Skip Headers
Oracle® Enterprise Manager Framework, Host, and Third-Party Metric Reference Manual
10g Release 2 (10.2)

Part Number B16230-01
Go to Documentation Home
Home
Go to Book List
Book List
Go to Table of Contents
Contents
Go to Master Index
Master Index
Go to Feedback page
Contact Us

Go to previous page
Previous
View PDF

9 Web Application Metrics

This chapter describes the Web Application metrics.

9.1 HTTP Content

The following sections lists the HTTP Content metrics, their descriptions, and user actions.

9.1.1 Average Connect Time

This metric measures the average connect time for all pages in the transaction. This is calculated as: Total Connect Time / Number of Connections Made. The Connect Time is one of the phases of a transaction that can help you isolate and fix response time problems.

User Action

The average connect time, when reviewed over a period of time, can indicate whether network congestion or other connectivity issues are the cause of poor Web application response time.

9.1.2 Average First Byte Time

This metric measures the average First Byte Time for all pages in the transaction. This metric is computed as: Total First Byte Time / Number of Requests Made (either to fetch HTML or content). The First Byte time is one of the phases of a transaction that can help you isolate and fix response time problems.

User Action

The average First Byte Time, when reviewed over a period of time, can indicate whether network congestion or other connectivity issues are the cause of poor Web application response time.

9.1.3 Average Response Time

A single transaction often accesses multiple Web pages. The Average Page Response metric calculates the average response time of the pages within a single transaction. This metric is calculated as: Total Transaction Time / Number of Pages in the Transaction. For example, if the transaction connects to four different Web pages, this metric will calculate the average response time for the four pages each time the transaction is run.

User Action

If a particular transaction continuously exceeds the Average Page Response threshold, use the Beacon Data page to test the transaction from other beacons and over a specific time period. Use this data to pinpoint any trends or specific beacons that generate the alerts. Display the Beacon Data page by clicking the value of a metric on the Transaction Performance Page.

9.1.4 Beacon Name

The beacon name if the name of the beacon for which the current metric data is being collected.

9.1.5 Broken Content

This metric is not currently collected by Oracle Enterprise Manager and is for internal use only.

9.1.6 Broken Count

This metric measures the number of errors encountered when displaying content for the pages accessed by the transaction. For example, missing GIF images or style sheets will increase the value of the Broken Count metric.

User Action

Use this metric to measure the quality of the pages being served by your Web application. For example, high values for the Broken Count metric can indicate that files have been moved or that relative links in the application are broken.

9.1.7 Computed Response Time

This metric represents the estimated response time for a client such as a browser, to fetch all the pages in a transaction. The computed response time is calculated as if the contents of every page (such as images and HTML style sheets) were fetched in parallel using multiple threads.

User Action

Use the Calculated Response Time to predict the response time that will be experienced by your average end user.

9.1.8 Connect Time

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. Connect Time is the first phase of a transaction and represents the time it takes for a connection to the Web server to be established for all requests.

User Action

Slow connect time has nothing to do with the content of the page itself. It is likely caused by a slow network or a busy Web server, which prevents the request from getting to the Web server in a timely manner.

9.1.9 Content Time

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. This metric measures the Content Time, which is the amount of time taken to transfer page content to the browser. Page content includes images and style sheets, as opposed to the HTML coding for the page.

User Action

Consider reducing the size of images or other contents of the page

9.1.10 DNS Time

This metric is not supported for this version of Enterprise Manager

9.1.11 First Byte TIme

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. This metric measures the First Byte Time, which is the total time taken between the last byte of the request sent and the first byte of the response received by the server for all requests made. This includes the network latency and the time for the server to respond.

User Action

As with the Connect Time and Redirect Time, this metric can help you pinpoint whether or not the page content or Web application software is causing the slow response time, as opposed to the actual time it takes to transfer one byte of information to the browser.

9.1.12 HTML Bytes

This metric provides information about the amount of data transferred during the selected transaction. For each transaction, this metric provides the total bytes of HTML code transferred from all the Web pages accessed by the transaction.

User Action

This metric can help you isolate the cause of any performance problems identified by this transaction. Be sure to consider the number of HTML bytes when you compare the response time of your Web Application transactions.

9.1.13 HTML Content

This metric serves as a container for a set of metrics that provide you with information about the content of the Web pages you are monitoring, as well as response time information.

