Transcript SEDS-Lite: Using Open Source Tools (R, BIRT and MySQL)

SEDS-Lite:
Using Open Source Tools
(R, BIRT and MySQL) to Report
and Analyze Performance Data
Igor Trubin, PhD
http://www.itrubin.blogspot.com
SCMG’11
Agenda
o SEDS-Lite project: Open source based implementation of Statistical Exception
Detection System - SEDS-Lite Introduction
o Open Source and “free” tools for Capacity Management – review
o Control Charts as a good test of using different tools for Capacity
management - Power of Control Charts and IT-Chart Concept (Part 1)
o R-system to built Control Charts (System Management by Exception:
o
o
o
o
o
o
Power of Control Charts Mar 25, 2009)
How To Build IT-Control Chart - Use the Excel Pivot Table!
BIRT based Control Chart
BIRT Data Cubes Usage for Performance Data Analysis
Building IT-Control Chart by BIRT against Data from the MySQL
Database
EV-Control Chart (to build using BIRT)
Simplified Exception Detection using BIRT: UCL=LCL : How many
standard deviations do we use for Control Charting? Use ZERO!
SCMG’11
2
SEDS-Lite
Open source based implementation of Statistical Exception
Detection System - SEDS-Lite Introduction
SCMG’11
Open Source and “free” tools for Capacity Management
Previous reviews:
o MeasureIT - Issue 2.05 - Open Source Capacity &
Performance Management Tools for Windows & Unix
Systems (2004 - Brian Johnson )
o Capacity Planning and Performance Monitoring with Free
Tools (for mspcmg 2011! - Adrian Cockcroft)
The missed point is BI and Generic DB open source tools usage
for Performance Analysis and Reporting
o BIRT (Business Intelligence and Reporting Tools)
o MySQL
o NoSQL
SCMG’11
4
Control Charts as a good test of using different tools for
Capacity management
What is the Control Chart?
o Power of Control Charts and IT-Chart Concept (CMG’09 -’10)
SCMG’11
5
R-system to built Control Charts
CMG’09 workshop “ Power of Control Charts “, SCMG - Mar 25, 2009
o EXAMPLE3: Monthly Profile
R download: http://www.r-project.org/
The input data is Unix File
Space Utilization:
R-script
(published
on my
blog):
(FYI: qcc: An R package for
quality control charting :
http://cran.rproject.org/web/packages/qcc/index.html )
SCMG’11
6
R-system to built Control Charts
CMG’09 workshop “ Power of Control Charts “, SCMG - Mar 25, 2009
SCMG’11
7
BIRT based Control Chart
What is BIRT – see BIRT Report Designer Tutorial Demo
There is a tool that generates control charts against performance
data using BIRT:
The IT-Control Chart would give much more value for analyzing time
stamped historical data. Is that possible to build using BIRT?
SCMG’11
8
BIRT based IT-Control Chart: Live demo
The case where the input data source is CSV file:
But how to
calculate that
UCL and LCL
numbers!?
SCMG’11
9
How to calculate that UCL and LCL numbers!?
Use EXCEL Pivot Table
Weekhour calculation:
Pivot Table creation (IT-Chart Builder)
SCMG’11
Pivot Chart built
10
How to calculate that UCL and LCL numbers!?
BIRT Data Cubes Usage for Performance Data Analysis
Data behind IT Control chart is a Cube:
Input data is raw
stamped measures
The result (in
the form of
BIRT report
designer
preview) is on
the picture at
left
SCMG’11
11
How to calculate that UCL and LCL numbers!?
BIRT Data Cubes Usage for Performance Data Analysis
(1) The Reference set or base-line is based on the input raw data with
some filtering and computed columns (weekday and weekhour)
(2) the Actual data set is the same but having the different filter:
(raw[“date”] Greater “2011-04-02”)
SCMG’11
12
How to calculate that UCL and LCL numbers!?
BIRT Data Cubes Usage for Performance Data Analysis
(3) To combine both data
sets for comparing base-line
vs. actual, the “Data Set1” is
built as a “Joint Data Set” by
the following BIRT Query
builder:
(4) Then the Data Cube was
built in the BIRT Data Cube
Builder with the structure
shown on the following
screen:
(5) The next step is building
report starting with Cross
Table (which is picked as an
object from BIRT Report
designer “Pallete”):
The picture above shows also what fields are
SCMG’11
chosen from Cube to Cross table.
13
How to calculate that UCL and LCL numbers!?
BIRT Data Cubes Usage for Performance Data Analysis
(6) The final step is dropping “Chart” object from “Palette” and adding
UCL calculation using Expression Builder for additional Value (Y) Series:
SCMG’11
14
BIRT Data Cubes Usage for Performance Data Analysis
To see the
result one needs
just to run the
report or to use
a "preview' tab
on the
report designer
window:
SCMG’11
15
How to calculate that UCL and LCL numbers!?
