Performance benchmarks and timing for data and event management operations

This topic presents timing approximations for the following operations in an environment that includes data collection and event management:

The following environment is used as a sample for the timing estimates.

Sample data collection and event management environment

Metric

Amount or value

Number of devices

20 K

Number of groups

2100

Number of monitor instance

200 K

Number of intelligent events

5 K per day

Number of attributes

1,500 K

Database size

60 GB

Average database growth per day

2 GB

Backup and restore, Infrastructure Management SAP SQL Anywhere ASA database

  • Infrastructure Management stores up to seven database restore log files (six files from Monday through Saturday, one file with the full backup)
  • Average size of each incremental log file: 13 GB
  • Time to restore each incremental log file: 30 minutes
    Infrastructure Management stores up to seven database restore log files (six from Monday through Saturday, and one that comes with the full backup)

The following table shows the amount of time necessary to restore a Infrastructure Management SAP SQL Anywhere ASA Database.

Backup files

Time to restore

1 incremental backup file

30 minutes

1 full backup + 1 incremental backup files

1 hour

1 full backup + 6 incremental backup files

3.5 hours

Start up time for the Infrastructure Management Server

The start up time for the BMC TrueSight Infrastructure Management Server with a SAP SQL Anywhere database and an Oracle database is 17 to 20 minutes.

Input/output operations per second for the Infrastructure Management Server and database

The input/output operations per second (IOPS) in a BMC TrueSight Infrastructure Management Server network is often measured by the number of read and write bytes per second for computer storage devices, such as the hard disk.

The following table describes the environment that is used as a sample for the timing estimates in a large deployment.

Sample data collection and event management environment

Metric

Amount or value

Number of devices

20k

Number of groups

2100

Number of monitor instance

200k

Number of intelligent events

5k per day

Number of attributes

1,500k

Database size

60 GB

Average database growth per day

2 GB

In this environment, BMC TrueSight Infrastructure Management Server and database take approximately 448 IOPS. BMC recommends up to 600 IOPS.

If multiple Infrastructure Management servers are connected to the same Oracle instance, the IOPS requirement is 600 * number of Infrastructure Management servers.

For example, in an environment with 3 Infrastructure Management servers, the IOPS requirement is 600*3 = 1800 IOPS.


The following tables show the IOPS over time for the database and server.

IOPS of the Infrastructure Management SAP SQL Anywhere database

MetricsValue
IO Data Bytes/sec3,602,277.436
IO Data Operations/sec448.489
IO Read Bytes/sec196,918.007
IO Write Bytes/sec3,405,359.429

IOPS of the Infrastructure Management Oracle database

TablespaceReadsAv Reads/sAv Rd (ms)Av Blks/RdWritesAv Writes/sBuffer WaitsAv Buf Wt (ms)
PNET_TS2,075,578163.888.5010,506,998833,127,9240.22
UNDOTBS149700.021.001,355,19911486,9990.09
SYSAUX34,57205.452.8943,754000.00
SYSTEM33,49605.191.6310,74604130.46
TEMP75201.2113.51841000.00
USERS49500.001.00495000.00
PNET_TS_TMP2015.001.000000.00

Network bandwidth utilization between a Infrastructure Management Server and Integration Service

In BMC TrueSight Infrastructure Management Server network, the bandwidth is often measured as the amount of data that is carried from the BMC TrueSight Infrastructure Management Server to the Integration Service over a specified period of time.

The following values show the network bandwidth utilization between the BMC TrueSight Infrastructure Management Server and Integration Service:

  • 265 KB per second
  • Average: 284 KB per second
  • Minimum: 106 KB per second
  • Maximum: 461 KB per second

Timing for the Integration Service failover

The following table describes the failover duration from one Integration Service to another.

Failover duration from Integration Service 1 to Integration Service 2

PATROL Agents

Integration Service load for failover

Integration Service failover duration
InstancesAttributes
50020k260k

5 minutes to 8 minutes

900

40k

500k

7 minutes to 12 minutes

Related topics

Performance benchmarks and timing for event and impact management operations

Performance benchmarks and timing for the Cloud Lifecycle Management integration

Was this page helpful? Yes No Submitting... Thank you

Comments