9.1.14 HTTP Response

This metric is a container for a set of metrics you can use to measure the performance of your Web Application transactions. It indicates how quickly the pages respond to user requests.

9.1.15 HTML Time

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. This metric measures the HTML Time, which is the amount of time it takes to transfer the HTML coding of the page to the browser. This metric does not include the time spent transferring images or other page content, for example.

User Action

Slow HTML time could indicate problems with your HTML coding. Check the source code for your Web Application page to see if there are ways to streamline or improve the logic of the HTML code.

9.1.16 Page Content Bytes

This metric provides information about the amount of data transferred during the selected transaction. For each transaction, this metric provides the number of bytes that represent page content such as images and style sheets.

User Action

This metric can help you isolate the cause of any performance problems identified by this transaction. Be sure to consider the total number of bytes when you compare the response time of your Web Application transactions. Pages with many images or complex style sheets will return a high value for the Page Content Bytes metric.

9.1.17 Page Content Count

This metric is not currently collected by Oracle Enterprise Manager and is for internal use only.

9.1.18 Redirect Count

This metric is not currently collected by Oracle Enterprise Manager and is for internal use only.

9.1.19 Redirect Time

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. Some pages automatically redirect the HTTP request to another page. Redirect time represents the total time of all redirects within a transaction. The time taken to redirect the request can affect the overall response time of the page.

User Action

Significant time taken to redirect the HTTP request removes the possibility that the page content or the Web application software is causing a slow response time alert. If the redirect is causing the performance problems, consider alternative solutions to sending the user to another HTML page.

9.1.20 Request Count

This metric is not currently collected by Oracle Enterprise Manager and is for internal use only.

9.1.21 Slowest Response Time

A single transaction often accesses multiple Web pages. This metric indicates the maximum response time measured for a particular page within a transaction. The slowest page response time can be monitored for a specific transaction and from a specific beacon over a period of time.

You can set a threshold for this metric so that Enterprise Manager will generate an alert if the slowest page response for a particular transaction exceeds a value you specify when it is run from a specific beacon.

9.1.22 Status

This metric returns a value of 1 if the selected beacon was successfully able to run the transaction for this Web application target.

User Action

There are several possible causes to a failed transaction. First, check the availability of the Web application and host for the Web application target.

Second, check the availability of the Oracle Agent for this beacon.

9.1.23 Status Description

If the beacon is unable to run the transaction successfully, this metric returns a description of the error that prevented the transaction from running.

User Action

If you are reviewing the metric results from the All Metrics page, review the Value column of the Status Description table. The error description should offer clues about why the transaction failed.

9.1.24 Total Bytes

This metric provides information about the amount of data transferred during the selected transaction. For each transaction, this metric provides the total number of bytes transferred from all the Web pages accessed by the transaction.

User Action

This metric can help you isolate the cause of any performance problems identified by this transaction. Be sure to consider the total number of bytes when you compare the response time of your Web Application transactions.

9.1.25 Total Response TIme

Total transaction time indicates the overall time spent to process the transaction. This includes all the phases of the transaction, including Connect Time, Redirect Time, First Byte Time, HTML Time, and Content Time. This metric calculates total transaction time by assuming all contents of a page are fetched in a serial manner.

User Action

Use the total response time metric to identify problem pages. After you identify a page or transaction that's slow to respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.1.26 Transaction Name

The transaction name is the name of the transaction for which the current metric data is collected.

9.1.27 Transfer Rate

The transfer rate indicates how quickly data is being transferred from the Web server to the client browser. This is computed as: Total Kilobytes Received / Total Transaction Time.

User Action

Slow transfer rate can be caused by network congestion or other connectivity issues.

9.1.28 Web Application

You can use Oracle Enterprise Manager to view performance and availability metrics for your Web applications.

9.2 HTTP Step Group

The following sections lists the HTTP Step Group metrics, their descriptions, and user actions.

9.2.1 [HTTP Step Group] Broken URL Count

This metric measures the number of errors encountered when displaying content for the pages accessed by the step group. For example, missing GIF images or style sheets will increase the value of the Broken Count metric.

User Action

Use this metric to measure the quality of the pages being served by your Web application. For example, high values for the Broken Count metric can indicate that files have been moved or that relative links in the application are broken.

