Operations Manager Architecture And Application Performance Monitoring

Operations Manager Architecture

1.a.The operations manager architecture is intended to give a strong establishment to base-level accessibility and execution monitoring, while at the same time setting a base requirement on foundation and resources of personnel. It comprises various components such as (Araujo,2011):

• Management server which runs Data Access Server, Agent manager and consolidator. A management group can comprise of more than one server depending on the environment size of the computing.
• Agent which are used in gathering and investigating data and executing instructions received from the operations manager.
• Database where all the information of event and logic rules are stored. The information is gathered and executed for the group of management and also retains data whose term is short.

The Report Operator profile contains privileges for accessing reports by users. It grants users the permission to view reports in regard to their scope of configuration( Bicer,2011). The user can write, delete, read and update favorite reports, read and run reports among others.

b.The Read-Only Operator profile contains privileges intended for users requiring access on read only. It allows users access views plus view alerts. The user can read tasks of console, alerts, resolution state, events, rules of discovery, enumerate diagnostics results among others.

The Administrator profile contains all rights regarding Operations Manager. The profile of the Administrator includes all privileges contained in all other profiles. The administrator can create, delete and update a state of a resolution, deploy and uninstall an agent among others.

2.a.Agents gather configuration data and assembles data for reporting and analysis, determines the state of health of objects monitored and performs tasks requested.

b.Active directory – Operations Manager enables the use of active directory in assigning agent-managed computers to management groups.

Manually install – In this method, the setup is run manually on the agent or deployed via a distribution tool.

c.Operation console is used in installation of one or more agents. The console can also be used to control agents that have been installed using the Discovery Wizard. It can apply patches, update versions of agents and configure the server of management that the agent reports to.

3.Management pack templates offer Monitoring wizards which allow creation of complete scenarios of monitoring with insignificant input. The wizard makes the required rules, monitors and even focuses to execute the specific situation. You can adjust the setup of the wizard itself on the off chance that you need to change how the monitoring is being performed. The available templates include(Guo,2014):

• Process monitoring template which allows you to monitor if a certain process is running on a computer.
• TCP port template which monitors the accessibility of an application that is available over TCP.
• which monitors the availability of one or more web application URLs and runs these monitoring tests from internal locations.

4.

Management groups are created after installation of operations manager. They comprise of operational database, management server and a database of reporting data warehouse.

• The management server administers the management group and communicates with the database.
• The operational database contains all design information for the management group and stores all checking information that is gathered and prepared for the management group.

Objects are discovered and monitored by use of the following procedure:

• The Operations Manager searches for computers to manage.
• Computers that meet the specified criteria are identified.
• Install agent on the discovered computer.
• Send configuration to agent on request for example, the management server sends the agent the operating system classes to search for operating systems.
• Discover objects. The agent sends back information to the management server that the operating system is present.
• Send monitoring logic to agent. The agent gets all the monitoring logic about the operating system.
• Send monitoring data to the management server.
• As change occurs, monitoring logic is updated.

5.

1. Select Authoring workspace in the Operations console.
2. Select Distributed Applications and then Choose Create a new distributed application.
3. Input a name under the box of Name to name the distributed application.
4. Choose a template regarding point of start under the box of Template for the distributed application.
5. Select the management pack and then click OK.
6. Start the Operations console
7. Click Monitoring, and then select Discovered Inventory.
8. Click Change Target Type then select View all targets.
9. Select Windows Server 2008 Logical Disk, and then click OK
10. Select one of the instances
11. Select Change Target Type.
12. Select View all targets
13. Select Logical Disk (Server), and then click OK
14. Again select Change Target Type. Select View all targets.
15. Select Logical Disk, and then click OK.
16. Select one of the disks and then click the link on the computer name part of the Path Name property shown in the Detail View.

6.

Server side application performance monitoring utilizes WebLOAD which gathers server-side execution information from working frameworks, application servers, web servers and database servers to enable you to recognize the main driver of issues. Server side information is shown continuously as a component of WebLOAD’s outcomes examination reports. WebLOAD’s worked in reconciliation with application execution administration (APM) apparatuses assists you do cause investigation above and beyond(Moran,2012). Additionally, WebLOAD underpins standard APIs through which servers uncover execution information, for example, RSTATD, Monitor, JMX, Windows Performance SNMP, JDBC and SSH. When you distinguish an execution issue utilizing WebLOAD you can change to your APM device and relate the WebLOAD exchange to the server-side occasion. You can precisely distinguish occasions behind bottlenecks and rapidly settle issues.

Management Pack Templates

Client side application performance monitoring enables measuring real time needed by an application to finish a transaction. The following approaches can be used to monitor from a client’s side:

Synthetic Transactions method utilizes simulated transactions so as to gauge the reaction time of an application server and to confirm the got reactions by contrasting them with already recorded reference transactions(Ontaneda,2012). A few test system operators, going about as customers in a system, send solicitations to the application server of intrigue and determine the time expected to finish an exchange. In case the reaction time surpasses a configurable edge or the got server reaction is not correct, the operators advise the manager by creating events.

GUI based solutions enables metering real client transactions and avoids the requirement for getting to the customer application’s source code. As each client ask for the two begins and closures changing a GUI component at the customer side, just watching GUI events convey the required data about begin and end points of client transactions. The software product introduced at customer site gadgets assembles the transaction information of interest from a client’s perspective

7.

