Website Monitoring – Lesson 1
Part A – Knowledge evidence
- Research the functionality and features of Microsoft Windows Performance Monitor in order to:
- Monitor Physical Disk Performance
- Monitor Processor Performance
- Monitor Memory Performance
- Monitor Networking Performance
- Research the functionality and features of a suitable software or application in order to:
- Monitor Internet Speed Performance of ISP
- Research the following functionality and features of a suitable software or application in order to:
- Monitor the Uptime of a website
- Monitor the Full page load
- Monitor the Transaction
Assessment:
- Upload your answers onto the web-based technical journal before 30/5/2017.
Further reading & research
- http://www.windowscentral.com/how-use-performance-monitor-windows-10
- http://www.makeuseof.com/tag/use-windows-performance-monitor-like-power-user/
- https://www.tpg.com.au/support/speed-troubleshooting
- http://www.speedtest.net/
- https://developers.google.com/speed/docs/insights/about
Part B – Performance evidence
Task 1 – Monitor Computer / Network Performance using Microsoft Performance Monitor
- Use the Windows key + R keyboard shortcut to open the Run command, type perfmon, and click OK to open. See an example below:
- Explain the results in the System Summary:
- How much resources (memory, network interface, hard disk, CPU) are used/available for use on the computer you are monitoring?
- Could you use this tool to monitor a remote server?
- Take a screen grab of the System Summary
- Open another resource-hungry application (Photoshop for example) and an educational YouTube video clip then compare the results with the screen grab you just took in the previous step:
- What is the new value of Memory [Available Mbytes]? What would happen if there was no memory left available? This scenario could be caused by a DOS attack or an old computer running out of resources.
- What is the new value of Network Interface [Bytes Total/sec]?
- On the left panel of the dashboard, expand Monitoring Tools then click Performance Monitor. Observe the graph showing values of Processor Information [% Processor Time]. Open a YouTube video and observe the changes. At which point(s) should we consider upgrading to a more powerful processor: 33%, 50% or 66%?
Now add another counter to monitor the network performance
- Click Add (or the green + sign) then select then select Network Adapter-Bytes Total/sec.
- Continue to another counter Memory [Available Mbytes]. Customise the Counter display (different colour for instance). Observe the changes as you add / remove applications.
Up to this point we know how to monitor performances in real time. Next we learn how to record the results for later viewing or reporting.
Now use Data Collector Sets to set and record performance (Admin access required).
- On the left panel of the dashboard, expand Data Collector Sets
- Right click User Defined – New – Data Collector Set. Enter a (your) name.
- Select Create data logs and tick the Performance counter option.
- Add the three counters for similar to the previous page:
- Configure the interval of 5 seconds
- Choose a safe place to store this Data Collector Set
- Add the three counters similar to the previous page: Memory, Network Adapter and Processor
In order to set a stop after a period of time:
- Right click the Data Set Collector and select Properties
- Click on the Stop Condition tab
- Check the Overall duration option and set the limits as needed.
In order to schedule a set to run automatically:
- Right click the Data Set Collector and select Properties
- Click on the Schedule tab
- Click Add. On “Folder Action”, define the schedule as required. Click OK.
In addition to the above steps to create your own sets you can also set up two built-in sets for System Diagnostics and System Performance.
Now use Reports to view data performance previously recorded.
- On the left panel of the dashboard, expand Reports – User Defined.
If you’re opening one of the System reports, you will see more information which may assist with your reporting.
- Right click the Data Set Collector and select Properties
- Click on the Schedule tab
- Click Add. On “Folder Action”, define the schedule as required. Click OK.
In addition to the above steps to create your own sets you can also set up two built-in sets for System Diagnostics and System Performance.
Now use Reports to view data performance previously recorded.
- On the left panel of the dashboard, expand Reports – User Defined.
If you’re opening one of the System reports, you will see more information which may assist with your reporting.
Task 2 – Run ISP Performance Speed Test
- Visit speedtest.net
- Click [BEGIN TEST] (do not click [START SCAN] for this lesson)
- Wait for the results then record Date, Time, Download & Upload results (see sample below):
| Testing tool | www.speedtest.net | Test condition: | ||
| Tester | Dean Nguyen | |||
| Location | Date / Time | Ping (ms) | Download (Mbps) | Upload (Mbps) |
| Sydney TAFE
Room ? |
10/10/2016
10am |
12 | 169 | 78 |
| Home? | ||||
| Office * | ||||
- Record measurements over three locations approximately at the same times (over different days of course. * Ask your supervisor for permission if testing at work
- Include both upload and download speed test results which are recorded in a spread sheet.
- Analyse and compare the differences. Answer the following:
- Which location has best / worst speed?
- Why is the result for upload so different to download?
- What would be acceptable (download/upload) speeds for home or office?
- Upload your answers on the web-based technical journal before 30/5/2017.
Part B – Performance evidence
Click Monitors drop-down menu then select Uptime Monitors then HTTP
—————————————————————————————————————————————–
Part A – Knowledge evidence
- Research the functionality and features of Microsoft Windows Performance Monitor in order to:
Monitor Physical Disk Performance
Performance Monitor is an administrative tool for reviewing system performance, both historically from log files or in real time which provides a visual display of inbuilt performance counters either accepted by default or selected individually to create a custom Data Collection Set.
