A Framework forApplication Performance Monitoring

and Dynamic Software Analysis—Preview for Invited Demo/Poster Presentation—

André van Hoorn, J. Waller, and W. Hasselbring

Software Engineering GroupUniversity of Kiel, Germany

April 24, 2012 @ ICPE ’12, Boston, MA

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 1 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Use cases in research and practice:

• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Use cases in research and practice:

• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Analysis configuration (Web GUI)

Analysis

Plugins

Use cases in research and practice:

• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Analysis configuration (Web GUI)

Analysis

Plugins

Invo

catio

ns/m

inut

e [x

100

0]

Calendar time (hh:mm)

Workload Anomaly Detection

0.0

0.5

1.0

1.5

2.0

Ano

mal

y sc

ore

0.0

0.25

0.50

0.75

SRV0::

@3:..Bookstore

SRV0::

@1:..Catalog

SRV0::

@2:..CRM

SRV1::

@1:..Catalog

searchBook()

getBook(..)

getOffers()

getBook(..)

<<execution container>>SRV0<<deployment component>>@3:..Bookstore <<deployment component>>@2:..CRM

<<deployment component>>@1:..Catalog

<<execution container>>SRV1<<deployment component>>@1:..Catalog

searchBook()

getOffers()

1635

getBook(..)

543

getBook(..)

1092

573

1062

$ 1635

Visualizations

Use cases in research and practice:

• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Analysis configuration (Web GUI)

Analysis

Plugins

Invo

catio

ns/m

inut

e [x

100

0]

Calendar time (hh:mm)

Workload Anomaly Detection

0.0

0.5

1.0

1.5

2.0

Ano

mal

y sc

ore

0.0

0.25

0.50

0.75

SRV0::

@3:..Bookstore

SRV0::

@1:..Catalog

SRV0::

@2:..CRM

SRV1::

@1:..Catalog

searchBook()

getBook(..)

getOffers()

getBook(..)

<<execution container>>SRV0<<deployment component>>@3:..Bookstore <<deployment component>>@2:..CRM

<<deployment component>>@1:..Catalog

<<execution container>>SRV1<<deployment component>>@1:..Catalog

searchBook()

getOffers()

1635

getBook(..)

543

getBook(..)

1092

573

1062

$ 1635

Visualizations

Use cases in research and practice:

• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Analysis configuration (Web GUI)

Analysis

Plugins

Invo

catio

ns/m

inut

e [x

100

0]

Calendar time (hh:mm)

Workload Anomaly Detection

0.0

0.5

1.0

1.5

2.0

Ano

mal

y sc

ore

0.0

0.25

0.50

0.75

SRV0::

@3:..Bookstore

SRV0::

@1:..Catalog

SRV0::

@2:..CRM

SRV1::

@1:..Catalog

searchBook()

getBook(..)

getOffers()

getBook(..)

<<execution container>>SRV0<<deployment component>>@3:..Bookstore <<deployment component>>@2:..CRM

<<deployment component>>@1:..Catalog

<<execution container>>SRV1<<deployment component>>@1:..Catalog

searchBook()

getOffers()

1635

getBook(..)

543

getBook(..)

1092

573

1062

$ 1635

Visualizations

Use cases in research and practice:• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Analysis configuration (Web GUI)

Analysis

Plugins

Invo

catio

ns/m

inut

e [x

100

0]

Calendar time (hh:mm)

Workload Anomaly Detection

0.0

0.5

1.0

1.5

2.0

Ano

mal

y sc

ore

0.0

0.25

0.50

0.75

SRV0::

@3:..Bookstore

SRV0::

@1:..Catalog

SRV0::

@2:..CRM

SRV1::

@1:..Catalog

searchBook()

getBook(..)

getOffers()

getBook(..)

<<execution container>>SRV0<<deployment component>>@3:..Bookstore <<deployment component>>@2:..CRM

<<deployment component>>@1:..Catalog

<<execution container>>SRV1<<deployment component>>@1:..Catalog

searchBook()

getOffers()

1635

getBook(..)