9.2.2 [HTTP Step Group] Broken URL Details

This metric is not currently collected by Oracle Enterprise Manager and is for internal use only.

9.2.3 [HTTP Step Group] Connect Time (ms)

Enterprise Manager breaks down each step group into individual phases. Performance metrics for each phase of the step group can help you pinpoint the cause of a slow response time alert. Connect Time is the total time spent in the transaction connecting to the server. There may be multiple connections made during a transaction. Time spend connecting for requests that result in redirects count as Redirect Time rather than Connect Time.

User Action

Significant Connect Time values are usually caused by a slow network or a busy Web server. Significant Connect Time values may also indicate that there are too many connections made during the transaction. Consider enabling HTTP persistent connections if the application does not already have them enabled.

9.2.4 [HTTP Step Group] DNS Time

This metric is not supported for this version of Enterprise Manager.

9.2.5 [HTTP Step Group] First Byte Time (ms)

Enterprise Manager breaks down each step group into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. This metric measures the First Byte Time, which is the total time taken between the last byte of the request sent and the first byte of the response received by the server for all requests made. This includes the network latency and the time for the server to respond.

User Action

As with the Connect Time and Redirect Time, this metric can help you pinpoint whether or not the page content or Web application software is causing the slow response time, as opposed to the actual time it takes to transfer one byte of information to the browser. A high First Byte Time suggests that there may be high network latency between the agent and the service. Some applications generate an entire page before sending the first byte of that page. For such applications, a high First Byte Time could also indicate that the servers are taking a long time to generate each page.

9.2.6 [HTTP Step Group] First Byte TIme per Page (ms)

This is the First Byte Time divided by the number of pages in the step group.

User Action

A high First Byte Time per Page suggests that there may be high network latency between the agent and the service. Some applications generate an entire page before sending the first byte of that page. For such applications, a high First Byte Time could also indicate that the servers are taking a long time to generate each page.

9.2.7 [HTTP Step Group] HTML Time (ms)

Enterprise Manager breaks down each step group into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. This metric measures the HTML Time, which is the amount of time it takes to transfer the HTML coding of the page to the browser. This metric does not include the time spent transferring images or other page content.

User Action

Slow HTML time could indicate that the application is taking a long time to finish generating each page. Alternatively, slow HTML time could indicate that network bandwidth between the agent and the service is low.

9.2.8 [HTTP Step Group] Non-HTML Time (ms)

This is the amount of time it takes to transfer the non-HTML content such as images to the browser.

User Action

Slow Non-HTML time could indicate that the application is taking a long time to generate images. Alternatively, slow HTML time could indicate that network bandwidth between the agent and the service is low. Consider reducing the number of distinct images in the application.

9.2.9 [HTTP Step Group] Perceived Slowest Page Time (ms)

The amount of time that it would take a Web browser to play the slowest page in a step group. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be.

User Action

Use this metric to identify problem pages. After you identify a page or transaction that's slow to response respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem

9.2.10 [HTTP Step Group] Perceived Time per Page (ms)

The average amount of time that it would take a Web browser to play each page in the step group. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be. Because it is normalized on a per-page basis, Perceived Time per Page is also a good metric for comparing the relative performance of different transactions.

User Action

Use this metric to identify problem transactions. After you identify a transaction that's slow to respond to user requests, you can drill down and analyze each phase of the step group to isolate and repair the problem.

9.2.11 [HTTP Step Group] Perceived Total Time (ms)

The amount of time that it would take a Web browser to play the step group. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be.

User Action

Use this metric to identify problem transactions. After you identify a step group that's slow to respond to user requests, you can drill down and analyze each phase of the step group to isolate and repair the problem.

9.2.12 [HTTP Step Group] Redirect Time (ms)

Enterprise Manager breaks down each step group into individual phases. Performance metrics for each phase of the step group can help you pinpoint the cause of a slow response time alert. Some pages automatically redirect the HTTP request to another page. Redirect time represents the total time of all redirects within a step group. The time taken to redirect the request can affect the overall response time of the page.

User Action

Significant time taken to redirect the HTTP request. If the redirect is causing the performance problems, consider alternative solutions to sending the user to another HTML page.

9.2.13 [HTTP Step Group] Status

Indicates whether the Web transaction was successful.