System Monitor provides physical hard disk counters to achieve improved physical hard dive performance by:
- Troubleshooting latency frustrations.
- Detecting network bottlenecks.
- Identifying server performance problems.
- Uncovering intermittent faults.
- Planning the capacity of servers and subnets.
- Setting alerts to identify issues early
- Creating baselines when activity is low.
- Understanding the effect of workload on resources.
This allows load balancing to ensure the best use of resources is achieved and identifies where the minimum investment results in the maximum performance improvements.
Performance Monitor provides numerous features to make data collecting easier but still requires a methodical approach to ensure vital clues are not omitted. Analysis should always include checking the major resource objects i.e. the processor, memory, disk and network.
Once the appropriate counters are selected the actual techniques are relatively easy to apply to generate the data report. The collected data hence needs to be interpreted using the following skills:
- Knowledge of network principles and server components
- Ability to notice something out of the ordinary
- Experience and confidence with performance counters
For hard drive performance monitoring, at a minimum, the following counters should be monitored:
https://technet.microsoft.com/en-au/library/cc938959.aspx
- PhysicalDisk\ % Disk Time
The reports the percentage of time that the selected disk drive is busy servicing read or write requests. As this counter-data can extend over multiple samples, and hence over-estimate disk usage, this value should be compared against % Idle Time for a more accurate assessment
- LogicalDisk\ % Free Space
This counter reports the percentage of free disk space to the total usable space on the logical volume. In calculating the _Total instance, the %Free Space counters recalculate the sum as a percentage for each disk. There is no % Free Space counter for the PhysicalDisk object.
- PhysicalDisk\ Disk Reads/sec and Disk Writes/sec
The % Disk Read Time and % Disk Write Time counters can exaggerate disk time as they report busy time based on the duration of the I/O request, which includes time spent on activities other than reading to or writing from the disk.
The busy time for all requests is then summed and divided by the elapsed time of the sample interval. If multiple requests are in process at a time, the total request time is greater than the time of the sample interval and as a result, reported disk utilization can exceed actual utilization.
Additionally, the following counters can be monitored:
- PhysicalDisk\ Avg. Disk Bytes/Transfer
This counter reflects the time a hard dive requires to complete requests. High values may indicate the hard disk controller is continually retrying to access the hard disk due to access failures. These failures increase average hard disk transfer time.
- PhysicalDisk\ Avg. Disk sec/Transfer
Here values exceeding 20 kilobytes (KBs) indicate the hard dive is performing well. Low values result when an application accesses an inefficient hard disk. Better disk I/O performance is achieved after registry settings are tuned, defragmented and aligned .
- PhysicalDisk\ Disk Bytes/sec
As the operating system does not collect performance data for the logical hard disk by default, diskperf –yv must be entered at the command prompt to direct the hard disk performance statistics driver, to report data for logical drives or storage volumes.
To improve system performance PhysicalDisk counters, monitor the activity of physical disks including USB drives, to collect current performance data that can be reviewed and used to set the identified parameters from this analysis to improve disk performance. These counters can also be used to monitor disk space, efficiency and disk operations in detail.
LogicalDisk object counters do not appear in the user interface until enabled.
PhysicalDisk object counters are by default enabled on the operating system and shown in the Performance console on the user interface. To obtain performance data on logical volumes, use the diskperf command enables the LogicalDisk performance counters. The LogicalDisk object counters do not appear in the user interface until enabled
Establish disk performance baseline.
A system’s performance baseline is the level of performance that can be reliably expected during normal usage. Once a baseline is established, it identifies when a system is experiencing performance issues, because counter levels disagree with the baseline readings.
Analyze disk counter data that might suggest a disk bottleneck.
Specific disk counters provide data about disk paging activity, disk utilization, queuing of requests and rates of throughput. Observing these counters helps to determine when a disk bottleneck is developing. When the cause of a disk bottleneck has been determined, steps to correct the problem by changing disk-system configuration, defragmenting disks, upgrading hardware or other tuning methods can be applied.
Research the functionality and features of Microsoft Windows Performance Monitor in order to monitor Processor Performance.
Performance Monitor is administrative tool for standalone servers allowing a comprehensive profile, of server resources management to be complied for analysis.
The analysis of the server’s current performance begins with clicking the Performance Monitor tab under Monitoring Tools on the program’s main panel, as shown below in Figure 1.
Figure 1. Begin the server’s current performance analysis by clicking the Performance Monitor tab under Monitoring Tools on the program’s main panel
To investigate processor performance for a particular set of programs, open each program consecutively, noting it’s impact on the system’s resources in the resulting graphs. By default, the graph displayed in Performance Monitor measures Processor time (the time the processor is busy executing active programs, shown in percentages) and provides a measure of how hard the processor is working. See Figure 2. below.
Figure 2. A display of the computer’s performs while using a certain set of apps and consecutively opening further programs and noting their impact on the system’s resources.