543

getBook(..)

1092

573

1062

$ 1635

Visualizations

Use cases in research and practice:• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage

• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Analysis configuration (Web GUI)

Analysis

Plugins

Invo

catio

ns/m

inut

e [x

100

0]

Calendar time (hh:mm)

Workload Anomaly Detection

0.0

0.5

1.0

1.5

2.0

Ano

mal

y sc

ore

0.0

0.25

0.50

0.75

SRV0::

@3:..Bookstore

SRV0::

@1:..Catalog

SRV0::

@2:..CRM

SRV1::

@1:..Catalog

searchBook()

getBook(..)

getOffers()

getBook(..)

<<execution container>>SRV0<<deployment component>>@3:..Bookstore <<deployment component>>@2:..CRM

<<deployment component>>@1:..Catalog

<<execution container>>SRV1<<deployment component>>@1:..Catalog

searchBook()

getOffers()

1635

getBook(..)

543

getBook(..)

1092

573

1062

$ 1635

Visualizations

Use cases in research and practice:• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis

• (Self-)adaptation control• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Analysis configuration (Web GUI)

Analysis

Plugins

Invo

catio

ns/m

inut

e [x

100

0]

Calendar time (hh:mm)

Workload Anomaly Detection

0.0

0.5

1.0

1.5

2.0

Ano

mal

y sc

ore

0.0

0.25

0.50

0.75

SRV0::

@3:..Bookstore

SRV0::

@1:..Catalog

SRV0::

@2:..CRM

SRV1::

@1:..Catalog

searchBook()

getBook(..)

getOffers()

getBook(..)

<<execution container>>SRV0<<deployment component>>@3:..Bookstore <<deployment component>>@2:..CRM

<<deployment component>>@1:..Catalog

<<execution container>>SRV1<<deployment component>>@1:..Catalog

searchBook()

getOffers()

1635

getBook(..)

543

getBook(..)

1092

573

1062

$ 1635

Visualizations

Use cases in research and practice:• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Analysis configuration (Web GUI)

Analysis

Plugins

Invo

catio

ns/m

inut

e [x

100

0]

Calendar time (hh:mm)

Workload Anomaly Detection

0.0

0.5

1.0

1.5

2.0

Ano

mal

y sc

ore

0.0

0.25

0.50

0.75

SRV0::

@3:..Bookstore

SRV0::

@1:..Catalog

SRV0::

@2:..CRM

SRV1::

@1:..Catalog

searchBook()

getBook(..)

getOffers()

getBook(..)

<<execution container>>SRV0<<deployment component>>@3:..Bookstore <<deployment component>>@2:..CRM

<<deployment component>>@1:..Catalog

<<execution container>>SRV1<<deployment component>>@1:..Catalog

searchBook()

getOffers()

1635

getBook(..)

543

getBook(..)

1092

573

1062

$ 1635

Visualizations

Use cases in research and practice:• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations

• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Kieker: Example Workflow and Use CasesPreview for Invited Demo/Poster Presentation

Software system with monitoring instrumentation

Monitoring probe

Measurement

Monitoring records

Monitoring log/stream

Analysis configuration (Web GUI)

Analysis

Plugins

Invo

catio

ns/m

inut

e [x

100

0]

Calendar time (hh:mm)

Workload Anomaly Detection

0.0

0.5

1.0

1.5

2.0

Ano

mal

y sc

ore

0.0

0.25

0.50

0.75

SRV0::

@3:..Bookstore

SRV0::

@1:..Catalog

SRV0::

@2:..CRM

SRV1::

@1:..Catalog

searchBook()

getBook(..)

getOffers()

getBook(..)

<<execution container>>SRV0<<deployment component>>@3:..Bookstore <<deployment component>>@2:..CRM

<<deployment component>>@1:..Catalog

<<execution container>>SRV1<<deployment component>>@1:..Catalog

searchBook()

getOffers()

1635

getBook(..)

543

getBook(..)

