Module 3: Monitoring Server Performance

Overview
Multimedia: The Primary Server Subsystems Monitoring Server Memory Monitoring Processor Usage Monitoring Disks Monitoring Network Usage

Multimedia: The Primary Server Subsystems

Server environments described in this presentation: file and print servers, application servers, and domain servers

Server subsystems: memory, processor, disk, and network

Important point to watch for: Which subsystem has the greatest effect on each server environment?

Lesson: Monitoring Server Memory

Why Monitor Server Memory?

How to Identify and Resolve Memory Bottlenecks

How to Monitor Memory

Why Monitor Server Memory?

Monitor server memory to determine whether any of the following conditions exist:
Memory bottlenecks

Insufficient memory
Excessive paging Memory leaks

How to Identify and Resolve Memory Bottlenecks

Memory counter
Pages/sec

Acceptable average range

020
Minimum of 5% of total memory Less than physical RAM Remain steady, no increase

Desired value
Low

Action
Find the process that is causing paging Add RAM Find the process that is using RAM Add RAM Find the process that is using RAM Add RAM
Check for memory leak in application Find the process that is causing paging Add RAM

Available Bytes Committed Bytes Pool Nonpaged Bytes

High

Low
Not applicable

Page Faults/sec

Below 5

Low

How to Monitor Memory

Your instructor will demonstrate how to:

Monitor memory by using Performance Monitor memory by using Task Manager

Practice: Monitoring Server Memory

In this practice, you will configure System Monitor to monitor selected memory counters

Lesson: Monitoring Processor Usage

What Is Processor Usage? How to Identify and Resolve Processor Bottlenecks How to Monitor Processor Usage

What Is Processor Usage?

Percentage of time that the processor is working Monitor to detect processor bottlenecks Tool
Task Manager Performance

Counter
CPU Usage %Processor Time Graph

Display
Percentage of elapsed time to run non-idle thread

How to Identify and Resolve Processor Bottlenecks

Processor counter
% Processor Time System: Processor Queue Length Server Work Queues: Queue Length

Acceptable average range

Less than 85%

Desired value Action

Low Find process using excessive processor time Upgrade or add another processor Upgrade or add additional processor Find process using excessive processor time Upgrade or add another processor Find controller card generating interrupts

Less than 10

Low

Less than four Depends on processor

Low Low

Interrupts/sec

How to Monitor Processor Usage

Your instructor will demonstrate how to:

Monitor processor usage by using Performance Monitor processor usage by using Task Manager

Practice: Monitoring Processor Usage

In this practice, you will configure System Monitor to monitor selected processor counters

Lesson: Monitoring Disks

Why Monitor Disks? How to Identify and Resolve Disk Bottlenecks How to Monitor Disks

Why Monitor Disks?

Monitor disks to determine:

Presence of disk bottlenecks

Need for disk defragmentation

Need for additional or faster disks Presence of excessive paging

Disk efficiency

How to Identify and Resolve Disk Bottlenecks

Desired high or low Action value
Low Monitor to see if paging is occurring Upgrade disk subsystem Upgrade disk subsystem Upgrade disk subsystem Upgrade disk subsystem

Physical disk counter

% Disk Time

Acceptable average range

Under 50%

Current Disk Queue Length Avg. Disk Bytes/Transfer Disk Bytes/sec

02 Baseline or higher Baseline or higher

Low High High

How to Monitor Disks

Your instructor will demonstrate how to monitor disks

Practice: Monitoring Disks

In this practice, you will configure System Monitor to monitor selected disk counters

Lesson: Monitoring Network Usage

What Is Network Usage? How to Identify and Resolve Network Bottlenecks How to Monitor Network Usage

What Is Network Usage?

Percentage of network bandwidth in use on the segment being monitored

Monitoring network usage helps you detect network bottlenecks

Bottlenecks in network communications directly affect the experience of the user at the client workstation and the entire network Typical causes of network bottlenecks are: Overloaded server

Overloaded network
Loss of network integrity

How to Identify and Resolve Network Bottlenecks

Network interface counter

Network Utilization (in Task Manager) Network Interface: Bytes Sent/sec Network Interface: Bytes Total/sec

Acceptable average range

Generally lower than 30% Baseline or higher

Desire high Action or low value

Low High Low Upgrade network adapter or physical network Perform further analysis to determine cause of problem Upgrade or add another adapter. Upgrade network adapter or physical network

Baseline or higher Less than 50% of the capacity of the bandwidth of the network card

High

Server: Bytes Received/Sec

NA

How to Monitor Network Usage

Your instructor will demonstrate how to monitor performance by using:

Performance Task Manager

Practice: Monitoring Network Usage

In this practice, you will configure System Monitor to monitor selected network counters

Guidelines for Using Counters and Thresholds

Subsystem Counter
Monitor page faults Monitor available RAM

Threshold
Over 5 per second Less than 5% of total

Monitor committed bytes

% Processor time, % Privileged Time, % User Time System: Processor Queue Length Server Work Queues: Queue Length % Disk Time Current Disk Queue Length Server: Bytes Total/sec, Network Interface: Bytes Total/sec

More than physical RAM

Above 85%

Above 2 Above 2 If more than 50%, check for excessive paging Greater than 2
Higher than the baseline number

Best Practices for Monitoring Server Performance

Set up Performance Logs and Alert Keep monitoring overhead low Analyze performance results and establish a performance baseline

Set alerts
Perform system tuning Monitor trends for capacity planning and add or upgrade components

Lab A: Monitoring Server Performance

In this lab, you will configure System Monitor to view selected processor, memory, network, and disk counters