9.2.14 [HTTP Step Group] Status Description

If the beacon is unable to run the transaction successfully, this metric returns a description of the error that prevented the transaction from running.

9.2.15 [HTTP Step Group] Time per Connection (ms)

This is the Connect Time divided by the number of connections made while playing a step group.

User Action

Slow Time per Connection has nothing to do with the content of the page itself. It is likely caused by a slow network or a busy Web server, which prevents the request from getting to the Web server in a timely manner. Transactions that use HTTPS will typically have a much higher Time per Connection than transactions that use HTTP.

9.2.16 [HTTP Step Group] Total Time (ms)

Indicates the overall time spent in processing the step group. This includes all the phases of the transaction, including Connect Time, Redirect Time, First Byte Time, HTML Time, and Non-HTML Time. This metric calculates total transaction time by assuming all contents of a page are fetched in a serial manner.

User Action

Use the Total Time Metric to identify problem transactions. After you identify a transaction that's slow to respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.2.17 [HTTP Step Group] Transfer Rate (KB per second)

The transfer rate indicates how quickly data is being transferred from the Web server to the client browser. This is computed as: Total Kilobytes Received / Total Transaction Time.

User Action

Slow transfer rate can be caused by network congestion or other connectivity issues.

9.3 HTTP Transaction

The following sections lists the HTTP Transaction metrics, their descriptions, and user actions.

9.3.1 [HTTP Transaction] Connect Time (ms)

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. Connect Time is the total time spent in the transaction connecting to the server. There may be multiple connections made during a transaction. Time spend connecting for requests that result in redirects count as Redirect Time rather than Connect Time.

User Action

Significant Connect Time values are usually caused by a slow network or a busy Web server. Significant Connect Time values may also indicate that there are too many connections made during the transaction. Consider enabling HTTP persistent connections if the application does not already have them enabled.

9.3.2 [HTTP Transaction] DNS Time

This metric is not supported for this version of Enterprise Manager.

9.3.3 [HTTP Transaction] First Byte Time (ms)

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. This metric measures the First Byte Time, which is the total time taken between the last byte of the request sent and the first byte of the response received by the server for all requests made. This includes the network latency and the time for the server to respond.

User Action

As with the Connect Time and Redirect Time, this metric can help you pinpoint whether or not the page content or Web application software is causing the slow response time, as opposed to the actual time it takes to transfer one byte of information to the browser. A high First Byte Time suggests that there may be high network latency between the agent and the service. Some applications generate an entire page before sending the first byte of that page. For such applications, a high First Byte Time could also indicate that the servers are taking a long time to generate each page.

9.3.4 [HTTP Transaction] First Byte Time per Page (ms)

This is the First Byte Time divided by the number of pages in the transaction.

User Action

A high First Byte Time per Page suggests that there may be high network latency between the agent and the service. Some applications generate an entire page before sending the first byte of that page. For such applications, a high First Byte Time could also indicate that the servers are taking a long time to generate each page.

9.3.5 [HTTP Transaction] HTML Time (ms)

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. This metric measures the HTML Time, which is the amount of time it takes to transfer the HTML coding of the page to the browser. This metric does not include the time spent transferring images or other page content.

User Action

Slow HTML time could indicate that the application is taking a long time to finish generating each page. Alternatively, slow HTML time could indicate that network bandwidth between the agent and the service is low.

9.3.6 [HTTP Transaction] Non-HTML Time (ms)

This is the amount of time it takes to transfer the non-HTML content such as images to the browser.

User Action

Slow Non-HTML time could indicate that the application is taking a long time to generate images. Alternatively, slow HTML time could indicate that network bandwidth between the agent and the service is low. Consider reducing the number of distinct images in the application.

9.3.7 [HTTP Transaction] Perceived Slowest Page Time (ms)

The amount of time that it would take a web browser to play the slowest page in the transaction. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be.

User Action

Use this metric to identify problem pages. After you identify a page or transaction that's slow to response respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.3.8 [HTTP Transaction] Perceived Time per Page (ms)

The average amount of time that it would take a Web browser to play each page in the transaction. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be. Because it is normalized on a per-page basis, Perceived Time per Page is also a good metric for comparing the relative performance of different transactions.