To obtain a more comprehensive analysis, further counters can be added to capture other data by clicking the green plus sign (in the task bar above the graph) to open the “Add Counters” window. See Figure 2.
Opening a counter, reveals several options that can be selected either individually or collectively and activated by clicking on the green “Add” button . The “Add Counters” window allows selection from an extensive list of counters, organised by type, to monitor data in real time (see Figure.3 below). Once finished selecting the counters and the objects to be monitored, click the Add button.
Figure 3. The “Add Counters” window allows selection from an extensive list of counters, organised by type, to monitor data in real time
Newly added counters will display on the right side of the window (see Figure 4) and when double clicked they are added to the graph displayed by Performance Monitor (see Figure 5).
Figure 4. Newly added counters will display on the right side of the window.
Figure 5. When new counters are added, their impact is displayed in the graphical output generated by Performance Monitor
The Processor counter set selected will display data, in real time they change the computer’s processing capacity. Displayed data includes the total amount of non-idle time consumed performing User mode operations (%User Time) or the numbers of interrupts to which the processor has responded (Interrupts/sec). Data output can be displayed in a variety of formats by clicking the Change graph type button and choosing the Histogram bar or Report options. See Figure 7.
Further changes to how the data is displayed, can be obtained by clicking the Properties button (or CTRL+Q) to the Performance Monitor Properties window, where the display of each counter can be customised for color, type of lines, etc. Both the Data and the Graph tabs can be used for customisation. See Figure 7.
Figure 7. Customisation of displayed data can be achieved through the the Properties button (or CTRL+Q) to the Performance Monitor Properties window.
The limitation of the data generated by Performance Monitor‘s graphical reports is its highly technical nature making it difficult to understand by novices. As an aid, a description of any counter function can be obtained by selecting that counter in the Add Counters dialog and clicking Explain.
If you are looking for a good knowledge base, explaining all the gibberish terms like %DPC Time or Page Faults/sec, read this entry on Microsoft’s TechNet: Performance Monitor Counters. There you will find complete information about each counter found in the standard list of reports.
References:
1. http://www.digitalcitizen.life/basics-about-working-performance-monitor
2. https://www.serverdensity.com/monitor/windows/how-to/
Research the functionality and features of Microsoft Windows Performance Monitor in order to monitor Memory Performance
Monitor Networking Performance
Issues due inadequate memory often appear as problems in other parts of the system. Adequate memory requirements should first be verified before checking other components.
To determine if the current amount of server memory is sufficient Performance Monitor may be used. Logging the following counters allows determination if any performance bottlenecks associated with memory:
- Memory: Available Bytes. Ensure at least ten percent of available memory is reserved for periods of peak use.
- Memory: Page Faults/sec, Memory: Pages Input/sec, and Memory: Page Reads/sec. If a process requests a page in memory and the system cannot find it at the requested location, this constitutes a page fault:
If the page is elsewhere in memory, the fault is called a soft page fault. Most processors can handle large numbers of soft faults without delay issues.
If the page must be retrieved from disk, the fault is called a hard page fault. However, hard faults can cause significant delays.
Page Faults/sec is the overall rate at which the processor handles faulted pages, including both hard and soft page faults.
Pages Input/sec is the total number of pages read from disk to resolve hard page faults.
Page Reads/sec is the number of times the disk was read to resolve hard page faults.
Pages Input/sec will be greater than or equal to Page Reads/sec and gives a good estimate of the hard page fault rate. If these numbers are low, the server is responding quickly to requests . If the numbers are high, it may be because excessive memory is dedicated to the caches, providing insufficient memory for the remainder of the system. The amount of RAM on the server needs to be increased, though lowering cache sizes is also an effective option.
- Memory: Cache Bytes, Internet Information Services Global: File Cache Hits %, Internet Information Services Global: File Cache Flushes, and Internet Information Services Global: File Cache Hits.
The first counter, Memory: Cache Bytes, reveals the size of the File System Cache, which by default is set at 50% of available physical memory. As Internet Information Services (IIS) will automatically reduce cache if it is running out of memory, note the direction in which this counter trends.
The second counter is the ratio of cache hits to total cache requests and reflects how well the settings for the IIS File Cache are working. For a site principally consisting of static files, 80% or more cache hits is considered efficient.
Compare logs for the last two counters, IIS Global: File Cache Flushes and IIS Global: File Cache Hits, to determine if objects are being flushed out of the cache at an appropriate rate. If flushes are occurring too slowly, memory may be wasted. Alternatively if flushes are occurring too quickly, objects may be flushed from cache more often than necessary.
- Page File Bytes: Total.
This counter reflects the size of the system’s paging file(s): the larger the paging file, the more memory the system has commited.
Striping a paging file across separate physical drives improves paging file performance (apply to drives that do not contain site’s content or log files). The paging file on the system drive should be at least twice the size of physical memory, so that the system can write the entire contents of RAM to disk if a crash occurs.
