17
TECHNICAL WHITE PAPER BMC Control-M Workload Automation Performance Results for Version 8.0.00.100

BMC Control-M Workload Automation Performance Results for Version

  • Upload
    others

  • View
    5

  • Download
    0

Embed Size (px)

Citation preview

Page 1: BMC Control-M Workload Automation Performance Results for Version

TECHNICAL WHITE PAPER

BMC Control-M Workload Automation Performance Results for Version 8.0.00.100

Page 2: BMC Control-M Workload Automation Performance Results for Version

TABLE OF CONTENTS

EXECUTIVE SUMMARY 1

NEW DAY ENHANCEMENTS 2

DISCLAIMER 2

BMC CONTROL-M/SERVER PERFORMANCE TESTS RESULTS 3

» Test Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

» Environment Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

NEW DAY ENHANCEMENTS TEST METHODOLOGY 4

NEW DAY ENHANCEMENT TEST 5

NEW DAY ENHANCEMENTS SUMMARY TESTS RESULTS 5

» New day enhancements test were executed are as follows: . . . . . . . . . . . . . . . 5

ORDER 20K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD ON THE LOCAL DATABASE 6

ORDER 20K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD ON THE MIRROR DATABASE 7

ORDER 20K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD ON A REMOTE DATABASE 8

ORDER 40K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD 9

ORDER 100K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD 14

WHERE TO GET THE LATEST PRODUCT INFORMATION 15

Page 3: BMC Control-M Workload Automation Performance Results for Version

1

INTRODUCTIONBMC Software is informing users and potential users of BMC Control-M Workload Automation version 8.0.00.100

about a series of tests. The test results are relevant to decisions about the platforms required to support various

levels of BMC Control-M production. If you have questions about these tests, contact BMC Software Customer

Support at (800) 537-1813 (United States or Canada) or call your local support center.

EXECUTIVE SUMMARYAs businesses continue to expand their usage of BMC Control-M Workload Automation, it becomes increasingly

important to provide information about the expected performance for this solution. BMC Software conducted

a series of enterprise-scale tests to demonstrate the performance of BMC Control-M Workload Automation.

These tests were conducted in BMC R&D labs prior to releasing version 8.0.00.100.

This white paper provides the following information to assist customers in achieving peak performance:

» Quantitative test results, as a baseline » Guidelines for hardware sizing » System configuration recommendations

Realistic user scenarios and workloads derived from customer cases were used for the tests. The results

demonstrate that by using BMC Workload Automation version 8.0.00.100, organizations can achieve tangible

operational improvements.

The focus of this series of tests is designed to measure the duration of the New Day procedure. The changes

introduced with version 8.0.00.100 have reduced the time required for both the Order and Download processes.

This document describes the dramatic performance improvements observed during New Day processing.

Page 4: BMC Control-M Workload Automation Performance Results for Version

2

NEW DAY ENHANCEMENTSThe New Day procedure is an automated CONTROL-M/Server function. It is performed at the beginning of each

day and prepares the Control-M/Server environment to manage the production workload for the coming day.

Jobs in folders with the order method of “Automatic” are placed into the Active Environment (Ordered) during this

procedure.

During New Day, Control-M/Server housekeeping is also performed. In order to reduce the time necessary for

New Day processing, some customers choose to order some jobs via user dailies (set the folder Order Method to

“Specific User Daily”).

The New Day enhancements described here significantly reduce the time required for processing folders with the

Automatic order method thus enabling customers can reduce or eliminate User Dailies thereby simplifying the

management of their environments.

PRE-ORDERFolders with the Automatic order method are identified and pre-processed in the background in advance of the

New Day procedure. Preprocessing performs in advance much of the work done during New Day. The resulting

data is staged in temporary database tables during pre-order and moved to the Active tables when the actual

New Day procedure is run.

The pre-order is enabled by default when the configuration parameter NEWDAY_MODE is set to a value of 1.

DOWNLOADWhen many jobs exist in the active jobs database the download time is increased. You can reduce the time

required to download all the required data to the Control-M/Server by enabling Parallel Download.

To enable Parallel Download, set the following parameter in config.dat file:

DOWNLOAD_THREADS (the number of CPU - 1) (default is 3)

DISCLAIMERBMC Software is providing these test results to help users and potential users of BMC Control-M Workload

Automation to determine the resources needed for using this product. BMC Software intends to provide

meaningful and accurate information, but does not guarantee the information is relevant or appropriate to your

computing environment. The results of the tests are based on specific job attributes that may not reflect your

specific scenario.

BMC Software does not make any warranty, expressed or implied, with respect to the information in this

document, nor guarantees the accuracy, completeness, usefulness, or adequacy of this information. BMC

Software will not be liable for any damages, including special, indirect, or consequential damages, arising out of

or in connection with the use of the information provided in this document.

Page 5: BMC Control-M Workload Automation Performance Results for Version

3

BMC CONTROL-M/SERVER PERFORMANCE TESTS RESULTSBMC Control-M/Server sizing tests were executed in different permutations and volumes. The results were compared between

version 8.0.00 and 8.0.00.100 using various application configurations, such as SSL and database modes.

