Abstract; Introduction - HP ML570 - ProLiant - G2 Tuning Manual

Novell NetWare 6 performance tuning guidelines for ProLiant servers

abstract

introduction

IT (information technology) professionals and administrators are constantly under intense
pressure to cut cost, protect current investments, and meet immediate performance
demands without service interruption. Achieving these requirements within tight budgetary
constraints is not an easy task. In order to optimize your investment, it is important the
server operates at the highest performance level possible.
To achieve optimum performance, a server must contain key subsystem components in
addition to being configured and tuned correctly. This document provides information on
the important subsystem components to consider and suggestions on how to configure and
tune them.
Given that the operating system and hardware components contribute to overall server
performance, the focus of this document will be on the configuration and tuning of the
following key components:
network operating system (NOS) – NetWare 6
hardware components – disk, network, memory, processor, bus interface
The NetBench and WebBench test results sections of this document also provide sample
performance results to reinforce some of the recommendations and suggestions provided in
this document.
Determining the correct hardware / software mix for the user, takes in depth knowledge of
the environment in which the server will be deployed. This key element is vital in order to
properly tune and configure the server to maximize performance and eliminate
bottlenecks. Knowledge of the user environment includes the following aspects:
user needs – what the user requires in terms of security, storage space, etc.
user workload characterization – types of application that will be running
user performance requirements – level of response times / throughput expectation
budget – the amount of money allocated for the project
A thorough understanding of the user environment aids the IT professional or administrator
in selecting the correct hardware and software mix that meets the requirements without
exceeding the budget.
In this document the Ziff-Davis industry standard benchmark is used as the workload
generator. The NetBench and WebBench test suites are used to simulate "real-world" file
I/O requests and web application workloads respectively. The operating system platform
chosen is Novell NetWare 6 and the hardware selected is the ProLiant ML570 Generation
2 (G2) server.
Although the server used in this document is the ProLiant ML570 G2, the knowledge
gained as a result of this study can easily be applied to any HP servers deployed in similar
environments.
3