Migration guide for Rational DOORS Next Generation 6.0

Authors: GustafSvensson, VaughnRokosz
Date: May 29, 2015
Build basis: Rational Doors Next Generation (RDNG) 6.0

Introduction

This document summarizes the results of our performance testing of the upgrade to IBM® Rational® DOORS® Next Generation (RDNG) 6.0. We look at the time required to upgrade from RDNG 5.0.2 to 6.0 in a variety of configurations, and identify the factors that most influence upgrade time, including:

• the total number of artifacts in the repository
• the number of comments and links
• the impact of baselines in the 5.0.x data
• the network latency to the database server
• using Oracle vs. DB2

We also look at how the size of the database is impacted by the upgrade.

Upgrade times in your environments may be different that what we observed. Please be sure to review our standard disclaimer.

Summary of test results

General considerations

There are two general considerations for upgrade performance:

• Keeping the database statistics updated
• Adjusting the heap size for the repotools-rm command

Database statistics

Because there are significant changes to the schema of the DNG database for the 6.0 release, it is important to be sure that the database statistics are kept up to date. Some of the new DNG tables will start out empty and then grow during upgrade. This change in table size can degrade performance of the SQL queries that are executed during upgrade if your database is not configured to update statistics automatically.

If you update your database statistics manually, you may need to update the statistics during the upgrade process. You can use the following commands:

DB2

DB2 REORGCHK UPDATE STATISTICS ON TABLE ALL 

Oracle

EXEC DBMS_STATS.gather_schema_stats('RMSCHEMA');

(where RMSCHEMA is the name of the RM database).

Repotools JVM heap size

For larger repositories (> 100K artifacts), the default heap size in the repotools-rm command is too small. The upgrade will fail unless you raise the heap size from the default of 1.5G to something closer to 8GB (up to a maximum of 24G) Our tests also suggest that the 8G heap size can improve upgrade performance by 20%, for smaller repositories.

The repotools script was updated with the following line; VMARGS="$VMARGS -Xmx8G"

Impact of repository size and shape

The main factor that impacts upgrade time is the total number of artifacts of the following types: requirements, comments, and links. It takes roughly 5 minutes per 100K artifacts to upgrade with no baselines present. In our largest test with 4.9 million total requirements, comments, and links, the migration time for those artifacts was 4 hours.

A secondary factor that impacts upgrade time is the number of artifacts with comments. There is an additional upgrade step which deals with comments. In our largest test, we migrated 1.6 million comments associated with 551,000 artifacts and this took 71 minutes. The total upgrade time for this configuration was 5 hours and 26 minutes.

Baselines and reviews will add additional cost. There are several factors that determine the time it takes to migrate baselines and reviews. The number of artifacts included in baselines and reviews do not impact the migration time but the number of artifacts with revisions do. Each baseline and review also adds extra processing time.

When we upgraded a repository with 2 baselines per module (where each baseline included 10% of the module artifacts), the time to upgrade doubled.

Database growth during upgrade

You can expect the DNG database to grow at least 10% during the upgrade process. The 6.0 release introduces several major new features, and these features involve changes to the database schema. In our testing the database size grew with 0.4GB for every 100,000 artifacts. In our most extreme result (with almost 5 million artifacts), the size of the DNG database nearly doubled (from 22G to 40G). Customer data shows this ratio can vary.

Impact of network latency to the database server

We looked at the impact of increasing the network latency between the system where we ran the upgrade to the database server. There is a linear increase in total upgrade times as the network latency increases, and the upgrade times double for each additional 2 milliseconds of network latency.

For best upgrade performance, be sure that the network connection to the database service is fast.

Comparing Oracle and DB2

We compared the upgrade times for two systems: one using DB2 and one using Oracle. There was no significant difference in upgrade times.

Appendix A: Standard disclaimer

The information in this document is distributed AS IS. The use of this information or the implementation of any of these techniques is a customer responsibility and depends on the customer’s ability to evaluate and integrate them into the customer’s operational environment. While each item may have been reviewed by IBM for accuracy in a specific situation, there is no guarantee that the same or similar results will be obtained elsewhere. Customers attempting to adapt these techniques to their own environments do so at their own risk. Any pointers in this publication to external Web sites are provided for convenience only and do not in any manner serve as an endorsement of these Web sites. Any performance data contained in this document was determined in a controlled environment, and therefore, the results that may be obtained in other operating environments may vary significantly. Users of this document should verify the applicable data for their specific environment.

Performance is based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual throughput or performance that any user will experience will vary depending upon many factors, including considerations such as the amount of multi-programming in the user’s job stream, the I/O configuration, the storage configuration, and the workload processed. Therefore, no assurance can be given that an individual user will achieve results similar to those stated here.

This testing was done as a way to compare and characterize the differences in performance between different versions of the product. The results shown here should thus be looked at as a comparison of the contrasting performance between different versions, and not as an absolute benchmark of performance.

