Kofax Communication Server 9.2.0 User Manual

Page 55

Advertising
background image

Environment Guide

Version 9.02.00

© Copyright Kofax, Inc. All information is subject to change without notice.

55

In the example above

, one logical processor (2 GHz) is assigned to the virtual machine. The “Virtual

machine reserve (percentage):” is set to 30 percent. Effectively, this machine has one processor (2 GHz)
and 615 MHz are reserved for this virtual machine. The virtual machine was assigned 512 MB RAM.

Supported KCS Components

All KCS components supported on VMware are also supported on Hyper-V. For more information, see KCS

Components Supported on Virtual Environment.

Similar restrictions and hints like for VMware apply also for Hyper-V. See chapters Restrictions and Hints.

The KCS core, i.e. TCOSS, must be installed on a dedicated virtual machine.

All other KCS applications connecting to the TCOSS must run on other virtual machines than TCOSS. It is

recommended to connect their VNICs (virtual network adapters) to the same network adapter as the VM of

TCOSS.

Tests and Performance Observations

Hyper-V was tested on an Intel XEON E5405@2GHz (Quad Core) computer with 7 GB memory.

The virtual machine was assigned one logical processor (2 GHz). The virtual machine reserve was not

configured; however, not much else was happening on the host computer. The virtual machine was

assigned 2048 MB memory. The host operating system was Windows Server 2008 R2; the guest operating

system was Windows Server 2008 R2 as well.

Several KCS server package 7.86.00 components were installed on the host, most prominently TCOSS

with 60 fax channels, connected with two LS1V2, each with 30 fax channels.

In such a configuration with 60 fax channels running in parallel, test ran successfully with an average CPU

usage of the virtual machine at around 13 percent.

Similar configuration with 30 FoIP channels was tested successfully as well.

During tests it was observed that extensive disk operations on the host (e.g., copying large volume of data)

may result in a dramatic performance decrease on the guest.

Advertising