6.2/0.7.0/0.7.1 
 JStorm 0.7.1/0.9.0/0.9.1/0.9.2/0.9.3/… 
Our Duties 
 Application Development 
 JStorm System Development 
 JStorm System Operation 
Longda Feng 
Alibaba
Who are Using JStorm 
Many small Chinese companies are using 
JStorm 
Longda Feng 
Alibaba
How Big? 
More than 3000 servers 
More than 3 trillion messages per day 
Longda Feng 
Alibaba
What is JStorm? 
JStorm is a distributed programming 
framework 
 Similar to Hadoop MapReduce but designed 
for real-time/in-memory scenarios 
 Users can build powerful distributed 
applications from very simple APIs 
Longda Feng 
Alibaba
What is JStorm? 
Redesigned Storm in Java. 
 Proved stable running in huge clusters. 
 Much faster 
 Much more powerful 
Longda Feng 
Alibaba
Basic Conception 
Pipe-lined data processing 
Longda Feng 
Alibaba
Advantage 1 
Easy learning: 
 Simple Building Blocks: Topology/Spout/Bolt 
APIs 
 Out of Box RPC/Fault-tolerance/Real-time 
Data Grouping & Combining 
Longda Feng 
Alibaba
Advantage 2 
Excellent Scalability 
 Horizontally Scalable 
 DAG-based 
 Adjustable parallelism of each component 
Longda Feng 
Alibaba
Stable 
Guarantees Fault-Tolerance 
 No Single Point of Failure 
• Nimbus HA 
• Any Supervisor can be shutdown 
 New worker will be spawned and replace the 
failed one automatically 
Longda Feng 
Alibaba
Accuracy 
Acking framework guarantees no lost of 
data 
Transaction framework guarantees data 
accuracy. 
Longda Feng 
Alibaba
Scenarios 
Stateless Computation 
 All data come from Tuple 
Use Cases: 
 Log Analysis 
 Pipe-lined System 
 Message converter 
 Statistical Analysis 
 Real-time Recommendation Algorithm 
Longda Feng 
Alibaba
Longda Feng 
Alibaba 
Why start JStorm 
Storm community is not as active as we’ve 
expected 
 Tailored for enterprise environment 
 Fixed critical bugs in Storm 
 Provided professional technical support, 
improved app development pace. 
 Reduced operational cost. 

How Many Versions? 
https://github.com/alibaba/JStorm/releases 
 0.9.6(2014/9/22) 
 0.9.5.1(2014/9/14) 
 0.9.5 (2014/8/27) 
 0.9.4.1 (2014/8/15) 
 0.9.4(2014/7/18) 
 0.9.3.1 (2014/5/31) 
 0.9.3 (2014/5/10) 
 0.9.2 (2014/4/8) 
 0.9.1(2014/1/24) 
 0.9.0(2013/12/30) 
 0.7.1(2013/4/28) 
Longda Feng 
Alibaba
JStorm is a superset of Storm 
The program run in Storm can run in 
JStorm without changing code 
Longda Feng 
Alibaba
More stable (1) -- nimbus HA 
Nimbus HA 
 Dual-Nimbus HA 
Longda Feng 
Alibaba
More stable (2) -- RPC 
Netty supports 2 RPC modes 
 Async 
 Sync 
• Sending speed keeps up with the receiving speed, 
therefore the data flow is more stable. 
Longda Feng 
Alibaba
More stable(3) – resource isolation 
Malicious Worker won’t mess up with 
others 
 Supported CPU Isolation with cgroups 
 Supported Memory Isolation 
Resources quota can be enforced on each 
group (before 0.9.5) 
Longda Feng 
Alibaba
More stable(4) -- Monitor 
Monitor every component in your 
Topology 
 Many more metrics(70+) than storm 
Supported user-defined metrics 
Supported user-defined alerts 
Longda Feng 
Alibaba
More stable (5) – CPU usage 
Better utilizing CPU resource 
 Improved disruptor implementation 
• Drop CPU usage from 300% to 10% when 
processing queue is full 
 Avoid CPU spin-waiting 
• Relocating nextTuple/ack/fail work to a different 
thread 
Longda Feng 
Alibaba
More stable(6) -- more catch 
Add try-catch in any place. 
 Nimbus/supervisor main thread 
 Spout/bolt initialization/cleanup 
 All IO operation, serialization/deserialization 
 All ZK operation 
Longda Feng 
Alibaba
More stable(7) -- ZK 
Reduced unnecessary ZK usage: 
 Removed useless watcher 
 Increased ZK heartbeat frequency 
 Detect failed worker without a full scan of the 
entire ZK directory 
Longda Feng 
Alibaba
More stable(8) -- other 
Improved GC Tuning. 
Guaranteed that all workers killed after kill 
command is issued 
Guaranteed single supervisor/nimbus per 
instance 
Avoid excessive use of local ports by 
Netty client 
。。。 
Longda Feng 
Alibaba
More powerful scheduler 
Balancing Tasks with regard of : 
 CPU 
 Memory 
 Net 
Longda Feng 
Alibaba
CPU assignment 
By default assign each worker a single 
CPU slot 
Application can be configured to utilize 
more slots 
Why: 
 Some task creates extra threads to do other 
things in Alimama, one CPU slot doesn’t meet 
requirement 
Longda Feng 
Alibaba
Memory Usage 
Default worker memory is 2G 
Application can be configured to utilize 
more memory slots 
Why: 
 In Alipay Mdrill application, Solr bolt will apply 
much more memory 
Longda Feng 
Alibaba
Smarter Balancing 
With JStorm Scheduler: 
 Tasks that exchange data heavily tend to be 
assigned to the same worker to avoid 
networking cost. 
Longda Feng 
Alibaba
User Defined Scheduler 
User define task run one designated 
worker 
 User can setting how many CPU slot /memory 
slot will be used 
Why: 
 In Taobao TAE project, some bolts want to 
run in user defined-nodes 
Longda Feng 
Alibaba
Task on Different Node 
Task of one component can be scheduled 
to run on different nodes 
Why: 
 In ALIPAY Mdrill, Solr bolt must run different 
node 
Longda Feng 
Alibaba
Task on Single Node 
All tasks can be scheduled to run on a 
single node. 
Why: 
 In Taobao TLog, there are many small jobs, in 
order to reduce network cost, all task of one 
job must run on single node. 
Longda Feng 
Alibaba
Old Assignment 
“Last Assignment Policy” 
 By default , a task will run on the machine it 
runs previous time 
Why: 
 In Alibaba CDO, When restart one application, 
user wanted to reuse old workers 
Longda Feng 
Alibaba
Pluginable 
Be able to run on: 
Hadoop yarn(more stable than storm) 
Alibaba Apsara Clould System 
Alibaba Elastic Resource Pool 
Longda Feng 
Alibaba
Classloader 
Resolved application jar-confliction with 
JStorm 
Longda Feng 
Alibaba
More convenient UI 
More useful stats collected and displayed. 
Browse Worker Log in UI 
Longda Feng 
Alibaba
Support libjar 
Don’t need assembly all dependency jars 
into one jar 
 Submit libjar with libjar parameter 
 Support worker.classpath 
Longda Feng 
Alibaba
Faster 
6 Servers (24core/98G) 
18 Spout/18 Bolt/18 Acker 
Longda Feng 
Alibaba 
9280598 
10818815 
Throughput vs workers 
9065965 
6819139 
5610201 
6243680 
6830500 
5595900 5474180 
3379800 
12000000 
10000000 
8000000 
6000000 
4000000 
2000000 
0 
0 10 20 30 40 50 60 
polltuples/10s 
workers 
jstorm 
storm
JStorm 41W/S Sending Speed 
Longda Feng 
Alibaba
Storm 41W/S Sending Speed 
Longda Feng 
Alibaba
Why Faster 
Reduce memory-copying by zeroMq 
Dedicated Deserializing Thread 
Better Tuned Sampling Logic 
Better Tuned Acking Framework 
Better Tuned GC 
Longda Feng 
Alibaba
Other Improvement 
 More than 100 improvements 
 https://github.com/alibaba/JStorm/blob/master/history.md 
 Fixed assign topology competition 
 Reset rebalance/reassigned worker timeout as 4 minutes 
 Graceful worker shutdown 
 Improvement on thrift server 
 Avoid mistakenly killing of worker while rebalancing jobs. 
 。。。。 
Longda Feng 
Alibaba
More document 
https://github.com/alibaba/JStorm/wiki 
Google-group:JStorm-user@googlegroups.com 
Wangwang:JStorm 
QQ:228374502 
Laiwang: JStorm 
Longda Feng 
Alibaba
Join us 
Welcome to Join us 
 zhongyan.feng@alibaba-inc.com 
Longda Feng 
Alibaba
Company 
LOGO 
纪君祥(Longda Feng)