1092

573

1062

$ 1635

Visualizations

Use cases in research and practice:• Online/offline performance evaluation and feedback, e.g.,

• Continuous monitoring of application behavior and usage• Performance anomaly detection and diagnosis• (Self-)adaptation control

• Extraction of software architectural (performance) models and visualizations• Simulation (replaying previously monitored stimuli; measurement, logging, and analysis)

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 2 / 3

Characteristics, Features, Extension PointsPreview for Invited Demo/Poster Presentation

Kieker Framework

• Modular, flexible, and extensible architecture(Probes, records, readers, writers, filters etc.)

• Pipes-and-filters framework for analysis configuration• Distributed tracing (logging, reconstruction, visualization)

• Low overhead (designed for continuous operation)

• Evaluated in lab and industrial case studies (since 2006)

Kieker is open-source software (Apache License, V. 2.0)

http://kieker-monitoring.net

Kieker is distributed as part of SPEC® RG'srepository of peer-reviewed tools for

quantitative system evaluation and analysishttp://research.spec.org/projects/tools.html

Kieker.Monitoring

Periodic

Sam

plingJM

X

Interface

Logging

Tim

e S

ource

Monitoring Controller

Logging

Seria-lization

Deseria-lization

Kieker.Analysis

Analysis Controller

Mo

nito

ring

Reco

rds Current time

Resource utilization

CPU utilization

Operation execution

Control-flow events

<your monitoring record type>

Mo

nito

ring

Writers

Mo

nito

ring

Pro

bes/S

am

plers

Tim

e So

urces

Database (SQL)

Named pipe

<your monitoring writer>

Asynchr. w

ritersS

yncr.

System time

<your time source>

Control-flow

tracing SpringAspectJ

Servlet CXF/SOAP

Manual instrumentation

<your interception technology>

Resource

monitoring

<your technology>

CPU utilization

Memory usage

Sigar

Servlet

<your monitoring probe>

File system

Database (SQL)

Java Messaging Service (JMS)

Java Management Ext. (JMX)

File system

<your monitoring reader>

File system

Database (SQL)

Mo

nito

ring

Read

ers

<your visualization>

Sequence diagrams

Dependency graphs

Call graphs

Visualization

Architecture reconstr.

Pipe-and-filter fram

ewo

rk

An

alysis/Visu

alization

Plu

gin

s

<your trace analysis>

Trace analysis

<your reconstruction plugin>

<your analysis plugin/tool>

Java Messaging Service (JMS)

Java Management Ext. (JMX)

Output stream

Memory/swap usage

Real-time replayerNamed pipe

Monitoring Record

Monitoring Reader

Monitoring Reader

Monitoring Reader

Monitoring Writer

MonitoringLog/Stream

Monitoring Probe

Monitoring Probe

Monitoring Probe

Analysis /Visualization

Plugin

Analysis /Visualization

Plugin

Analysis /Visualization

Plugin

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 3 / 3

Characteristics, Features, Extension PointsPreview for Invited Demo/Poster Presentation

Kieker Framework

• Modular, flexible, and extensible architecture(Probes, records, readers, writers, filters etc.)

• Pipes-and-filters framework for analysis configuration• Distributed tracing (logging, reconstruction, visualization)

• Low overhead (designed for continuous operation)

• Evaluated in lab and industrial case studies (since 2006)

Kieker is open-source software (Apache License, V. 2.0)

http://kieker-monitoring.net

Kieker is distributed as part of SPEC® RG'srepository of peer-reviewed tools for

quantitative system evaluation and analysishttp://research.spec.org/projects/tools.html

Kieker.Monitoring

Periodic

Sam

plingJM

X

Interface

Logging

Tim

e S

ource

Monitoring Controller

Logging

Seria-lization

Deseria-lization

Kieker.Analysis

Analysis Controller

Mo

nito

ring

Reco

rds Current time

Resource utilization

CPU utilization

Operation execution

Control-flow events

<your monitoring record type>

Mo

nito

ring

Writers

Mo

nito

ring

Pro

bes/S

am

plers

Tim

e So

urces

Database (SQL)

Named pipe

<your monitoring writer>

Asynchr. w

ritersS

yncr.