TEST ENVIRONMENTBoth BMC Control-M/Server and BMC Control-M/Agent version 8.0.00.100 were installed on the same host. BMC Control-M/Server

was tested in the following UNIX® and MS Windows environments:

Server OS SPEC Server Model # Cores # Chips # Cores Per Chips RAM

AIX® 7.1 400 IBM®, 8202-E4B PowerPC (Power7, 3 GHz) 12 3 4 30 GB

HP-UX Itanium B.11.31

102HP Integrity RX6600 (Intel Itanium 2 9100, 1.6 GHz)

8 4 2 16 GB

SUN 5.10 179Solaris Enterprise M4000 (SPARC64VII+, 2.66 GHz)

16 4 4 32 GB

Windows Server 2008 R2 64BIT

200Dell PowerEdge R710 (Intel Xeon E5530, 2.4 GHz)

16 2 8 48 GB

Linux® RedHat 6.1 64BIT

317Dell PowerEdge R710 (Intel Xeon E5645, 2 GHz)

24 2 12 48 GB

NOTE: SPEC published at:

SPEC CPU2600 CINT_rate http://www.spec.org

ENVIRONMENT CONFIGURATION » BMC Control-M/Server and a dedicated database were installed on the same computer using local disk storage. » No other applications ran on the computer during the tests. » There were no firewalls in any of the environments.

Page 6: BMC Control-M Workload Automation Performance Results for Version

4

NEW DAY ENHANCEMENTS TEST METHODOLOGY

New

Day

Enh

ance

men

t Tes

t

ITEM DETAILS

Test Type New Day Enhancements (Pre-Order and Parallel Download)

Capacity 20K, 40K and 100K jobs

Data Composition

Each unit contains a chain of 100 smart folders; each smart folder contains 10 jobs (1100 jobs in total). For job dependency and resource utilization details, see below.

Unit characteristics: » Each unit contains 100 smart folders (to simulate the effect of folder scheduling). » In each unit, no more than 10 jobs ran simultaneously. » Each smart folder contains 10 relatively trivial jobs to be scheduled and executed. For example, there

were no cyclic jobs and no periodic calendars. » Each job contains 10 AutoEdit variables. » Each unit contains 10,000 AutoEdit variables. » Each unit contains DO MAIL for 5 percent of the jobs. » Each unit contains LATE TIME SHOUT for 20 percent of the jobs. » Each unit contains LATE SUB SHOUT for 20 percent of the jobs. » Each job contains 1 DO SHOUT message. » Each unit contains 1000 DO SHOUT message. » Each unit contains 2 control resources were used in shared mode. » In each smart folder, 4 of the 10 jobs used the same control resource in shared mode. » 1 quantitative resource was used for each smart folder (100 per unit). » The first job in each folder used the same quantitative resource (quantity = 1). So, the first job in each

folder could not start while another folder’s first job was running. » Each unit contains 5000 in conditions. » Each unit contains 1000 out conditions. » Each job submitted an echo command.

Product BMC Control-M Workload Automation products: » BMC Control-M/Enterprise Manager server » BMC Control-M/Server » BMC Control-M/Agent

Deployment Consolidate on a single machine

Tested Components

BMC Control-M/Server was installed on individual computers and Control-M/Agents were installed on multiple individual machines.

BMC Control-M/Server configured with local, mirror and remote databases. » Database vendors were:

– Oracle® 11g – PostgreSQL 8.3.7 – MS SQL Server 2008 – Sybase 15.5

» No other applications ran on that computer during the tests. » BMC Control-M/Enterprise Manager ran on a different computer and used a different database. » There were no firewalls in any of the environments. » All of the BMC Control-M/Agents ran on computers other than the host for BMC Control-M/Server.

System Parameters

Most system parameters were set to default values that maximized throughput: » BMC Control-M/Server clean statistics were were not tested. » The debug level was set to default. » The following system parameters were set for all tests:

– NEWDAY_MODE 1 (default) – SEL_IO_THREADS 8 – POST_PROCESS_THREAD_POOL_SIZE 6

Test Goals » Approve the product has no regression from a performance aspect » Approve the product has improvements in the order time and download time » Determine database capacity influence » Quantitative test results, as a baseline » System configuration recommendations

Page 7: BMC Control-M Workload Automation Performance Results for Version

5

NEW DAY ENHANCEMENTS SUMMARY TESTS RESULTSNew day enhancements were tested in with several jobs volumes, OS platforms and Databases.

Test scenario includes pre-order and parallel download to Control-M/EM.

New day enhancements test results showed significant improvement in New Day procedure speed in every tested

component.

Pre-order test results showed reduced time ordering jobs with the Automatic order method by preparing required jobs in

advance. The actual “Order” process was reduce to seconds.

