Electrical computers and digital processing systems: multicomput – Computer network managing – Computer network monitoring
Reexamination Certificate
1998-05-11
2001-11-20
Coulter, Kenneth R. (Department: 2758)
Electrical computers and digital processing systems: multicomput
Computer network managing
Computer network monitoring
C702S186000
Reexamination Certificate
active
06321263
ABSTRACT:
FIELD OF THE INVENTION
The present invention is generally related to the field of network system service, and particularly to an end-user based application availability and response monitoring and alerting system, method and program product. More specifically, the present invention enables the monitoring of availability and response time or other desired performance metrics of an application program from the perspective of an end-user utilizing the application program over a distributed computing network. Additionally, the invention provides a readily accessible reporting system for dynamically communicating the real-time results of the application program monitoring. The monitoring system is implemented such that it is not platform specific, does not impact the performance of the monitored applications, and permits easy maintenance. The reporting system enables real-time analysis of the monitoring results, via graphical display, at a variety of levels of granularity available to any user of the network from a central repository thereon. Additionally, the invention includes facilities for establishing performance thresholds for the application program and for determining when the established thresholds have been violated which may indicate performance aberrations of the program. The invention further provides alert signals indicative of the threshold violation to a service or support entity so as to quickly provide the appropriate services to the poorly functioning application.
BACKGROUND OF THE INVENTION
The pervasive current trend in data processing system design is toward utilization of a distributed computing environment wherein an end-user accesses application programs and data over one or more interconnected networks each including multiple interconnected computers. In a typical distributed computing environment, the desktop computers or network computers used by the end-user community, are connected as clients over local area networks (LANs) to a server, which in turn may connect to other such servers locally or remotely. For example, a business enterprise may maintain several interconnected LANs at each of its geographically separate offices. LAN servers at a given office are each interconnected with one another and are further interconnected over wide area networks (WANs) to the servers in the networks of the remote offices.
Businesses have increasingly adopted this computing model in order to allay the cost of operating, maintaining and upgrading separate isolated “piece-part” computing systems. The interconnected networks characterizing this distributed computing model facilitate the prioritization of applications and data, with mission-critical applications and data residing on high-end, high-bandwidth servers, and less important applications and data assigned to correspondingly lower-end servers. In addition, such a highly distributed processing model will typically incorporate features which ensure that the system will continue to function properly and will be continuously available notwithstanding the failure or maintenance of a single or even multiple servers.
Implementation of such a complex, distributed computing model, while offering numerous advantages for its users, presents correspondingly complex network management problems for its network administrators. Heterogeneous operating systems may be implemented in the interconnected networks. Different applications may be running on separate servers as well as different versions or releases of the same application. Failures occurring over localized or distributed portions of the network are not uniformly reported and accordingly corrective actions may be substantially delayed.
In many instances an information technology services (IT) organization either within or outside of the enterprise is charged with the responsibility for managing the distributed computing environment. Typically, a service level agreement (SLA) with such an agency specifies an expected level of application availability and response time for the users of such a network. Adherence to these expected baseline levels is required to fulfill contractual obligations and the failure to achieve these baselines may directly result in the loss of a customer's business. Accordingly, an application monitoring system which provides real-time data regarding application availability and response time would be an invaluable asset to such an organization.
A number of network management tools have been developed to assist the network manager in monitoring the performance of a distributed computing system. For example, the product known as System Performance Monitor/2 available from International Business Machines Corporation (hereinafter “IBM”, IBM is the present assignee hereof) provides a graphical interface for depicting the performance of various hardware resources in a processing system, however this product does not indicate the availability and response of a software application to an end-user, and does not permit in depth analysis of the results of the monitoring data. The IBM Netfinity(R) Manager software provides network monitoring of server resources as well as operating system resources at the client level, however it also subsists at the server level and does not monitor client-based access to application programs. Accordingly, it does not provide the IT professional with information needed to assess whether the aforementioned baseline levels, many of which are specified from the perspective of an end-user or client of the network, are being achieved.
A number of passive monitoring systems exist for gathering available data from servers and/or clients in a distributed computing system.
For example, U.S. Pat. No. 4,858,152 to Estes for “Operator Access To Monitoring Applications” (issued Aug. 15, 1989 and assigned to the present assignee) teaches a microcomputer-based monitoring system for concurrently monitoring a plurality of host applications running on a mainframe computer, for summarizing the monitored information and for graphically displaying the information on the display screen of a microcomputer system as well as to provide an alarm mechanism for indicating the attainment of user-defined thresholds. The Multiple System Application Monitor (MSAM) taught by Estes receives existing summarized information from the host machine and reduces the information to an accurate picture of the applications running on the host.
Likewise, the U.S. Pat. No. 5,483,468 to Chen et al. for “System and Method For Concurrent Recording And Displaying Of System Performance Data” (issued Jan. 9, 1996 and assigned to the present assignee) teaches a performance monitoring tool for interactive selection of performance statistics across a network. The tool incorporates a data supplier daemon which runs on a server to store statistical information which is selectively supplied to a data consumer program which in turn negotiates the reporting of the desired statistics. One advantage offered by the Chen et al. patent is that the data consumer program need not include any prior information regarding the statistics maintained by the data supplier daemon. The Chen et al. patent provides a mechanism for capturing system data and recording the data for subsequent play-back.
The aforementioned patents, while offering valuable information to a network manager, do not, by themselves, test application availability or response times, but rather they depend upon data being generated by other parts of the system. In the case of Estes, the information is already available at the host for provision to the microcomputer, and in Chen et al., the system statistical data is captured at the server and provided to the data collector. Thus, in both cases these monitoring tools do not generate relevant client-based availability information and are constrained to collecting and reporting pre-existing information on system performance. If no relevant data on application availability and response time from a client's perspective is previously available for these tools, they will not
Luzzi Joseph
Reps Steven M.
Zhu Gengxin
Coulter Kenneth R.
Ehrlich Marc A.
Gonzalez Floyd A.
International Business Machines - Corporation
LandOfFree
Client-based application availability does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Client-based application availability, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Client-based application availability will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2606592