- Memory: Pool Paged Bytes, Memory: Pool Nonpaged Bytes, Process: Pool Paged Bytes: Inetinfo, Process: Pool Nonpaged Bytes: Inetinfo, Process: Pool Paged Bytes: dllhost# n , and Process: Pool Nonpaged Bytes: dllhost.
Memory: Pool Paged Bytes and Memory: Pool Nonpaged Bytes monitors the pool space for all processes on the server. The other counters listed here monitor the pool space used directly by IIS 5.0, either by the Inetinfo process or by the Dllhost processes (in which isolated or pooled applications run) instantiated on your server.
The system’s memory pools hold objects created and used by applications and the operating system. Ensure that counters are monitored for all instances of Dllhost on the server; otherwise, inaccurate readings of pool space will be used by IIS.
The contents of the memory pools are accessible only in privileged mode being the kernel of the operating system can directly use the memory pools; user processes cannot. On servers running IIS 5.0, threads that service connections are stored in the non-paged pool along with other objects used by the service, such as file handles and sockets.
Apart from adding more RAM, the following techniques may also enhance memory performance:
Improve data organization (e.g. disk mirroring or striping)
Re-time the IIS File Cache, deleting unnecessary features
Change the balance of the File System Cache to the IIS 5.0 Working Set.
Enlarge paging files
Replace CGI applications with ISAPI or ASP applications
References:
1. https://technet.microsoft.com/en-us/library/dd722744.aspx
2. http://www.appadmintools.com/documents/windows-performance-counters-explained/
3. http://windowsitpro.com/windows/jsi-tip-4792-evaluating-memory-and-cache-usage
Research the functionality and features of Microsoft Windows Performance Monitor in order to monitor Networking Performance
Traditionally, network monitoring was predominately used to watch the communication paths used to transmit and receive data to ensure performance was as expected. Today, this process has expanded to include the supervision and measurement of a computer network’s performance and activity on beyond just communications paths.
The most significant advantage of Windows Performance Monitor over tools like System Monitor, Performance Logs and Alerts and Server Performance Advisor is that it combines the functionality of all these tools into a an interface with common methods for defining the data to be collected. The Data Collector Set makes a group of counters portable which can be selected and set as described above.
Performance Monitor allows the setting of counters for specific CPU, memory, disk and network events. A review of the output from Performance Monitor hence allows the identification of possible issues. Possible counters to include in monitoring networking performance may include:
- PhysicalDisk: Disk Write Bytes/sec
- Processor: % Idle Time
- Processor: Interrupts/sec
- System: Threads
- Memory: % Committed Bytes In Use
- Memory: Page Faults/sec
- PhysicalDisk: Disk Read Bytes/sec
- PhysicalDisk: Disk Reads/sec
- PhysicalDisk: Disk Writes/sec
Reference:
1. https://technet.microsoft.com/en-us/library/dd744567(v=ws.10).aspx
2. web1.muirfield-h.schools.nsw.edu.au/technology/…/documents/1738_reading.doc
Research the functionality and features of a suitable software or application in order to monitor Internet Speed Performance of ISP.
NDT (Network Diagnostic Test)
NDT (Network Diagnostic Test) is a speed and diagnostic tool that reports upload and download internet speeds, attempts to determine what problems limit speeds and provides a detailed diagnostic reporting on the results. See Figure 8.
While the diagnostic messages are most useful for expert users, they can also help novice users by allowing them to provide issue reports to their ISP provider.
Figure 8: An example of Internet speed Testing conducted on MLab.
Internet speed Testing was conducted at work over 3 days at 11.00 am using the “Speedtest” utility. Daily results varied and are presented in the table below:
| 28/05/2017 | 29/05/2017 | 30/05/2017 |
| 13 ms
DOWNLOAD: 36.23 Mbps UPLOAD: 1.08 Mbps |
11 ms
DOWNLOAD: 36.31 Mbps UPLOAD: 1.12 Mbps |
12 ms
DOWNLOAD: 27.47 Mbps. UPLOAD: 1.04 Mbps |
Table 1. Internet speed Testing Using the “Speedtest” utility over 3 consecutive days for the same system.
References:
1. http://www.speedtest.net/
2. https://www.measurementlab.net/tests/
Research the following functionality and features of a suitable software or application in order to monitor the Up-time of a website
5 Free Services to Monitor Site Uptime
Functionality and Features are summarised beneath each service
1. UptimeRobot
Number of Sites you can Monitor: 50
Regularity of Checks: Every 5 minutes
Methods of Alerts: E-mail, SMS, RSS, Twitter coming soon
UptimeRobot checks a site header’s status codes every five minutes and if the code returns a problem, does further checks. If issues persist it notifies the site owner.
2. Pingdom
Number of Sites you can Monitor: 1
Regularity of Checks: User set, from 1 minute upwards
Methods of Alerts: E-mail, SMS (up to 20 per month), push alerts via iPhone app
Pingdom, normally a paid application, offers a very basic free account. Although the free service only allows the monitoring of one site, it’s worthy of mention as it comes with a free iPhone app that provides a mobile solution for monitoring a website’s status.