Download test results showed reduced time downloading to Control-M/Enterprise Manager by performing the process

in parallel . Even greater improvement can be realized by increasing the number of threads used concurrently, set via

system parameter “DOWNLOAD_THREADS”.

BMC Control-M/Server consumed ~30% more database space when the pre-order option was enabled.

NEW DAY ENHANCEMENTS TEST WERE EXECUTED ARE AS FOLLOWS: » Order 20K jobs in different modes: pre-order and parallel download on the local database » Order 20K jobs in different modes: pre-order and parallel download on the mirror database » Order 20K jobs in different modes: pre-order and parallel download on a remote database » Order 40K jobs in different modes: pre-order and parallel download » Order 100K jobs in different modes: pre-order and parallel download

Page 8: BMC Control-M Workload Automation Performance Results for Version

6

ORDER 20K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD ON THE LOCAL DATABASE

1. BMC Control-M/Server installed on Solaris with Oracle 11 g (11.2) database.

Figure 1:

Figure 2:

Page 9: BMC Control-M Workload Automation Performance Results for Version

7

ORDER 20K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD ON THE MIRROR DATA-BASE

2. BMC Control-M/Server installed on Solaris with local Oracle 11 g (11.2) database and mirror Oracle 11 g (11.2) database

installed on Windows Server 2008 R2.

Figure 3:

Figure 4:

» Network latency between local and mirror databases is up to 4 ms.

Page 10: BMC Control-M Workload Automation Performance Results for Version

8

ORDER 20K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD ON A REMOTE DATA-BASE

3. BMC Control-M/Server installed on Solaris with local Oracle 11 g (11.2) database and remote Oracle 11 g (11.2)

database installed on Windows Server 2008 R2.

Figure 5:

Figure 6:

» Network latency between local and remote databases is up to 4 ms.

Page 11: BMC Control-M Workload Automation Performance Results for Version

9

ORDER 40K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD

4. BMC Control-M/Server installed on Solaris with local Oracle 11 g (11.2) database.

Figure 7:

Figure 8:

Page 12: BMC Control-M Workload Automation Performance Results for Version

1 0

5. BMC Control-M/Server installed on Windows with local MS SQL 2008 database.

Figure 9:

Figure 10:

Page 13: BMC Control-M Workload Automation Performance Results for Version

1 1

6. BMC Control-M/Server installed on HP-UX with local PostgreSQL 8.3.7 database.

Figure 11:

Figure 12:

Page 14: BMC Control-M Workload Automation Performance Results for Version

1 2

7. BMC Control-M/Server installed on Linux Red Hat with local PostgreSQL 8.3.7 database.

Figure 13:

Figure 14:

Page 15: BMC Control-M Workload Automation Performance Results for Version

1 3

8. BMC Control-M/Server installed on AIX with remote Sybase 15 database.

Figure 14:

Figure 15:

Page 16: BMC Control-M Workload Automation Performance Results for Version

1 4

ORDER 100K JOBS IN DIFFERENT MODES: PRE-ORDER AND PARALLEL DOWNLOAD

9. BMC Control-M/Server installed on AIX with local PostgreSQL 8.3.7 database.

Figure 16:

Figure 17:

Page 17: BMC Control-M Workload Automation Performance Results for Version

BMC, BMC Software, and the BMC Software logo are the exclusive properties of BMC Software, Inc , Are registered with the U S Patent and Trademark Office, and may be registered or pending registration in other countries All other BMC trademarks, service marks, and logos may be registered or pending registration in the U S or in other countries AIX and IBM are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both UNIX is the registered trademark of The Open Group in the US and other countries Linux is the registered trademark of Linus Torvalds Oracle and Java are registered trademarks of Oracle and/or its affiliates Other names may be trademarks of their respective owners All other trademarks or registered trademarks are the property of their respective owners © 2013 BMC Software, Inc All rights reserved Origin date 05/2013

*443132*

WHERE TO GET THE LATEST PRODUCT INFORMATIONTo view the latest BMC Software documents, visit the BMC Customer Support page at http://www.bmc.com/support_

home . BMC Software distributes printed copies of flashes, technical bulletins, and release notes with most product

shipments, as indicated on your shipping list. In addition, all notices are available on the Customer Support page,

including any notices that BMC Software issues after you receive your product shipment. You will not receive new

notices by mail. However, by subscribing to proactive notification, you can receive e-mail messages that direct you to

those notices. For more information about proactive notification, refer to the Customer Support page.

BUSINESS RUNS ON IT. IT RUNS ON BMC SOFTWARE.Business runs better when IT runs at its best. Tens of thousands of IT organizations around the world -- from small

and mid-market businesses to the Global 100 -- rely on BMC Software (NASDAQ: BMC) to manage their business

services and applications across distributed, mainframe, virtual and cloud environments. BMC helps customers cut

costs, reduce risk and achieve business objectives with the broadest choice of IT management solutions, including

industry-leading Business Service Management and Cloud Management offerings. For the four fiscal quarters ended

March 31, 2013, BMC revenue was approximately $2.2 billion.