More Related Content

What's hot

Introduction to Hadoop - The Essentials
Introduction to Hadoop - The EssentialsIntroduction to Hadoop - The Essentials
Introduction to Hadoop - The Essentials
Fadi Yousuf
 
Hadoop 2.0 Architecture | HDFS Federation | NameNode High Availability |
Hadoop 2.0 Architecture | HDFS Federation | NameNode High Availability | Hadoop 2.0 Architecture | HDFS Federation | NameNode High Availability |
Hadoop 2.0 Architecture | HDFS Federation | NameNode High Availability | Edureka!
 
Hadoop
Hadoop Hadoop
Hadoop
ABHIJEET RAJ
 
Hadoop Backup and Disaster Recovery
Hadoop Backup and Disaster RecoveryHadoop Backup and Disaster Recovery
Hadoop Backup and Disaster Recovery
Cloudera, Inc.
 
Introduction to Data Analyst Training
Introduction to Data Analyst TrainingIntroduction to Data Analyst Training
Introduction to Data Analyst TrainingCloudera, Inc.
 
Hadoop Interview Questions and Answers by rohit kapa
Hadoop Interview Questions and Answers by rohit kapaHadoop Interview Questions and Answers by rohit kapa
Hadoop Interview Questions and Answers by rohit kapa
kapa rohit
 
Boston Hadoop Meetup, April 26 2012
Boston Hadoop Meetup, April 26 2012Boston Hadoop Meetup, April 26 2012
Boston Hadoop Meetup, April 26 2012
Daniel Abadi
 
Design, Scale and Performance of MapR's Distribution for Hadoop
Design, Scale and Performance of MapR's Distribution for HadoopDesign, Scale and Performance of MapR's Distribution for Hadoop
Design, Scale and Performance of MapR's Distribution for Hadoop
mcsrivas
 
Hadoop
HadoopHadoop
SQL-on-Hadoop Tutorial
SQL-on-Hadoop TutorialSQL-on-Hadoop Tutorial
SQL-on-Hadoop Tutorial
Daniel Abadi
 
Introduction to Hadoop and Cloudera, Louisville BI & Big Data Analytics Meetup
Introduction to Hadoop and Cloudera, Louisville BI & Big Data Analytics MeetupIntroduction to Hadoop and Cloudera, Louisville BI & Big Data Analytics Meetup
Introduction to Hadoop and Cloudera, Louisville BI & Big Data Analytics Meetup
iwrigley
 
HUG August 2010: Best practices
HUG August 2010: Best practicesHUG August 2010: Best practices
HUG August 2010: Best practices
Hadoop User Group
 
Introduction to Hadoop
Introduction to HadoopIntroduction to Hadoop
Introduction to Hadoop
Dr. C.V. Suresh Babu
 
An Introduction to Hadoop and Cloudera: Nashville Cloudera User Group, 10/23/14
An Introduction to Hadoop and Cloudera: Nashville Cloudera User Group, 10/23/14An Introduction to Hadoop and Cloudera: Nashville Cloudera User Group, 10/23/14
An Introduction to Hadoop and Cloudera: Nashville Cloudera User Group, 10/23/14
iwrigley
 
Hadoop_Its_Not_Just_Internal_Storage_V14
Hadoop_Its_Not_Just_Internal_Storage_V14Hadoop_Its_Not_Just_Internal_Storage_V14
Hadoop_Its_Not_Just_Internal_Storage_V14
John Sing
 
Hadoop World 2011: Hadoop and RDBMS with Sqoop and Other Tools - Guy Harrison...
Hadoop World 2011: Hadoop and RDBMS with Sqoop and Other Tools - Guy Harrison...Hadoop World 2011: Hadoop and RDBMS with Sqoop and Other Tools - Guy Harrison...
Hadoop World 2011: Hadoop and RDBMS with Sqoop and Other Tools - Guy Harrison...
Cloudera, Inc.
 