3. Mon.itor.us
Number of Sites you can Monitor: 1
Regularity of Checks: Every 30 minutes
Methods of Alerts: IM, SMS, E-mail, RSS
Mon.itor.us is the free associated service to Monitis. It a simple set-up for external monitoring, alerts, tracking of failures and generates a daily, weejkly or monthly report of the uptime/downtime of a website.
4. InternetSeer
Number of Sites you can Monitor: 1
Regularity of Checks: Every hour
Methods of Alerts: E-mail, SMS, pager (!)
InternetSeer claims to be the largest website monitoring service, utilised for over 1.7 million sites globally. While critised for excessive email marketing, in addition to the ability to set up alerts for multiple contacts, the service generates a free weekly report.
5. Uptrends
Number of Sites you can Monitor: 1
Regularity of Checks: Every 30 minutes
Methods of Alerts: On-site info only
Uptrends is more suited to those wanting to make uptime data readily available to their users, rather than an alert-based service.
Once installed and embedded Uptrends monitors website up-time every 30 minutes from worldwide checkpoints generating an up-time report of the last 24 hours, last 7 days, month or year as required.
Reference:
1. http://mashable.com/2010/04/09/free-uptime-monitoring/#Q_PI2anA0kq3
Research the following functionality and features of a suitable software or application in order to monitor the Full page load.
Full page load time is measured from the moment a request is sent from the client until the page and all elements are loaded and displayed in the web browser to provide an estimate of the latency period that a client experiences when visiting a specified website.
Unacceptable delays can significantly drop customer conversion rates, causing them to leave even before they have looked at the website. For a global company, it is important to analyze the page load time from different geographic locations.
Anturis Software maintains a number of Public Agents internationally for use in performing full page load time monitoring, to see how the measurements differ for different customers around the world.
The Anturis Full Page monitor enables:
- Tracking the time required for a WebKit browser to load a web page, including all images, JavaScript and CSS elements. Additionally, a waterfall chart of all page elements (images, CSS, JavaScript, etc) is generated showing the sequence in which the elements are loaded and the time taken to load each element. This enables analyses of which elements affect page load time the most.
- Monitor full page load time from different global locations .
- Set custom load time thresholds for errors and warnings.
- Be alerted when a threshold is exceeded, HTTP status is not as expected or when page content is missing.
Reference:
1. https://anturis.com/full-page-load-monitoring/
Research the following functionality and features of a suitable software or application in order to monitor the Transaction.
Monitis Transaction Monitoing
There is always a risk that site applications such as shopping carts, registration pages, etc are functioning improperly even when the total site appears fine and operational.
The Monitis transaction monitoring service ensures the site’s e-commerce engine is functioning properly, by simulating a client’s interaction with the website, continually monitoring web and application transactions sending immediate alerts when any step in the online process becomes dysfunctional.
Extra Information
Monitis Software is a total solution by combining Website uptime,
Full page load and Transaction monitoring to enhance the end-user experience.
Uptime Monitoring
Monitis allows free checking of availability and response time of websites, applications, cloud and hosted services around the clock with with completely cloud based tools, meaning no software or hardware is required. Features include:
- Down to 1 minute monitoring intervals and 2 years historical archive
- 30+ global check locations from major markets worldwide
- Multiple protocols, such as HTTP, HTTPS, PING, DNS
- Email checks using SMTP, POP3 and IMAP protocols
- Public IP checks using TCP, UDP and ICMP protocols
- VoIP checks using SIP protocol
- IPv6 support
- Webpage content checks
- Instant failure alerts via email, instant message, Text/SMS or live voice, mobile apps
- Detailed daily, weekly, monthly and ad hoc reporting with service-level metrics per interval
Full page load monitoring.
Monitis features:
- Load times of all images, CSS, JavaScript, RSS, Flash, frames, etc.
- Detailed timelines of each individual request
- Alerts when elements of any page don’t load(e.g., “404 error”)
- Breakdowns of page loads per components
Part B – Performance evidence
Task 1 – Monitor Computer / Network Performance using Microsoft Performance Monitor
- Explain the results in the System Summary:
- How much resources (memory, network interface, hard disk, CPU) are used/available for use on the computer you are monitoring?
- Take a screen grab of the System Summary
The first screen of Performance Monitors is a system summary which includes some technical stats about the CPU and memory.
Memory
% Committed Bytes: is a measure of the demand for virtual memory.
Figure 9: Indicates only 3.42 GB of RAM is presently in use with a remaining capacity of 12.89 Mb (approx. 79%).
In the image above, 3.42/16.00 GB of Ram (approx 21%) is in current use leaving a further 12.89 GB available . This indicates the number of bytes allocated by processes and to which the operating system has committed a RAM page frame and/or a page slot in the pagefile.
Cache faults/sec is an indication of the times the Server has attempted to find a page in the cache, but being unsuccessful has loaded it from disk instead. The higher the faults/sec the higher the incidence is of seeking to load the file from disk. In this example it occurs 9.0/sec. See Figure 10.
To detect if the faults/sec is affecting system performance, monitor wait stats for IO/Disk. If there are high waits in conjunction with a high cache fault/sec number, then performance issues probably exist.