Appendix B: Summary of all test runs

The test results

Repository	Database vendor	Total migration time	Artifacts with comments / time to migrate	Time to migrate 10,000 artifacts with comments	Total number of artifacts	Time to migrate 100,000 artifacts	Database size before/after migration	Database size change GB/%	Database size increase for 100,000 artifacts	Latency
1 project shape 1	DB2 10.1	40 minutes	85,930 / 11 minutes	1 minute 17 seconds	823,441 versions took 27 minutes	3 minutes 17 seconds	4.9GB / 7.98GB	3.08GB / 63%	0.374GB	0.2 ms
6 projects shape 1	DB2 10.1	5 hours 26 minutes	550,580 / 71 minutes	1 minuted 17 seconds	4,940,678 versions took 4 Hours	4 minutes 51 seconds	22.24GB / 40.28GB	18.04GB / 81%	0.365GB	0.2 ms
1 project shape 2	DB2 10.1	1 hour 15 minutes	80,625 / 13 minutes	1 minute 37 seconds	634,101 versions took 1 hour 7 minutes	9 minutes 28 sec	9.61GB / 11.95GB	2.34GB / 24%	0.369GB	0.2 ms
3 projects shape 2	DB2 10.1	3 hours 38 minutes	170,555 / 28 minutes	1 minute 39 sec	1,483,437 took 2 hours 38 minutes	10 minutes 39 seconds	23.19GB / 28.55GB	5.36GB / 23%	0.361GB	0.2 ms
Comparing DB2 and Oracle
1 project shape 2	DB2 10.1	2 hours 6 minutes	65,555 / 13 minutes	1 minute 59 seconds	552,106 took 1 hours 45 minutes	19 minutes	8.95GB / 11.13GB	2.18GB / 24%	0.395GB	0.2 ms
1 project shape 2	Oracle 12c	1 hour 47 minutes	69,889/ 8 minutes 41 seconds	1 minute 15 seconds	704,488 took 1 hours 36 minutes	13 minutes 36 seconds	8.94GB / 10.82GB	1.88GB / 21%	0.267GB	0.2 ms
Latency tests
1 project shape 2	DB2 10.1	1 hour 21 minutes	80,625 / 4 minutes 45 seconds	35 seconds	633,750 versions took 1 hour 15 minutes	11 minutes 50 seconds	10.18GB / 12.22GB	2.04GB / 20%	0.322GB	0.2 ms
1 project shape 2	DB2 10.1	2 hours 43 minutes	80,625 / 9 minutes 18 seconds	1 minute 12 seconds	633,750 versions took 2 hours 31 minutes	23 minutes 50 seconds	10.18GB / 12.22GB	2.04GB / 20%	0.322GB	2 ms
1 project shape 2	DB2 10.1	4 hours 8 minutes	80,625/ 14 minutes 58 seconds	1 minute 51 seconds	633,750 versions took 3 hours 51 minutes	36 minutes 27 seconds	10.18GB / 12.22GB	2.04GB / 20%	0.322GB	4 ms
Migrations done using customer data
Customer A - 56 projects	DB2 10.1	48 minutes	9,385 / 1 minute	1 minute	100,357 versions took 44 minutes	43 minutes 51 sec	17.6GB / 19.1GB	1.5GB / 9%	1.494GB	0.22 ms
Customer B - 9 projects	DB2 10.1	46 minutes	0 comments	Not applicable	513,371 versions took 42 minutes	8 minutes 11 sec	8.6GB / 9.2GB	0.6GB / 7%	0.117GB	0.22 ms
Customer C - 119 projects	DB2 10.1	10 hours 18 minutes	6394 / 2 minutes 57 seconds	4 minutes 37 seconds	810,359 versions took 10 hours 7 minutes	1 hour 16 minutes	26.18GB / 29.5GB	3.32GB / 7%	0.41GB	0.22 ms

Data volume and shape

For our migration tests we used the following project types

Artifact type	Project shape 1	Project shape 2	Customer A	Customer B	Customer C
Modules	52	52	161	452	1880
Folders	119	119	3743	774	15,612
Requirement artifacts	1,118	86,171	19,659	40,005	122,434
Module artifacts	85,000	85,000	5,201	123,792	37,896
Comments	260,582	244,132	9,385	0	21,395
Links	304,029	151,195	20,425	93,174	404,548
Collections	14	14	285	37	1,848
Tags	300	350	546	10	2,220
Views	100	200	1026	23	2,676
Terms	238	238	521	601	1,683
Reviews	0	0	342	0	3,884
Baselines	0	115	57	0	1,166

Appendix C: Topology

The topology that was used in our testing is based on the standard (E1) Enterprise - Distributed / Linux / DB2 topology

Key:

• RM Server: Rational DOORS Next Generation Server
• JTS: Jazz™ Team Server

All server machines and test clients are located on the same subnet. The LAN has 1000 Mbps of maximum bandwidth and 0.2 ms average latency in ping.

Tested hardware and software configurations

The specifications of machines used for most of the upgrade tests are listed below:

Function	Machine Type	CPU / Machine	Total # of CPU Cores/Machine	Memory/Machine	Disk	Disk capacity	Network interface	OS and Version
RDNG Server	IBM System x3550 M4	2 x Intel Xeon E5-2640 2.5GHz (six-core)	24	32GB	RAID 5 -- SAS Disk x 4	279GB	Gigabit Ethernet	Red Hat Enterprise Linux Server release 6.3 (Santiago)
Database Server	IBM System x3650 M4	2 x Intel Xeon E5-2640 2.5GHz (six-core)	24	64GB	RAID 10 -- SAS Disk x 16	279GB	Gigabit Ethernet	Red Hat Enterprise Linux Server release 6.3 (Santiago)

The tests comparing upgrade times between Oracle and DB2 used virtual systems with the following characteristics:

Function	Machine Type	CPU / Machine	Total # of CPU Cores/Machine	Memory/Machine	Disk capacity	Network interface	OS and Version
RDNG Server	VM image	8	8	32GB	80GB	Gigabit Ethernet	Red Hat Enterprise Linux Server release 6.6 (Santiago)
Database Server	VM image	8	8	16GB	80GB	Gigabit Ethernet	Red Hat Enterprise Linux Server release 6.6 (Santiago)

For more information

• Performance Datasheets: CLM

About the authors

GustafSvensson

---

Questions and comments:

• What other performance information would you like to see here?
• Do you have performance scenarios to share?
• Do you have scenarios that are not addressed in documentation?
• Where are you having problems in performance?

Warning: Can't find topic Deployment.PerformanceDatasheetReaderComments