Shared slides-edbt-keynote-03-19-13
Shared slides-edbt-keynote-03-19-13Shared slides-edbt-keynote-03-19-13
Shared slides-edbt-keynote-03-19-13
Daniel Abadi
 
Apache hadoop technology : Beginners
Apache hadoop technology : BeginnersApache hadoop technology : Beginners
Apache hadoop technology : Beginners
Shweta Patnaik
 
Apache kudu
Apache kuduApache kudu
Apache kudu
Asim Jalis
 

What's hot (20)

Introduction to Hadoop - The Essentials
Introduction to Hadoop - The EssentialsIntroduction to Hadoop - The Essentials
Introduction to Hadoop - The Essentials
 
Hadoop 2.0 Architecture | HDFS Federation | NameNode High Availability |
Hadoop 2.0 Architecture | HDFS Federation | NameNode High Availability | Hadoop 2.0 Architecture | HDFS Federation | NameNode High Availability |
Hadoop 2.0 Architecture | HDFS Federation | NameNode High Availability |
 
Hadoop
Hadoop Hadoop
Hadoop
 
Hadoop Backup and Disaster Recovery
Hadoop Backup and Disaster RecoveryHadoop Backup and Disaster Recovery
Hadoop Backup and Disaster Recovery
 
Introduction to Data Analyst Training
Introduction to Data Analyst TrainingIntroduction to Data Analyst Training
Introduction to Data Analyst Training
 
Hadoop Interview Questions and Answers by rohit kapa
Hadoop Interview Questions and Answers by rohit kapaHadoop Interview Questions and Answers by rohit kapa
Hadoop Interview Questions and Answers by rohit kapa
 
Boston Hadoop Meetup, April 26 2012
Boston Hadoop Meetup, April 26 2012Boston Hadoop Meetup, April 26 2012
Boston Hadoop Meetup, April 26 2012
 
Design, Scale and Performance of MapR's Distribution for Hadoop
Design, Scale and Performance of MapR's Distribution for HadoopDesign, Scale and Performance of MapR's Distribution for Hadoop
Design, Scale and Performance of MapR's Distribution for Hadoop
 
Hadoop
HadoopHadoop
Hadoop
 
SQL-on-Hadoop Tutorial
SQL-on-Hadoop TutorialSQL-on-Hadoop Tutorial
SQL-on-Hadoop Tutorial
 
Introduction to Hadoop and Cloudera, Louisville BI & Big Data Analytics Meetup
Introduction to Hadoop and Cloudera, Louisville BI & Big Data Analytics MeetupIntroduction to Hadoop and Cloudera, Louisville BI & Big Data Analytics Meetup
Introduction to Hadoop and Cloudera, Louisville BI & Big Data Analytics Meetup
 
HUG August 2010: Best practices
HUG August 2010: Best practicesHUG August 2010: Best practices
HUG August 2010: Best practices
 
Introduction to Hadoop
Introduction to HadoopIntroduction to Hadoop
Introduction to Hadoop
 
An Introduction to Hadoop and Cloudera: Nashville Cloudera User Group, 10/23/14
An Introduction to Hadoop and Cloudera: Nashville Cloudera User Group, 10/23/14An Introduction to Hadoop and Cloudera: Nashville Cloudera User Group, 10/23/14
An Introduction to Hadoop and Cloudera: Nashville Cloudera User Group, 10/23/14
 
Hadoop_Its_Not_Just_Internal_Storage_V14
Hadoop_Its_Not_Just_Internal_Storage_V14Hadoop_Its_Not_Just_Internal_Storage_V14
Hadoop_Its_Not_Just_Internal_Storage_V14
 
Hadoop World 2011: Hadoop and RDBMS with Sqoop and Other Tools - Guy Harrison...
Hadoop World 2011: Hadoop and RDBMS with Sqoop and Other Tools - Guy Harrison...Hadoop World 2011: Hadoop and RDBMS with Sqoop and Other Tools - Guy Harrison...
Hadoop World 2011: Hadoop and RDBMS with Sqoop and Other Tools - Guy Harrison...
 
10c introduction
10c introduction10c introduction
10c introduction
 
Shared slides-edbt-keynote-03-19-13
Shared slides-edbt-keynote-03-19-13Shared slides-edbt-keynote-03-19-13
Shared slides-edbt-keynote-03-19-13
 
Apache hadoop technology : Beginners
Apache hadoop technology : BeginnersApache hadoop technology : Beginners
Apache hadoop technology : Beginners
 
Apache kudu
Apache kuduApache kudu
Apache kudu
 

Viewers also liked

Quant trading theory series: electronic markets
Quant trading theory series: electronic marketsQuant trading theory series: electronic markets
Quant trading theory series: electronic markets
Oliver Laslett
 
Tabobao Reinventing (E-)Commerce
Tabobao Reinventing (E-)CommerceTabobao Reinventing (E-)Commerce
Tabobao Reinventing (E-)Commerce
Farzan Dehbashi
 
Semiconductor industry in china20151126R1.2
Semiconductor industry in china20151126R1.2Semiconductor industry in china20151126R1.2
Semiconductor industry in china20151126R1.2xiaolong zheng
 
Monitoring Electronic Trading Environments using Spark by Fergal Toomey and P...
Monitoring Electronic Trading Environments using Spark by Fergal Toomey and P...Monitoring Electronic Trading Environments using Spark by Fergal Toomey and P...
Monitoring Electronic Trading Environments using Spark by Fergal Toomey and P...
Spark Summit
 
Alibaba & e-Commerce in China
Alibaba & e-Commerce in ChinaAlibaba & e-Commerce in China
Alibaba & e-Commerce in China
Ed Sander
 
Online Trading
Online TradingOnline Trading
Online Tradingkoolzub
 
TradeZilla - Trading system Design
TradeZilla - Trading system DesignTradeZilla - Trading system Design
TradeZilla - Trading system Design
Marketcalls
 
Electronic Trading: A Primer
Electronic Trading: A PrimerElectronic Trading: A Primer
Electronic Trading: A Primer
ManagedFunds
 
Competitive analysis of it service firms
Competitive analysis of it service firmsCompetitive analysis of it service firms
Competitive analysis of it service firmsSayan Maiti
 
The Alibaba Group Explained
The Alibaba Group ExplainedThe Alibaba Group Explained
The Alibaba Group Explained
Ranjan Roy
 