SQL script for MySQL Database (upload)
(1) That raw data (the same) need to be uploaded to some table
(CPUutil) in the MySQL schema (ServerMetric) by using the following
script (sqlScriptToUploadCSVforSEDS.sql):
SCMG’11
16
How to calculate that UCL and LCL numbers!?
SQL script for MySQL Database (Calculate)
(2) Then the output (result) data (ActualVsHistoric table) is built using
the following script (sqlScriptToControlChartforSEDS.sql):
SCMG’11
17
Building IT-Control Chart by BIRT
against Data from the MySQL Database
(3) In the BIRT, the connection to MySQL database is established (to
MySQLti with schema ServerMetrics to table ActualVsHistorical):
SCMG’11
18
Building IT-Control Chart by BIRT
against Data from the MySQL Database
(4) the chart is developed the same way like that was done in BIRT
based Control Chart post (slide 14):
SCMG’11
19
Building IT-Control Chart by BIRT
against Data from the MySQL Database
(5) In BIRT you can specify report parameters, that could be then a part
of any constants including for filtering (to change a baseline or to provide
server or metric names).
(6) Finally the
report should be
run to get the
following result,
which is almost
identical with the
one built for BIRT
based Control
Chart post (slide
15)
Lastly: R-system
can be used to run
the same SQL
script against
MySQL by using
Package
'RMySQL‘!!!!
SCMG’11
20
EV-Control Chart (to build using BIRT)
EV meta-metric introduced in 2001 as a measure of anomaly
severity. (The Exception Value Concept to Measure Magnitude of Systems Behavior Anomalies)
In CMG’08 paper “Exception Based Modeling
and Forecasting” That metric was plotted
using Excel to explain how it could be used
for a new trend starting point recognition:
It is the difference
(integral) between actual
data and control limits.
SCMG’11
21
EV-Control Chart (to build using BIRT)
Plotting that meta-metric and/or two their components (EV+ and EV-) over
time gives a valuable picture of system behavior.
The BIRT example of the EV-chart against the same sample data:
EV calculation
SCMG’11
22
EV-Control Chart (to build using BIRT)
The EXCEL example of the EV-chart against the same sample data:
SCMG’11
23
SEDS-Lite: Exception Detection (EV-based) using BIRT
Lite
version:
(UCL=LCL)
Below is example of
DB2-like SQL script
to implement that.
It is based on
real script
developed and
successfully
tested against
real data for
BIRT tool to get
exceptional
servers list:
SCMG’11
24
References
Igor Trubin: “Global
and Application Levels Exception Detection System, Based
on MASF Technique ”, Proceedings of the Computer Measurement Group, 2002.
(http://www.cmg.org/measureit/shared/trubin_02.pdf)
Linwood Merritt, Igor Trubin: “Disk
Subsystem Capacity Management Based on
Business Drivers I/O Performance Metrics and MASF”, Proceedings of the Computer
Measurement Group, 2003.
(http://regions.cmg.org/regions/ncacmg/downloads/june162004_session3.doc)
Linwood Merritt, Igor Trubin: : “Mainframe
Global and Workload Level Statistical
Exception Detection System, Based on MASF”, Proceedings of the Computer
Measurement Group, 2004. (http://www.cmg.org/membersonly/2004/papers/4179.pdf)
Igor Trubin: “Capturing Workload Pathology by Statistical
System”, Proceedings of the Computer Measurement Group, 2005.
(http://www.cmg.org/membersonly/2005/papers/5016.pdf)
Igor Trubin: “System Management by Exception,
Measurement Group, 2006.
(http://www.cmg.org/membersonly/2006/papers/6120.pdf)
Exception Detection
Part 6”, Proceedings of the Computer
Igor Trubin: “System Management by Exception, Part Final”, Proceedings of the
Computer Measurement Group,
2007.(http://regions.cmg.org/regions/scmg/fall_07/richmond/SEDSCMG2007_v4.pdf )
Igor Trubin: “Exception Based Modeling and Forecasting”, Proceedings of the Computer
Measurement Group, 2008. (http://itrubin.blogspot.com/2008/08/exception-based-modeling-and.html )
Igor Trubin: “IT-Control
Chart”, Proceedings of the Computer Measurement Group, 2010.
(http://itrubin.blogspot.com/2010/07/my-new-cmg10-paper-it-control-charts.html )
SCMG’11
25
Questions?
Everything is ready to build
SEDS-Lite:
Using Open Source Tools
(R, BIRT and MySQL)
Igor Trubin, PhD
http://www.itrubin.blogspot.com
SCMG’11
26