System time

<your time source>

Control-flow

tracing SpringAspectJ

Servlet CXF/SOAP

Manual instrumentation

<your interception technology>

Resource

monitoring

<your technology>

CPU utilization

Memory usage

Sigar

Servlet

<your monitoring probe>

File system

Database (SQL)

Java Messaging Service (JMS)

Java Management Ext. (JMX)

File system

<your monitoring reader>

File system

Database (SQL)

Mo

nito

ring

Read

ers

<your visualization>

Sequence diagrams

Dependency graphs

Call graphs

Visualization

Architecture reconstr.

Pipe-and-filter fram

ewo

rk

An

alysis/Visu

alization

Plu

gin

s

<your trace analysis>

Trace analysis

<your reconstruction plugin>

<your analysis plugin/tool>

Java Messaging Service (JMS)

Java Management Ext. (JMX)

Output stream

Memory/swap usage

Real-time replayerNamed pipe

Monitoring Record

Monitoring Reader

Monitoring Reader

Monitoring Reader

Monitoring Writer

MonitoringLog/Stream

Monitoring Probe

Monitoring Probe

Monitoring Probe

Analysis /Visualization

Plugin

Analysis /Visualization

Plugin

Analysis /Visualization

Plugin

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 3 / 3

Characteristics, Features, Extension PointsPreview for Invited Demo/Poster Presentation

Kieker Framework

• Modular, flexible, and extensible architecture(Probes, records, readers, writers, filters etc.)

• Pipes-and-filters framework for analysis configuration• Distributed tracing (logging, reconstruction, visualization)

• Low overhead (designed for continuous operation)

• Evaluated in lab and industrial case studies (since 2006)

Kieker is open-source software (Apache License, V. 2.0)

http://kieker-monitoring.net

Kieker is distributed as part of SPEC® RG'srepository of peer-reviewed tools for

quantitative system evaluation and analysishttp://research.spec.org/projects/tools.html

Kieker.Monitoring

Periodic

Sam

plingJM

X

Interface

Logging

Tim

e S

ource

Monitoring Controller

Logging

Seria-lization

Deseria-lization

Kieker.Analysis

Analysis Controller

Mo

nito

ring

Reco

rds Current time

Resource utilization

CPU utilization

Operation execution

Control-flow events

<your monitoring record type>

Mo

nito

ring

Writers

Mo

nito

ring

Pro

bes/S

am

plers

Tim

e So

urces

Database (SQL)

Named pipe

<your monitoring writer>

Asynchr. w

ritersS

yncr.

System time

<your time source>

Control-flow

tracing SpringAspectJ

Servlet CXF/SOAP

Manual instrumentation

<your interception technology>

Resource

monitoring

<your technology>

CPU utilization

Memory usage

Sigar

Servlet

<your monitoring probe>

File system

Database (SQL)

Java Messaging Service (JMS)

Java Management Ext. (JMX)

File system

<your monitoring reader>

File system

Database (SQL)

Mo

nito

ring

Read

ers

<your visualization>

Sequence diagrams

Dependency graphs

Call graphs

Visualization

Architecture reconstr.

Pipe-and-filter fram

ewo

rk

An

alysis/Visu

alization

Plu

gin

s

<your trace analysis>

Trace analysis

<your reconstruction plugin>

<your analysis plugin/tool>

Java Messaging Service (JMS)

Java Management Ext. (JMX)

Output stream

Memory/swap usage

Real-time replayerNamed pipe

Monitoring Record

Monitoring Reader

Monitoring Reader

Monitoring Reader

Monitoring Writer

MonitoringLog/Stream

Monitoring Probe

Monitoring Probe

Monitoring Probe

Analysis /Visualization

Plugin

Analysis /Visualization

Plugin

Analysis /Visualization

Plugin

A. van Hoorn, J. Waller, W. Hasselbring Kieker (ICPE ’12 Poster/Demo Preview) April 24, 2012 @ ICPE ’12, Boston, MA 3 / 3