Wipro presentation
Wipro presentationWipro presentation
Wipro presentationmanishkr90
 
WIPRO PROJECT FILE
WIPRO PROJECT FILEWIPRO PROJECT FILE
WIPRO PROJECT FILE
Harsh Yadav
 
Tcs ppt
Tcs pptTcs ppt
Tcs ppt
Aqsa Javed
 
Online trading ppt
Online trading ppt Online trading ppt
Online trading ppt petkarshwt
 
Fixed Income Trading System Architecture
Fixed Income Trading System ArchitectureFixed Income Trading System Architecture
Fixed Income Trading System Architecture
Khader Shaik
 
7 steps to cloud onboarding
7 steps to cloud onboarding7 steps to cloud onboarding
7 steps to cloud onboarding
Interxion
 
Indian IT Industry overview
Indian IT Industry overviewIndian IT Industry overview
LinkedIn Data Infrastructure (QCon London 2012)
LinkedIn Data Infrastructure (QCon London 2012)LinkedIn Data Infrastructure (QCon London 2012)
LinkedIn Data Infrastructure (QCon London 2012)
Sid Anand
 

Viewers also liked (19)

Quant trading theory series: electronic markets
Quant trading theory series: electronic marketsQuant trading theory series: electronic markets
Quant trading theory series: electronic markets
 
Tabobao Reinventing (E-)Commerce
Tabobao Reinventing (E-)CommerceTabobao Reinventing (E-)Commerce
Tabobao Reinventing (E-)Commerce
 
Semiconductor industry in china20151126R1.2
Semiconductor industry in china20151126R1.2Semiconductor industry in china20151126R1.2
Semiconductor industry in china20151126R1.2
 
Monitoring Electronic Trading Environments using Spark by Fergal Toomey and P...
Monitoring Electronic Trading Environments using Spark by Fergal Toomey and P...Monitoring Electronic Trading Environments using Spark by Fergal Toomey and P...
Monitoring Electronic Trading Environments using Spark by Fergal Toomey and P...
 
Alibaba & e-Commerce in China
Alibaba & e-Commerce in ChinaAlibaba & e-Commerce in China
Alibaba & e-Commerce in China
 
Online Trading
Online TradingOnline Trading
Online Trading
 
TradeZilla - Trading system Design
TradeZilla - Trading system DesignTradeZilla - Trading system Design
TradeZilla - Trading system Design
 
Electronic Trading: A Primer
Electronic Trading: A PrimerElectronic Trading: A Primer
Electronic Trading: A Primer
 
Competitive analysis of it service firms
Competitive analysis of it service firmsCompetitive analysis of it service firms
Competitive analysis of it service firms
 
The Alibaba Group Explained
The Alibaba Group ExplainedThe Alibaba Group Explained
The Alibaba Group Explained
 
Wipro presentation
Wipro presentationWipro presentation
Wipro presentation
 
WIPRO PROJECT FILE
WIPRO PROJECT FILEWIPRO PROJECT FILE
WIPRO PROJECT FILE
 
Tcs ppt
Tcs pptTcs ppt
Tcs ppt
 
Online trading ppt
Online trading ppt Online trading ppt
Online trading ppt
 
Fixed Income Trading System Architecture
Fixed Income Trading System ArchitectureFixed Income Trading System Architecture
Fixed Income Trading System Architecture
 
7 steps to cloud onboarding
7 steps to cloud onboarding7 steps to cloud onboarding
7 steps to cloud onboarding
 
Indian IT Industry overview
Indian IT Industry overviewIndian IT Industry overview
Indian IT Industry overview
 
LinkedIn Data Infrastructure (QCon London 2012)
LinkedIn Data Infrastructure (QCon London 2012)LinkedIn Data Infrastructure (QCon London 2012)
LinkedIn Data Infrastructure (QCon London 2012)
 
E commerce
E commerceE commerce
E commerce
 

Similar to Jstorm introduction-0.9.6

JStorm Introduction
JStorm IntroductionJStorm Introduction
JStorm Introduction
longda feng
 
Low latency in java 8 v5
Low latency in java 8 v5Low latency in java 8 v5
Low latency in java 8 v5
Peter Lawrey
 
Comparison of various streaming technologies
Comparison of various streaming technologiesComparison of various streaming technologies
Comparison of various streaming technologies
Sachin Aggarwal
 
Five cool ways the JVM can run Apache Spark faster
Five cool ways the JVM can run Apache Spark fasterFive cool ways the JVM can run Apache Spark faster
Five cool ways the JVM can run Apache Spark faster
Tim Ellison
 
2019 StartIT - Boosting your performance with Blackfire
2019 StartIT - Boosting your performance with Blackfire2019 StartIT - Boosting your performance with Blackfire
2019 StartIT - Boosting your performance with Blackfire
Marko Mitranić
 
Things You MUST Know Before Deploying OpenStack: Bruno Lago, Catalyst IT
Things You MUST Know Before Deploying OpenStack: Bruno Lago, Catalyst ITThings You MUST Know Before Deploying OpenStack: Bruno Lago, Catalyst IT
Things You MUST Know Before Deploying OpenStack: Bruno Lago, Catalyst IT
OpenStack
 
Low latency in java 8 by Peter Lawrey
Low latency in java 8 by Peter Lawrey Low latency in java 8 by Peter Lawrey
Low latency in java 8 by Peter Lawrey
J On The Beach
 
Java Performance and Profiling
Java Performance and ProfilingJava Performance and Profiling
Java Performance and Profiling
WSO2
 
Azul yandexjune010
Azul yandexjune010Azul yandexjune010
Azul yandexjune010yaevents
 
Basic architecuture and operation concept of Backlog and Cacoo
Basic architecuture and operation concept of Backlog and CacooBasic architecuture and operation concept of Backlog and Cacoo
Basic architecuture and operation concept of Backlog and Cacoo
Takashi Someda
 
Azure + DataStax Enterprise (DSE) Powers Office365 Per User Store
Azure + DataStax Enterprise (DSE) Powers Office365 Per User StoreAzure + DataStax Enterprise (DSE) Powers Office365 Per User Store
Azure + DataStax Enterprise (DSE) Powers Office365 Per User Store
DataStax Academy
 
Google App Engine for Java v0.0.2
Google App Engine for Java v0.0.2Google App Engine for Java v0.0.2
Google App Engine for Java v0.0.2
Matthew McCullough
 