User Action

Use this metric to identify problem transactions. After you identify a transaction that's slow to respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.3.9 [HTTP Transaction] Perceived Total Time

The amount of time that it would take a web browser to play the transaction. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be.

User Action

Use this metric to identify problem transactions. After you identify a transaction that's slow to respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.3.10 [HTTP Transaction] Redirect Time (ms)

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. Some pages automatically redirect the HTTP request to another page. Redirect time represents the total time of all redirects within a transaction. The time taken to redirect the request can affect the overall response time of the page.

User Action

Significant time taken to redirect the HTTP request. If the redirect is causing the performance problems, consider alternative solutions to sending the user to another HTML page.

9.3.11 [HTTP Transaction] Status

Indicates whether the Web transaction was successful.

9.3.12 [HTTP Transaction] Status Description

If the beacon is unable to run the transaction successfully, this metric returns a description of the error that prevented the transaction from running.

9.3.13 [HTTP Transaction] Time per Connection (ms)

This is the Connect Time divided by the number of connections made while playing a transaction.

User Action

Slow Time per Connection has nothing to do with the content of the page itself. It is likely caused by a slow network or a busy Web server, which prevents the request from getting to the Web server in a timely manner. Transactions that use HTTPS will typically have a much higher Time per Connection than transactions that use HTTP.

9.3.14 [HTTP Transaction] Total Time (ms)

Indicates the overall time spent to process the transaction. This includes all the phases of the transaction, including Connect Time, Redirect Time, First Byte Time, HTML Time, and Non-HTML Time. This metric calculates total transaction time by assuming all contents of a page are fetched in a serial manner.

User Action

Use the Total Time Metric to identify problem transactions. After you identify a r transaction that's slow to respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.3.15 [HTTP Transaction] Transfer Rate (KB per second)

The transfer rate indicates how quickly data is being transferred from the Web server to the client browser. This is computed as: Total Kilobytes Received / Total Transaction Time.

User Action

Slow transfer rate can be caused by network congestion or other connectivity issues.

9.4 HTTP User Action

The following sections lists the HTTP User Action metrics, their descriptions, and user actions.

9.4.1 [HTTP Step] Broken URL Content

This metric is not currently collected by Oracle Enterprise Manager and is for internal use only.

9.4.2 [HTTP Step] Connect Time (ms)

Enterprise Manager breaks down each step element. Performance metrics for each step element can help you pinpoint the cause of a slow response time alert. Connect Time is the total time spent in the transaction connecting to the server. There may be multiple connections made during a transaction. Time spend connecting for requests that result in redirects count as Redirect Time rather than Connect Time.

User Action

Significant Connect Time values are usually caused by a slow network or a busy Web server. Significant Connect Time values may also indicate that there are too many connections made during the transaction. Consider enabling HTTP persistent connections if the application does not already have them enabled.

9.4.3 [HTTP Step] DNS Time

This metric is not supported for this version of Enterprise Manager.

9.4.4 [HTTP Step] First Byte Time (ms)

Enterprise Manager breaks down each step into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. This metric measures the First Byte Time, which is the total time taken between the last byte of the request sent and the first byte of the response received by the server for all requests made. This includes the network latency and the time for the server to respond.

User Action

As with the Connect Time and Redirect Time, this metric can help you pinpoint whether or not the page content or Web application software is causing the slow response time, as opposed to the actual time it takes to transfer one byte of information to the browser. A high First Byte Time suggests that there may be high network latency between the agent and the service. Some applications generate an entire page before sending the first byte of that page. For such applications, a high First Byte Time could also indicate that the servers are taking a long time to generate each page.

9.4.5 [HTTP Step] First Byte Time per Page Element (ms)

This is the First Byte Time divided by the number of step elements.

User Action

A high First Byte Time per Page suggests that there may be high network latency between the agent and the service. Some applications generate an entire page before sending the first byte of that page. For such applications, a high First Byte Time could also indicate that the servers are taking a long time to generate each page.

9.4.6 [HTTP Step] HTML Time (ms)

Enterprise Manager breaks down each step element. Performance metrics for each step element can help you pinpoint the cause of a slow response time alert. This metric measures the HTML Time, which is the amount of time it takes to transfer the HTML coding of the page to the browser. This metric does not include the time spent transferring images or other page content.