Figure 10: Cache faults/sec is an indication of the incidences the Server has attempted to find a page in the cache, but being unsuccessful seeks and loads it from disk. The higher the faults/sec the higher the incidence is of seeking to load the file from disk.
The performance data collection engine collects data from the Bytes Total/sec performance counter of the Network Interface performance object to analyse performance data. If the Exchange Server Analyser collects any data from the Bytes Total/sec performance counter, the Exchange Server Analyser displays information about the collected data as baseline information.
The Bytes Total/sec performance counter indicates the performance of the network adapter. If the Bytes Total/sec counter reports high values, this indicates a large number of successful transmissions. For the Intel Dual Band Adaptor installed in this computer the transmission rate is 300 Mb/s. For a 100 Mbps network adaptor, the value of the Bytes Total/sec performance counter should be under 7 megabytes/second. In this case the counter value is 1.5 MB/sec allowing the conclusion the network adaptor has adequate unused resources. See Figure 10.
Comparing the Bytes Total/sec counter value with the value reported by the Network Interface\Current Bandwidth counter, reflects each adapter’s bandwidth. As the Bytes Total/sec rate approaching the maximum transfer rate, the probability of collisions on the network increases and is likely to impact performance by increasing the latency of packet transfer on the network.
The PhysicalDisk performance object monitors disk drives (physical hardware) on the computer and the counters are the sum of the access to all partitions on the physical drives.
The % Idle Time counter provides a precise measure of the time the disk remained in an idle state, meaning all the requests from the operating system to the disk had been completed and there is zero pending requests.
% Idle Time is calculated as follows: the system time stamps as an event when the disk goes idle, then stamps another event when the disk receives a new request. At the end of the capture interval, the percentage of the time spent in idle can be calculated . This counter ranges from 100 (meaning always Idle) to 0 (meaning never idle or always busy). In the example in Figure 11. the % idle time is 99.3% and inferring the hard drive C: is almost always idle.
Figure 11. Shoes the Ave Disk Queue Length is 0.009% indicating that no bottlenecks presently exist and that the idle time exceeds 99% for the physical drives.
The Avg. Disk Queue Length – indicates how often the disk is busy and when busy most of the time, a queue may form. The size of queue indicates if any bottlenecks are occurring around disk processing and slowing system performance. In the example shown in Figure 11 the Ave Disk Queue Length is 0.009 indicating that no bottlenecks presently exists due to the physical drives.
Figure 12: Processor Information.
Processor information:
% Interrupt Time is the time the processor spends receiving and servicing hardware interrupts during sample intervals and is an indirect indicator of the activity of devices that generate interrupts, such as the system clock, the mouse, disk drivers, data communication lines, network interface cards and other peripheral devices. When interrupts occur during active time, they cause processes and threads to be suspended until processing of the interrupt is complete.
These devices normally interrupt the processor upon completion of a task or in need of attention. A high % Interrupt Time usually indicates an issue with a hardware device. Figure 12 indicates there are presently no peripheral device interruptions and no peripheral device issues in need of attention.
% Processor Time is the percentage of elapsed time that the processor needs to execute a non-Idle thread. It is calculated by measuring the percentage of time that the processor spends executing the idle thread and then subtracting that value from 100%.
The counter is the primary indicator of processor activity, displaying the average percentage of busy time observed during sample intervals and is used to determine the usage of the computer’s processor. In the Figure 12 example, the processor is busy 4.06% of the time.
Parking Status dynamically selects a set of processors to remain idle based on the current power policy and their recent usage. The scheduler will attempt to honor this selection when it decides on which processors to run threads, allowing the parked cores to enter idle states where they consume little power and extend operational battery length for laptops. Currently, Figure 12 indicates all core processors are active.
References:
1. https://technet.microsoft.com/en-us/library/aa996576(v=exchg.80).aspx
2. https://technet.microsoft.com/en-us/library/cc938653.aspx
3. https://www.microsoft.com/technet/…/5d3d3ae5-235f-4284-9fae-08a3892bd573.msp…
4 . http://mpwiki.viacode.com/default.aspx?g=posts&t=109252
5. https://social.technet.microsoft.com/wiki/contents/articles/12984.understanding-processor-processor-time-and-process-processor-time.aspx
6. https://www.myfavoritegadgets.info/tipstricks/CoreParking.html
Could you use this tool to monitor a remote server?
Yes, but to view performance counters from a remote computer, the Performance Logs and Alerts firewall exception must be enabled on the remote computer.
Open another resource-hungry application (Photoshop for example) and an educational YouTube video clip then compare the results with the screen grab you just took in the previous step:
Figure 13: Perfmon stats with Youtube active and Photoshop opened
Comparison of Figure. 9 with Figure 13 indicates :
- The amount of Ram in use has increased from 21% to 22.05% decreasing the available RAM left from 12.89 GB to 12.39 GB .
- Cache faults/sec has deceased from 9.0/sec to 3.04 faults/sec indicating the Server has been more successful at finding required page in the cache, thereby needing to look for the page and load it from the disk drive fewer times.