Google App Engine for Java
Google App Engine for JavaGoogle App Engine for Java
Google App Engine for Java
Lars Vogel
 
MySQL Scalability and Reliability for Replicated Environment
MySQL Scalability and Reliability for Replicated EnvironmentMySQL Scalability and Reliability for Replicated Environment
MySQL Scalability and Reliability for Replicated Environment
Jean-François Gagné
 
On The Building Of A PostgreSQL Cluster
On The Building Of A PostgreSQL ClusterOn The Building Of A PostgreSQL Cluster
On The Building Of A PostgreSQL Cluster
Srihari Sriraman
 
murakumo Cloud Controller
murakumo Cloud Controllermurakumo Cloud Controller
murakumo Cloud Controller
Shingo Kawano
 
Java Performance and Using Java Flight Recorder
Java Performance and Using Java Flight RecorderJava Performance and Using Java Flight Recorder
Java Performance and Using Java Flight Recorder
Isuru Perera
 
Java Tuning White Paper
Java Tuning White PaperJava Tuning White Paper
Java Tuning White Paperwhite paper
 
[RakutenTechConf2013] [C-1] Rakuten new infrastructure
[RakutenTechConf2013] [C-1] Rakuten new infrastructure[RakutenTechConf2013] [C-1] Rakuten new infrastructure
[RakutenTechConf2013] [C-1] Rakuten new infrastructure
Rakuten Group, Inc.
 

Similar to Jstorm introduction-0.9.6 (20)

JStorm Introduction
JStorm IntroductionJStorm Introduction
JStorm Introduction
 
J storm
J stormJ storm
J storm
 
Low latency in java 8 v5
Low latency in java 8 v5Low latency in java 8 v5
Low latency in java 8 v5
 
Comparison of various streaming technologies
Comparison of various streaming technologiesComparison of various streaming technologies
Comparison of various streaming technologies
 
Five cool ways the JVM can run Apache Spark faster
Five cool ways the JVM can run Apache Spark fasterFive cool ways the JVM can run Apache Spark faster
Five cool ways the JVM can run Apache Spark faster
 
2019 StartIT - Boosting your performance with Blackfire
2019 StartIT - Boosting your performance with Blackfire2019 StartIT - Boosting your performance with Blackfire
2019 StartIT - Boosting your performance with Blackfire
 
Things You MUST Know Before Deploying OpenStack: Bruno Lago, Catalyst IT
Things You MUST Know Before Deploying OpenStack: Bruno Lago, Catalyst ITThings You MUST Know Before Deploying OpenStack: Bruno Lago, Catalyst IT
Things You MUST Know Before Deploying OpenStack: Bruno Lago, Catalyst IT
 
Low latency in java 8 by Peter Lawrey
Low latency in java 8 by Peter Lawrey Low latency in java 8 by Peter Lawrey
Low latency in java 8 by Peter Lawrey
 
Java Performance and Profiling
Java Performance and ProfilingJava Performance and Profiling
Java Performance and Profiling
 
Azul yandexjune010
Azul yandexjune010Azul yandexjune010
Azul yandexjune010
 
Basic architecuture and operation concept of Backlog and Cacoo
Basic architecuture and operation concept of Backlog and CacooBasic architecuture and operation concept of Backlog and Cacoo
Basic architecuture and operation concept of Backlog and Cacoo
 
Azure + DataStax Enterprise (DSE) Powers Office365 Per User Store
Azure + DataStax Enterprise (DSE) Powers Office365 Per User StoreAzure + DataStax Enterprise (DSE) Powers Office365 Per User Store
Azure + DataStax Enterprise (DSE) Powers Office365 Per User Store
 
Google App Engine for Java v0.0.2
Google App Engine for Java v0.0.2Google App Engine for Java v0.0.2
Google App Engine for Java v0.0.2
 
Google App Engine for Java
Google App Engine for JavaGoogle App Engine for Java
Google App Engine for Java
 
MySQL Scalability and Reliability for Replicated Environment
MySQL Scalability and Reliability for Replicated EnvironmentMySQL Scalability and Reliability for Replicated Environment
MySQL Scalability and Reliability for Replicated Environment
 
On The Building Of A PostgreSQL Cluster
On The Building Of A PostgreSQL ClusterOn The Building Of A PostgreSQL Cluster
On The Building Of A PostgreSQL Cluster
 
murakumo Cloud Controller
murakumo Cloud Controllermurakumo Cloud Controller
murakumo Cloud Controller
 
Java Performance and Using Java Flight Recorder
Java Performance and Using Java Flight RecorderJava Performance and Using Java Flight Recorder
Java Performance and Using Java Flight Recorder
 
Java Tuning White Paper
Java Tuning White PaperJava Tuning White Paper
Java Tuning White Paper
 
[RakutenTechConf2013] [C-1] Rakuten new infrastructure
[RakutenTechConf2013] [C-1] Rakuten new infrastructure[RakutenTechConf2013] [C-1] Rakuten new infrastructure
[RakutenTechConf2013] [C-1] Rakuten new infrastructure
 

Recently uploaded

急速办(bedfordhire毕业证书)英国贝德福特大学毕业证成绩单原版一模一样
急速办(bedfordhire毕业证书)英国贝德福特大学毕业证成绩单原版一模一样急速办(bedfordhire毕业证书)英国贝德福特大学毕业证成绩单原版一模一样
急速办(bedfordhire毕业证书)英国贝德福特大学毕业证成绩单原版一模一样
3ipehhoa
 
Latest trends in computer networking.pptx
Latest trends in computer networking.pptxLatest trends in computer networking.pptx
Latest trends in computer networking.pptx
JungkooksNonexistent
 
1比1复刻(bath毕业证书)英国巴斯大学毕业证学位证原版一模一样
1比1复刻(bath毕业证书)英国巴斯大学毕业证学位证原版一模一样1比1复刻(bath毕业证书)英国巴斯大学毕业证学位证原版一模一样
1比1复刻(bath毕业证书)英国巴斯大学毕业证学位证原版一模一样
3ipehhoa
 
APNIC Foundation, presented by Ellisha Heppner at the PNG DNS Forum 2024
APNIC Foundation, presented by Ellisha Heppner at the PNG DNS Forum 2024APNIC Foundation, presented by Ellisha Heppner at the PNG DNS Forum 2024
APNIC Foundation, presented by Ellisha Heppner at the PNG DNS Forum 2024
APNIC
 