Application performance monitoring involves any activity that observes the way an application or a website performs. There are several basic monitoring activities that can be carried out which include(Vechiola,2009)

• Checking the CPUs contained in the servers.
• Rates of errors of applications can also be tracked and monitored.
• Monitoring traffic of the network for identification of slowdowns.

  1.Usage of the analytics of Google for notifying the slow speeds of pages. To perform server side monitoring:

• Set up the Rackspace Monitoring Agent.
• Create a server-side configuration file.
• Update or delete a server-side configuration.

  2.To update the environment, update and save the YAML file, and after that you restart the agent. Using the updated alarms and checks, the target server is updated by the agent. The changes made via the API are overwritten after the agent restarts by the file of configuration

8.

1. Select Library under the console of Service Manager.
2. After Expanding the Library, Select
3. Click Create Group and Select Next.
4. Under the General page, the following can be done:

• Give a name to the given group.
• Type a description for the group.
• Select an unsealed management pack.

1. Click Add and then select a class in the Filter by class.
2. Input the criteria to use in searching in order to find the position of the object and then Select the button of filter.
3. Choose at least one item under the list of the Objects Available, Select Add and then Select OK.
4. Select Next under the page of Members Included.
5. Click Create then close after completion message appears.
6. Select Library under the console of Service Manager.
7. After Expanding the Library select Queues.
8. Click Create Queue.
9. To complete the process, the following is done:

1. Click Next.
2. Input a name under the box of Queue.
3. Select the button of ellipsis right after the box of Work item, Choose a class and then Select OK.
4. Choose the unsealed management pack then click Next.
5. Build the criteria for using in filtering items of work for the queue, and then Select Next.
6. Select Create queue and Close.

9. (Van horn,2012)

1. Click Administration under the console of Operations.
2. Select Settings
3. Select Alerts
4. Select New under the tab of States of Alert Resolutions.
5. In the state in step d), input a name for the state of resolution and choose a variable under the box of Unique ID, Select OK.
6. Under the settings of the Group of Global Management – Alerts, Select OK.
8. Select Packs of Integration under the Manager of Deployment, Select Deploy IP to Server Action.
9. Choose the pack of integration, select Next.
10. Input the runbook server name using installed Designer of the Runbook, Select Add and then Next.
11. Repeat step c severally. Click Next.
12. Select the Schedule installation check box, date and time in the list of installation of Perform.
13. Install Integration Packs.
14. Click Next then click Finish.

10.     

Applications can continue working in a center of data or a private cloud and then blast to form an open cloud in case the number being registered exceeds the limit. This is what is referred to as Cloud bursting. Its advantage is that, payments for the compute resources are only made when needed. Its only recommended for applications that perform highly and are not critical which handle information that is not sensitive. In order to meet request of the crest, an application can burst in the cloud after being conveyed locally or it can be transferred to the overall public cloud so as to make free the assets found nearby for the applications that are basic for the business(Smith,2009).

Cloud bursting performs great for applications that don’t rely upon an intricate application conveyance framework or reconciliation with different applications, parts and frameworks inward to the data center. This means that it does not support applications that depend on each other in the system hence that becomes one of the challenge of cloud bursting.

Another challenge is incompatibility with various environments and also its limited in availability in the management tools. Another issue is the data traffic which comes about in that, all the data has to be copied from the cloud and the data happens to be huge since current applications contain a lot of data. It does not also support complex infrastructure.

References

Araujo, N. S., Dhasmana, A., Giberson, L. G., Monterrubio, A., Parry, J. C., Polonsky, E., & Wahlert, B. (2011). U.S. Patent No. 8,015,563. Washington, DC: U.S. Patent and Trademark Office.

Bicer, T., Chiu, D., & Agrawal, G. (2011, September). A framework for data-intensive computing with cloud bursting. In Cluster computing (cluster), 2011 ieee international conference on (pp. 169-177). IEEE.

Guo, T., Sharma, U., Shenoy, P., Wood, T., & Sahu, S. (2014). Cost-aware cloud bursting for enterprise applications. ACM Transactions on Internet Technology (TOIT), 13(3),

Moran, M., Liubinskas, T., & Goral, J. (2012). U.S. Patent No. 6,801,940. Washington, DC: U.S. Patent and Trademark Office.

            Ontaneda, M., Keane, T. W., Duque, B. B., Bhandari, C., Wright, T., Voloshin, V., & McKinnon, C. (2011). U.S. Patent No. 7,979,893. Washington, DC: U.S. Patent and Trademark Office.

Oustiougov, M., Hough, M. E., Maru, V. D., & Yadav, A. K. (2011). U.S. Patent No. 7,895,574. Washington, DC: U.S. Patent and Trademark Office.

Smith, M. B. (2009). Exchange Server Operations. Monitoring Exchange Server 2007 with System Center Operations Manager, 413-457.

Vecchiola, C., Pandey, S., & Buyya, R. (2009, December). High-performance cloud computing: A view of scientific applications. In Pervasive Systems, Algorithms, and Networks (ISPAN), 2009 10th International Symposium on (pp. 4-16). IEEE.

Van Hoorn, A., Waller, J., & Hasselbring, W. (2012, April). Kieker: A framework for application performance monitoring and dynamic software analysis. In Proceedings of the 3rd ACM/SPEC International Conference on Performance Engineering (pp. 247-248). ACM.