Figure 14: Perfmon stats with YouTube active and Photoshop opened
Comparison of Figure.10 and Figure.11 with Figure.14 indicates:
- In this instance the counter value has increased from 1.5 MB/sec to 3.79 MB/sec showing the number of successful transmissions by the network adaptor increased in response to the demands of the two extra open applications.
- In the example in Figure 11. the % idle time was 99.3% and inferred the hard drive C: is almost always idle. In Figure 14. the % idle time has decreased to 64.01% in response to the extra processing requirements of YouTube and Photoshop.
- In the example shown in Figure 11 the Ave Disk Queue Length is 0.009 compared with Figure 14. where the ave queue length has increased to 0.362 indicating at a 40 fold increase in the queue length. Irrespective no bottlenecks have developed with the disk’s processing of requests due to the increased resources required to run the extra applications.
Figure 15. Processor Information with YouTube active and Photoshop opened
Comparison of Figure.12 and with Figure.15 indicates:
Processor information:
- There has been no change in % Interrupt Time and irrespective of the extra resource demand, no peripheral device interruptions or peripheral device issues are in need of attention.
- In Figure.15, % Processor Time has increased from 4.06% in Figure.12 to 12.59% in response to increased processing requests from the extra applications.
- There has been no change in core parking status between Figures. 12 and 15.
What is the new value of Memory [Available Mbytes]?
The new value of Available MB is 12.39 GB .
What would happen if there was no memory left available? This scenario could be caused by a DOS attack or an old computer running out of resources.
When a computer exhausts its RAM memory, it will begin using the hard drive space for “Virtual Memory”. However s the hard drive read/write speeds are much slower than the RAM read/write speeds a noticeable degradation in performance will be observed. The more “Virtual Memory” used, the slower the computer will perform until it fails.
What is the new value of Network Interface [Bytes Total/sec]?
The new value of Network Interface [Bytes Total/sec] is 3,793.8 Mb.
On the left panel of the dashboard, expand Monitoring Tools then click Performance Monitor. Observe the graph showing values of Processor Information [% Processor Time]. Open a YouTube video and observe the changes. At which point(s) should we consider upgrading to a more powerful processor: 33%, 50% or 66%?
Processor Information [% Processor Time].
Now add another counter to monitor the network performance
- Click Add (or the green + sign) then select then select Network Adapter-Bytes Total/sec.
- Continue to another counter Memory [Available Mbytes]. Customise the Counter display (different colour for instance). Observe the changes as you add / remove applications.
Up to this point we know how to monitor performances in real time. Next we learn how to record the results for later viewing or reporting.
Now use Data Collector Sets to set and record performance (Admin access required).
- On the left panel of the dashboard, expand Data Collector Sets
- Right click User Defined – New – Data Collector Set. Enter a (your) name.
- Select Create data logs and tick the Performance counter option.
- Add the three counters for similar to the previous page:
- Configure the interval of 5 seconds
- Choose a safe place to store this Data Collector Set
- Add the three counters similar to the previous page: Memory, Network Adapter and Processor
In order to set a stop after a period of time:
- Right click the Data Set Collector and select Properties
- Click on the Stop Condition tab
- Check the Overall duration option and set the limits as needed.
In order to schedule a set to run automatically:
- Right click the Data Set Collector and select Properties
- Click on the Schedule tab
- Click Add. On “Folder Action”, define the schedule as required. Click OK.
In addition to the above steps to create your own sets you can also set up two built-in sets for System Diagnostics and System Performance.
Now use Reports to view data performance previously recorded.
- On the left panel of the dashboard, expand Reports – User Defined.
If you’re opening one of the System reports, you will see more information which may assist with your reporting.
- Right click the Data Set Collector and select Properties
- Click on the Schedule tab
- Click Add. On “Folder Action”, define the schedule as required. Click OK.
In addition to the above steps to create your own sets you can also set up two built-in sets for System Diagnostics and System Performance.
Now use Reports to view data performance previously recorded.
- On the left panel of the dashboard, expand Reports – User Defined.
If you’re opening one of the System reports, you will see more information which may assist with your reporting.
Task 2 – Run ISP Performance Speed Test
- Visit speedtest.net
- Click [BEGIN TEST] (do not click [START SCAN] for this lesson)
- Wait for the results then record Date, Time, Download & Upload results
- Record measurements over three locations approximately at the same times (over different days of course.(see sample below):
| ting tool | www.speedtest.net | Test condition: | ||
| Tester | James Boian | |||
| Location | Date / Time | Ping (ms) | Download (Mbps) | Upload (Mbps) |
| Work | 28/05/2017
29/05/2017 30/05/2017 11am |
13
31 32
|
36.23
36.31 27.47
|
1.08
1.12 1.04
|
| Home | 28/05/2017
29/05/2017 30/05/2017 9pm |
39
12 13
|
45.26
44.78 45.71
|
1.78
1.61 1.85
|
|
Mobile Oatley |
28/05/2017
29/05/2017 30/05/2017 8.30am |
12
13 13
|
5.10
7.63 6.91
|
0.06
0.74 0.69
|
Include both upload and download speed test results which are recorded in a spread sheet.