Comptia N+ Standard Networking lesson guide
Comptia N+ Standard Networking lesson guideComptia N+ Standard Networking lesson guide
Comptia N+ Standard Networking lesson guide
GTProductions1
 
History+of+E-commerce+Development+in+China-www.cfye-commerce.shop
History+of+E-commerce+Development+in+China-www.cfye-commerce.shopHistory+of+E-commerce+Development+in+China-www.cfye-commerce.shop
History+of+E-commerce+Development+in+China-www.cfye-commerce.shop
laozhuseo02
 
The+Prospects+of+E-Commerce+in+China.pptx
The+Prospects+of+E-Commerce+in+China.pptxThe+Prospects+of+E-Commerce+in+China.pptx
The+Prospects+of+E-Commerce+in+China.pptx
laozhuseo02
 
JAVIER LASA-EXPERIENCIA digital 1986-2024.pdf
JAVIER LASA-EXPERIENCIA digital 1986-2024.pdfJAVIER LASA-EXPERIENCIA digital 1986-2024.pdf
JAVIER LASA-EXPERIENCIA digital 1986-2024.pdf
Javier Lasa
 
This 7-second Brain Wave Ritual Attracts Money To You.!
This 7-second Brain Wave Ritual Attracts Money To You.!This 7-second Brain Wave Ritual Attracts Money To You.!
This 7-second Brain Wave Ritual Attracts Money To You.!
nirahealhty
 
一比一原版(SLU毕业证)圣路易斯大学毕业证成绩单专业办理
一比一原版(SLU毕业证)圣路易斯大学毕业证成绩单专业办理一比一原版(SLU毕业证)圣路易斯大学毕业证成绩单专业办理
一比一原版(SLU毕业证)圣路易斯大学毕业证成绩单专业办理
keoku
 
1.Wireless Communication System_Wireless communication is a broad term that i...
1.Wireless Communication System_Wireless communication is a broad term that i...1.Wireless Communication System_Wireless communication is a broad term that i...
1.Wireless Communication System_Wireless communication is a broad term that i...
JeyaPerumal1
 
Bài tập unit 1 English in the world.docx
Bài tập unit 1 English in the world.docxBài tập unit 1 English in the world.docx
Bài tập unit 1 English in the world.docx
nhiyenphan2005
 
Meet up Milano 14 _ Axpo Italia_ Migration from Mule3 (On-prem) to.pdf
Meet up Milano 14 _ Axpo Italia_ Migration from Mule3 (On-prem) to.pdfMeet up Milano 14 _ Axpo Italia_ Migration from Mule3 (On-prem) to.pdf
Meet up Milano 14 _ Axpo Italia_ Migration from Mule3 (On-prem) to.pdf
Florence Consulting
 
Italy Agriculture Equipment Market Outlook to 2027
Italy Agriculture Equipment Market Outlook to 2027Italy Agriculture Equipment Market Outlook to 2027
Italy Agriculture Equipment Market Outlook to 2027
harveenkaur52
 
原版仿制(uob毕业证书)英国伯明翰大学毕业证本科学历证书原版一模一样
原版仿制(uob毕业证书)英国伯明翰大学毕业证本科学历证书原版一模一样原版仿制(uob毕业证书)英国伯明翰大学毕业证本科学历证书原版一模一样
原版仿制(uob毕业证书)英国伯明翰大学毕业证本科学历证书原版一模一样
3ipehhoa
 
test test test test testtest test testtest test testtest test testtest test ...
test test  test test testtest test testtest test testtest test testtest test ...test test  test test testtest test testtest test testtest test testtest test ...
test test test test testtest test testtest test testtest test testtest test ...
Arif0071
 
Internet of Things in Manufacturing: Revolutionizing Efficiency & Quality | C...
Internet of Things in Manufacturing: Revolutionizing Efficiency & Quality | C...Internet of Things in Manufacturing: Revolutionizing Efficiency & Quality | C...
Internet of Things in Manufacturing: Revolutionizing Efficiency & Quality | C...
CIOWomenMagazine
 
一比一原版(LBS毕业证)伦敦商学院毕业证成绩单专业办理
一比一原版(LBS毕业证)伦敦商学院毕业证成绩单专业办理一比一原版(LBS毕业证)伦敦商学院毕业证成绩单专业办理
一比一原版(LBS毕业证)伦敦商学院毕业证成绩单专业办理
eutxy
 
How to Use Contact Form 7 Like a Pro.pptx
How to Use Contact Form 7 Like a Pro.pptxHow to Use Contact Form 7 Like a Pro.pptx
How to Use Contact Form 7 Like a Pro.pptx
Gal Baras
 
guildmasters guide to ravnica Dungeons & Dragons 5...
guildmasters guide to ravnica Dungeons & Dragons 5...guildmasters guide to ravnica Dungeons & Dragons 5...
guildmasters guide to ravnica Dungeons & Dragons 5...
Rogerio Filho
 

Recently uploaded (20)

急速办(bedfordhire毕业证书)英国贝德福特大学毕业证成绩单原版一模一样
急速办(bedfordhire毕业证书)英国贝德福特大学毕业证成绩单原版一模一样急速办(bedfordhire毕业证书)英国贝德福特大学毕业证成绩单原版一模一样
急速办(bedfordhire毕业证书)英国贝德福特大学毕业证成绩单原版一模一样
 
Latest trends in computer networking.pptx
Latest trends in computer networking.pptxLatest trends in computer networking.pptx
Latest trends in computer networking.pptx
 
1比1复刻(bath毕业证书)英国巴斯大学毕业证学位证原版一模一样
1比1复刻(bath毕业证书)英国巴斯大学毕业证学位证原版一模一样1比1复刻(bath毕业证书)英国巴斯大学毕业证学位证原版一模一样
1比1复刻(bath毕业证书)英国巴斯大学毕业证学位证原版一模一样
 
APNIC Foundation, presented by Ellisha Heppner at the PNG DNS Forum 2024
APNIC Foundation, presented by Ellisha Heppner at the PNG DNS Forum 2024APNIC Foundation, presented by Ellisha Heppner at the PNG DNS Forum 2024
APNIC Foundation, presented by Ellisha Heppner at the PNG DNS Forum 2024
 
Comptia N+ Standard Networking lesson guide
Comptia N+ Standard Networking lesson guideComptia N+ Standard Networking lesson guide
Comptia N+ Standard Networking lesson guide
 