User Action

Slow HTML time could indicate that the application is taking a long time to finish generating each page. Alternatively, slow HTML time could indicate that network bandwidth between the agent and the service is low.

9.4.7 [HTTP Step] Non-HTML Time (ms)

This is the amount of time it takes to transfer the non-HTML content such as images to the browser.

User Action

Slow Non-HTML time could indicate that the application is taking a long time to generate images. Alternatively, slow HTML time could indicate that network bandwidth between the agent and the service is low. Consider reducing the number of distinct images in the application.

9.4.8 [HTTP Step] Perceived Slowest Page Element Time (ms)

The amount of time that it would take a Web browser to play the slowest step element. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be.

User Action

Use this metric to identify problem pages. After you identify a page or transaction that's slow to response respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem

9.4.9 [HTTP Step] Perceived Time per Page Element (ms)

The average amount of time that it would take a Web browser to play each page in a step. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be. Because it is normalized on a per-page basis, Perceived Time per Page is also a good metric for comparing the relative performance of different transactions.

User Action

Use this metric to identify problem transactions. After you identify a transaction that's slow to respond to user requests, you can drill down and analyze each phase of the step group to isolate and repair the problem.

9.4.10 [HTTP Step] Perceived Total Time (ms)

The amount of time that it would take a Web browser to play the step element. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be.

User Action

Use this metric to identify problem transactions. After you identify a step group that's slow to respond to user requests, you can drill down and analyze each phase of the step to isolate and repair the problem.

9.4.11 [HTTP Step] Redirect Time (ms)

Enterprise Manager breaks down each step element. Performance metrics for each step element can help you pinpoint the cause of a slow response time alert. Some pages automatically redirect the HTTP request to another page. Redirect time represents the total time of all redirects within a step. The time taken to redirect the request can affect the overall response time of the page.

User Action

Significant time taken to redirect the HTTP request. If the redirect is causing the performance problems, consider alternative solutions to sending the user to another HTML page.

9.4.12 [HTTP Step] Status

Indicates whether the Web transaction was successful.

9.4.13 [HTTP Step] Status Description

If the beacon is unable to run the transaction successfully, this metric returns a description of the error that prevented the transaction from running.

9.4.14 [HTTP Step] Time per Connection (ms)

This is the Connect Time divided by the number of connections made while playing a step element.

User Action

Slow Time per Connection has nothing to do with the content of the page itself. It is likely caused by a slow network or a busy Web server, which prevents the request from getting to the Web server in a timely manner. Transactions that use HTTPS will typically have a much higher Time per Connection than transactions that use HTTP.

9.4.15 [HTTP Step] Total Time (ms)

Indicates the overall time spent in processing the step. This includes all the phases of the transaction, including Connect Time, Redirect Time, First Byte Time, HTML Time, and Non-HTML Time. This metric calculates total transaction time by assuming all contents of a page are fetched in a serial manner.

User Action

Use the Total Time Metric to identify problem transactions. After you identify a transaction that's slow to respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.4.16 [HTTP Step] Transfer Rate (KB per second)

The transfer rate indicates how quickly data is being transferred from the Web server to the client browser. This is computed as: Total Kilobytes Received / Total Transaction Time.

 

Slow transfer rate can be caused by network congestion or other connectivity issues.

9.4.17 [HTTP Step] URL

This is the URL associated with the step.

9.5 HTTP Raw

The following sections lists the HTTP Raw metrics, their descriptions, and user actions.

9.5.1 HTTP Raw Broken URL Details

This metric is not currently collected by Oracle Enterprise Manager and is for internal use only.

9.5.2 HTTP Raw Connect Time (ms)

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction, step or step group can help you pinpoint the cause of a slow response time alert. Connect Time is the total time spent in the transaction connecting to the server. There may be multiple connections made during a transaction. Time spent connecting for requests that result in redirects count as Redirect Time rather than Connect Time.

User Action

Significant Connect Time values are usually caused by a slow network or a busy Web server. Significant Connect Time values may also indicate that there are too many connections made during the transaction. Consider enabling HTTP persistent connections if the application does not already have them enabled.

9.5.3 HTTP Raw DNS Time

This metric is not supported for this version of Enterprise Manager.

9.5.4 HTTP Raw First Byte Time (ms)

This is the First Byte Time divided by the number of pages in the step, step group, or transaction.