Figure: Upload and download speed test results recorded in a spread sheet.
Analyse and compare the differences. Answer the following:
Which location has best / worst speed?
Beverly Hills – Work desktop has the fastest download and upload speeds. Faster speeds due to newer internet cabling and closer proximity to exchange.
Oatley – Home desktop has slower download and upload speeds
Oatley – Home mobile has slowest download and upload speeds
Oatley location is limited in number of telephone connections to the area due to waterfront location and quality of copper wire between the telephone exchange and the home are old.
Other factors affecting the speed of the internet may include:
- the number of other people concurrently using the service, which can cause congestion on the copper wires
- the computer equipment used to access the internet, such as the router.
Why is the result for upload so different to download?
Users spend more time downloading items from internet which includes viewing web pages, videos or other multi-media files all of which are generally large and can take long periods to download. Faster download speeds suit most uses better to minimise download waiting periods.
Most users do not upload large files to the net (mainly social media comments or e-mails) and as these are usually small files and do not require lengthy upload times even at the lower speeds. As users, do not wait for the upload to be completed before undertaking their next task they prefer that Internet providers give higher priority to downloading speeds.
What would be acceptable (download/upload) speeds for home or office?
Acceptable internet speeds depend on he tasks to be accomplished and the number of users accomplishing the tasks. For example if at work a few people occasionally look something up online or send an email, much less bandwidth is required than if many users are constantly online uploading and downloading large graphics files.
Available internet connection options include:
- DSL – common speeds are 3 x .384, 3 x 3, and up to 45Mbps Download.
- Cable – common business class speeds are 5 x 1, 10 x 1, 30 x 5, or 50 x 10
- Fiber Optic Cable – any speed, but typically 10 X 10 and up
Reference:
1. https://www.ciinc.com/business-internet-speed/
Task 3. Website Performance Testing
Pagespeed
PageSpeed Insights measures the performance of a page for mobile and desktop devices. fetching the url twice: once with a mobile user-agent and hence with a desktop user-agent.
PageSpeed Insights checks to see if a page has applied common performance best practices and provides a score, which ranges from 0 to 100 points, and falls into one of the following three colour indicators categories:
- Green : The page applies most performance best practices and should deliver a good user experience.
- Yellow: The page is missing some common performance optimizations that may result in a slow user experience. Please investigate the recommendations below.
- Red: The page is not optimized and is likely to deliver a slow user experience.
See images below.
Figure: Results of Pagespeed mobile test applied to http://www.sydneytafe.deu.au
Figure: Results of Pagespeed desktop test applied to http://www.sydneytafe.deu.au
- Ensuring User Satisfaction
- Saving Business Revenue
- Lowering IT Operations Costs
- Increasing Operational Efficiency
- Safeguarding New Investments
- Website Monitoring: is the process of testing and verifying that end-users can interact with a website as expected. Website monitoring is therefore used to ensure website up-time, performance and functionality is as expected.
A website monitoring service can check other internet protocols besides HTTP pages and HTTPS such as FTP, SMTP, POP3, ActiveSync, IMAP, DNS, SSH, Telnet, SSL, TCP, PING, UDP, SOAP, Domain Name Expiry, SSL Certificate Expiry and a range of ports.
- Network Monitoring: is the use of a system that constantly monitors a computer network for slow or failing components ( e.g. routers, switches, firewalls, Wireless, Load Balancers, WAN Accelerators, Printers, UPS and Storage) and notifies the network administrator, via email, SMS , etc, if an outages or other issue requiring attention.
- Cloud Monitoring: is the management of cloud computing products and services. At a minimum, a cloud-management system should have the ability to:
- manage a pool of heterogeneous compute-resources
- provide access to end users
- monitor security
- manage resource allocation
- manage tracking
For composite applications, cloud management systems also encompass frameworks for workflow-mapping and -management.
- Server Monitoring: is the in-depth visibility of the system’s key performance indicators with the aim of identifying server issues and their root cause. Server monitoring may include:
- Server Monitoring
- Windows Monitoring
- Linux Monitoring
- Site24x7 Plugins
- IIS Monitoring
- SQL Monitoring
- SharePoint Server
- MS Exchange Server
- Linux Syslog Monitoring
- File and Directory Monitoring
- Log File Monitoring
- Docker Monitoring
- VMware Monitoring
- Cloud Monitoring
- Application Monitoring: is a process that ensures a software application, processes and performs in an expected manner and scope. This technique routinely identifies, measures and evaluates the performance of an application and provides the means to isolate and rectify any abnormalities or unexpected behavior.
References:
1. https://www.techopedia.com/definition/29133/application-monitoring
2. http://www.site24x7.com
3. https://en.wikipedia.org/wiki/Cloud_management
4. https://en.wikipedia.org/wiki/Website_monitoring
Part B – Performance evidence
Click Monitors drop-down menu then select Uptime Monitors then HTTP
Figure: 8 monitoring locations have been added across the US, Europe, Middle East, Asia and Australia as shown along the bottom of the figure
dsjboiancertiv2017technicaljournal.wordpress.com 