History+of+E-commerce+Development+in+China-www.cfye-commerce.shop
History+of+E-commerce+Development+in+China-www.cfye-commerce.shopHistory+of+E-commerce+Development+in+China-www.cfye-commerce.shop
History+of+E-commerce+Development+in+China-www.cfye-commerce.shop
 
The+Prospects+of+E-Commerce+in+China.pptx
The+Prospects+of+E-Commerce+in+China.pptxThe+Prospects+of+E-Commerce+in+China.pptx
The+Prospects+of+E-Commerce+in+China.pptx
 
JAVIER LASA-EXPERIENCIA digital 1986-2024.pdf
JAVIER LASA-EXPERIENCIA digital 1986-2024.pdfJAVIER LASA-EXPERIENCIA digital 1986-2024.pdf
JAVIER LASA-EXPERIENCIA digital 1986-2024.pdf
 
This 7-second Brain Wave Ritual Attracts Money To You.!
This 7-second Brain Wave Ritual Attracts Money To You.!This 7-second Brain Wave Ritual Attracts Money To You.!
This 7-second Brain Wave Ritual Attracts Money To You.!
 
一比一原版(SLU毕业证)圣路易斯大学毕业证成绩单专业办理
一比一原版(SLU毕业证)圣路易斯大学毕业证成绩单专业办理一比一原版(SLU毕业证)圣路易斯大学毕业证成绩单专业办理
一比一原版(SLU毕业证)圣路易斯大学毕业证成绩单专业办理
 
1.Wireless Communication System_Wireless communication is a broad term that i...
1.Wireless Communication System_Wireless communication is a broad term that i...1.Wireless Communication System_Wireless communication is a broad term that i...
1.Wireless Communication System_Wireless communication is a broad term that i...
 
Bài tập unit 1 English in the world.docx
Bài tập unit 1 English in the world.docxBài tập unit 1 English in the world.docx
Bài tập unit 1 English in the world.docx
 
Meet up Milano 14 _ Axpo Italia_ Migration from Mule3 (On-prem) to.pdf
Meet up Milano 14 _ Axpo Italia_ Migration from Mule3 (On-prem) to.pdfMeet up Milano 14 _ Axpo Italia_ Migration from Mule3 (On-prem) to.pdf
Meet up Milano 14 _ Axpo Italia_ Migration from Mule3 (On-prem) to.pdf
 
Italy Agriculture Equipment Market Outlook to 2027
Italy Agriculture Equipment Market Outlook to 2027Italy Agriculture Equipment Market Outlook to 2027
Italy Agriculture Equipment Market Outlook to 2027
 
原版仿制(uob毕业证书)英国伯明翰大学毕业证本科学历证书原版一模一样
原版仿制(uob毕业证书)英国伯明翰大学毕业证本科学历证书原版一模一样原版仿制(uob毕业证书)英国伯明翰大学毕业证本科学历证书原版一模一样
原版仿制(uob毕业证书)英国伯明翰大学毕业证本科学历证书原版一模一样
 
test test test test testtest test testtest test testtest test testtest test ...
test test  test test testtest test testtest test testtest test testtest test ...test test  test test testtest test testtest test testtest test testtest test ...
test test test test testtest test testtest test testtest test testtest test ...
 
Internet of Things in Manufacturing: Revolutionizing Efficiency & Quality | C...
Internet of Things in Manufacturing: Revolutionizing Efficiency & Quality | C...Internet of Things in Manufacturing: Revolutionizing Efficiency & Quality | C...
Internet of Things in Manufacturing: Revolutionizing Efficiency & Quality | C...
 
一比一原版(LBS毕业证)伦敦商学院毕业证成绩单专业办理
一比一原版(LBS毕业证)伦敦商学院毕业证成绩单专业办理一比一原版(LBS毕业证)伦敦商学院毕业证成绩单专业办理
一比一原版(LBS毕业证)伦敦商学院毕业证成绩单专业办理
 
How to Use Contact Form 7 Like a Pro.pptx
How to Use Contact Form 7 Like a Pro.pptxHow to Use Contact Form 7 Like a Pro.pptx
How to Use Contact Form 7 Like a Pro.pptx
 
guildmasters guide to ravnica Dungeons & Dragons 5...
guildmasters guide to ravnica Dungeons & Dragons 5...guildmasters guide to ravnica Dungeons & Dragons 5...
guildmasters guide to ravnica Dungeons & Dragons 5...
 

