Intel AS/400 RISC Server Geographic Mirroring Measurements, Synchronization on an idle system

Geographic Mirroring Measurements

NOTE: The information that follows is based on performance measurements and analysis done in the IBM Server Group Division laboratory. Actual performance may vary significantly from this test.

Synchronization on an idle system:

The following data shows the time required to synchronize 1 terabyte of data. This test case could vary greatly depending on the speed and latency of communication between the two systems. In the following measurements, the same switch was used for both the source and target systems. Also, large objects were synchronized were synchronized, which tend to synchronize faster than small objects.

Time Required in Hours to Synchronize 1 Terabyte of Data using High Priority in Asynchronous mode

	1 Gigabit Line	2 Gigabit Lines	3 Gigabit Lines	4 Gigabit Lines
Time(Hours)	2.75	1.4	1.25	1.1

*This case represents best case scenario. An environment with objects ¼ the size of the objects used in this test caused synchronization times 4x’s larger.

Effects of Resume Priority Settings on Synchronization of 1 Terabyte of data using 4 gigabit lines in Asynchronous Mode.

	Low	Medium	High
Time(Minutes)	96:00	75:00	63:00
Source System CPU Overhead	9%	12%	16%
Target System CPU Overhead	12%	16%	18%

*This case represents best case scenario. An enviroment with objects ¼ the size of the objects used in this test caused synchronization times 4x’s larger.

Switchable Towers using Geographic Mirroring:

The following data shows the time required to switch a geographic mirrored IASP that is synchronized from the source system to the target system.

Time Required to Switch Towers using Geographic Mirroring using Asynchronous Mode

	Inactive Switchovers	Active Switchovers
Time(Minutes)	6:00	6:00

Active State:

The following measurements show the CPU overhead from an System/User ASP baseline on the source system.

CPU Overhead caused by Geographic Mirroring

	Asynchronous	Asynchronous	Synchronous	Synchronous
	Geographic Mirroring	Geographic	Geographic Mirroring	Geographic
	Synchronization Stage	Mirroring	Synchronization Stage	Mirroring
Source System CPU Overhead	19%	24%	19%	24%
Target System CPU Overhead	13%	13%	13%	13%

•Geographic Mirroring Synchronization Stage: The number reflects the amount of CPU utilized while in the synchronization mode on a 1-line system.

•Asynchronous Geographic Mirroring: The number reflects the overhead caused by mirroring in asynchronous mode using 1-line system and running the CPW workload.

•Synchronous Geographic Mirroring: The number reflects overhead caused by mirroring in synchronous mode using 1-line and running the CPW workload

IBM i 6.1 Performance Capabilities Reference - January/April/October 2008
© Copyright IBM Corp. 2008	Chapter 21 High Availability Performance	331