User Action

A high First Byte Time per Page suggests that there may be high network latency between the agent and the service. Some applications generate an entire page before sending the first byte of that page. For such applications, a high First Byte Time could also indicate that the servers are taking a long time to generate each page.

9.5.5 HTTP Raw HTML Time (ms)

Enterprise Manager breaks down each step, step group, or transaction into individual phases. Performance metrics for each phase can help you pinpoint the cause of a slow response time alert. This metric measures the HTML Time, which is the amount of time it takes to transfer the HTML coding of the page to the browser. This metric does not include the time spent transferring images or other page content.

User Action

Slow HTML time could indicate that the application is taking a long time to finish generating each page. Alternatively, slow HTML time could indicate that network bandwidth between the agent and the service is low.

9.5.6 HTTP Raw Non-HTML Time (ms)

This is the amount of time it takes to transfer the non-HTML content such as images to the browser.

User Action

Slow Non-HTML time could indicate that the application is taking a long time to generate images. Alternatively, slow HTML time could indicate that network bandwidth between the agent and the service is low. Consider reducing the number of distinct images in the application.

9.5.7 HTTP Raw Perceived Slowest Page / Page Element Time (ms)

The amount of time that it would take a web browser to play the slowest page in the step, step group, or transaction. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be.

User Action

Use this metric to identify problem pages. After you identify a page or transaction that's slow to response respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.5.8 HTTP Raw Perceived Time per Page / Page Element (ms)

The average amount of time that it would take a Web browser to play each page in the step, step group, or transaction. This is a good metric for setting thresholds because it is the closest active measurement of what the user-experience is likely to be. Because it is normalized on a per-page basis, Perceived Time per Page is also a good metric for comparing the relative performance of different transactions.

User Action

Use this metric to identify problem transactions. After you identify a transaction that's slow to respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.5.9 HTTP Raw Perceived Total Time (ms)

Indicates the overall time spent to process the step, step group, or transaction. This includes all the phases of the step / step group / transaction, including Connect Time, Redirect Time, First Byte Time, HTML Time, and Non-HTML Time. This metric calculates total transaction time by assuming all contents of a page are fetched in a serial manner.

User Action

Use the Total Time Metric to identify problem transactions. After you identify a transaction that's slow to respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.5.10 HTTP Raw Redirect Time (ms)

Enterprise Manager breaks down each transaction into individual phases. Performance metrics for each phase of the transaction can help you pinpoint the cause of a slow response time alert. Some pages automatically redirect the HTTP request to another page. Redirect time represents the total time of all redirects within a transaction. The time taken to redirect the request can affect the overall response time of the page.

User Action

Significant time taken to redirect the HTTP request. If the redirect is causing the performance problems, consider alternative solutions to sending the user to another HTML page.

9.5.11 HTTP Raw Status

Indicates whether the Web transaction was successful.

9.5.12 HTTP Raw Status Description

If the beacon is unable to run the step, step group, or transaction successfully, this metric returns a description of the error that prevented the transaction from running.

9.5.13 HTTP Raw Time Per Connection

This metric measures the average connect time for all pages in the transaction. This is calculated as: Total Connect Time / Number of Connections Made. The Connect Time is one of the phases of a transaction that can help you isolate and fix response time problems.

User Action

The average connect time, when reviewed over a period of time, can indicate whether network congestion or other connectivity issues are the cause of poor Web application response time.

9.5.14 HTTP Raw Transfer Rate (KB per second)

The transfer rate indicates how quickly data is being transferred from the Web server to the client browser. This is computed as: Total Kilobytes Received / Total Transaction Time.

User Action

Slow transfer rate can be caused by network congestion or other connectivity issues.

9.5.15 HTTP Raw Total Time (ms)

Indicates the overall time spent to process the step, step group, or transaction. This includes all the phases of the transaction, including Connect Time, Redirect Time, First Byte Time, HTML Time, and Non-HTML Time. This metric calculates total transaction time by assuming all contents of a page are fetched in a serial manner.

User Action

Use the Total Time Metric to identify problem transactions. After you identify a r transaction that's slow to respond to user requests, you can drill down and analyze each phase of the transaction to isolate and repair the problem.

9.5.16 HTTP Raw URL

This is the URL associated with the step.