Jstorm introduction-0.9.6

  • 1. Company LOGO An Introduction of JStorm LongdaFeng(zhongyan.feng@alibaba-inc.com)
  • 2. Longda Feng Alibaba Agenda Background Basic Concept & Scenarios Why start JStorm? JStorm vs Storm Question and Answer.
  • 3. Who are we? JStorm Team was among one of the earliest that uses Storm in China.  Storm 0.5.1/0.5.4/0.6.0/0.6.2/0.7.0/0.7.1  JStorm 0.7.1/0.9.0/0.9.1/0.9.2/0.9.3/… Our Duties  Application Development  JStorm System Development  JStorm System Operation Longda Feng Alibaba
  • 4. Who are Using JStorm Many small Chinese companies are using JStorm Longda Feng Alibaba
  • 5. How Big? More than 3000 servers More than 3 trillion messages per day Longda Feng Alibaba
  • 6. What is JStorm? JStorm is a distributed programming framework  Similar to Hadoop MapReduce but designed for real-time/in-memory scenarios  Users can build powerful distributed applications from very simple APIs Longda Feng Alibaba
  • 7. What is JStorm? Redesigned Storm in Java.  Proved stable running in huge clusters.  Much faster  Much more powerful Longda Feng Alibaba
  • 8. Basic Conception Pipe-lined data processing Longda Feng Alibaba
  • 9. Advantage 1 Easy learning:  Simple Building Blocks: Topology/Spout/Bolt APIs  Out of Box RPC/Fault-tolerance/Real-time Data Grouping & Combining Longda Feng Alibaba
  • 10. Advantage 2 Excellent Scalability  Horizontally Scalable  DAG-based  Adjustable parallelism of each component Longda Feng Alibaba
  • 11. Stable Guarantees Fault-Tolerance  No Single Point of Failure • Nimbus HA • Any Supervisor can be shutdown  New worker will be spawned and replace the failed one automatically Longda Feng Alibaba
  • 12. Accuracy Acking framework guarantees no lost of data Transaction framework guarantees data accuracy. Longda Feng Alibaba
  • 13. Scenarios Stateless Computation  All data come from Tuple Use Cases:  Log Analysis  Pipe-lined System  Message converter  Statistical Analysis  Real-time Recommendation Algorithm Longda Feng Alibaba
  • 14. Longda Feng Alibaba Why start JStorm Storm community is not as active as we’ve expected  Tailored for enterprise environment  Fixed critical bugs in Storm  Provided professional technical support, improved app development pace.  Reduced operational cost. 
  • 15. How Many Versions? https://github.com/alibaba/JStorm/releases  0.9.6(2014/9/22)  0.9.5.1(2014/9/14)  0.9.5 (2014/8/27)  0.9.4.1 (2014/8/15)  0.9.4(2014/7/18)  0.9.3.1 (2014/5/31)  0.9.3 (2014/5/10)  0.9.2 (2014/4/8)  0.9.1(2014/1/24)  0.9.0(2013/12/30)  0.7.1(2013/4/28) Longda Feng Alibaba
  • 16. JStorm is a superset of Storm The program run in Storm can run in JStorm without changing code Longda Feng Alibaba
  • 17. More stable (1) -- nimbus HA Nimbus HA  Dual-Nimbus HA Longda Feng Alibaba
  • 18. More stable (2) -- RPC Netty supports 2 RPC modes  Async  Sync • Sending speed keeps up with the receiving speed, therefore the data flow is more stable. Longda Feng Alibaba
  • 19. More stable(3) – resource isolation Malicious Worker won’t mess up with others  Supported CPU Isolation with cgroups  Supported Memory Isolation Resources quota can be enforced on each group (before 0.9.5) Longda Feng Alibaba
  • 20. More stable(4) -- Monitor Monitor every component in your Topology  Many more metrics(70+) than storm Supported user-defined metrics Supported user-defined alerts Longda Feng Alibaba
  • 21. More stable (5) – CPU usage Better utilizing CPU resource  Improved disruptor implementation • Drop CPU usage from 300% to 10% when processing queue is full  Avoid CPU spin-waiting • Relocating nextTuple/ack/fail work to a different thread Longda Feng Alibaba
  • 22. More stable(6) -- more catch Add try-catch in any place.  Nimbus/supervisor main thread  Spout/bolt initialization/cleanup  All IO operation, serialization/deserialization  All ZK operation Longda Feng Alibaba
  • 23. More stable(7) -- ZK Reduced unnecessary ZK usage:  Removed useless watcher  Increased ZK heartbeat frequency  Detect failed worker without a full scan of the entire ZK directory Longda Feng Alibaba
  • 24. More stable(8) -- other Improved GC Tuning. Guaranteed that all workers killed after kill command is issued Guaranteed single supervisor/nimbus per instance Avoid excessive use of local ports by Netty client 。。。 Longda Feng Alibaba
  • 25. More powerful scheduler Balancing Tasks with regard of :  CPU  Memory  Net Longda Feng Alibaba
  • 26. CPU assignment By default assign each worker a single CPU slot Application can be configured to utilize more slots Why:  Some task creates extra threads to do other things in Alimama, one CPU slot doesn’t meet requirement Longda Feng Alibaba
  • 27. Memory Usage Default worker memory is 2G Application can be configured to utilize more memory slots Why:  In Alipay Mdrill application, Solr bolt will apply much more memory Longda Feng Alibaba
  • 28. Smarter Balancing With JStorm Scheduler:  Tasks that exchange data heavily tend to be assigned to the same worker to avoid networking cost. Longda Feng Alibaba
  • 29. User Defined Scheduler User define task run one designated worker  User can setting how many CPU slot /memory slot will be used Why:  In Taobao TAE project, some bolts want to run in user defined-nodes Longda Feng Alibaba
  • 30. Task on Different Node Task of one component can be scheduled to run on different nodes Why:  In ALIPAY Mdrill, Solr bolt must run different node Longda Feng Alibaba
  • 31. Task on Single Node All tasks can be scheduled to run on a single node. Why:  In Taobao TLog, there are many small jobs, in order to reduce network cost, all task of one job must run on single node. Longda Feng Alibaba
  • 32. Old Assignment “Last Assignment Policy”  By default , a task will run on the machine it runs previous time Why:  In Alibaba CDO, When restart one application, user wanted to reuse old workers Longda Feng Alibaba
  • 33. Pluginable Be able to run on: Hadoop yarn(more stable than storm) Alibaba Apsara Clould System Alibaba Elastic Resource Pool Longda Feng Alibaba
  • 34. Classloader Resolved application jar-confliction with JStorm Longda Feng Alibaba
  • 35. More convenient UI More useful stats collected and displayed. Browse Worker Log in UI Longda Feng Alibaba
  • 36. Support libjar Don’t need assembly all dependency jars into one jar  Submit libjar with libjar parameter  Support worker.classpath Longda Feng Alibaba
  • 37. Faster 6 Servers (24core/98G) 18 Spout/18 Bolt/18 Acker Longda Feng Alibaba 9280598 10818815 Throughput vs workers 9065965 6819139 5610201 6243680 6830500 5595900 5474180 3379800 12000000 10000000 8000000 6000000 4000000 2000000 0 0 10 20 30 40 50 60 polltuples/10s workers jstorm storm
  • 38. JStorm 41W/S Sending Speed Longda Feng Alibaba
  • 39. Storm 41W/S Sending Speed Longda Feng Alibaba
  • 40. Why Faster Reduce memory-copying by zeroMq Dedicated Deserializing Thread Better Tuned Sampling Logic Better Tuned Acking Framework Better Tuned GC Longda Feng Alibaba
  • 41. Other Improvement  More than 100 improvements  https://github.com/alibaba/JStorm/blob/master/history.md  Fixed assign topology competition  Reset rebalance/reassigned worker timeout as 4 minutes  Graceful worker shutdown  Improvement on thrift server  Avoid mistakenly killing of worker while rebalancing jobs.  。。。。 Longda Feng Alibaba
  • 42. More document https://github.com/alibaba/JStorm/wiki Google-group:JStorm-user@googlegroups.com Wangwang:JStorm QQ:228374502 Laiwang: JStorm Longda Feng Alibaba
  • 43. Join us Welcome to Join us  zhongyan.feng@alibaba-inc.com Longda Feng Alibaba