The invention described herein may be manufactured and used by or for the Government of the United States of America for governmental purposes without the payment of any royalties thereon or therefor.
This patent application makes reference to a pending U.S. patent application entitled UNIVERSAL CLIENT AND CONSUMER, application Ser. No. 10/263,295 by the same inventors as this application.
(1) Field of the Invention
The present invention relates in general to computer networks, and more specifically to a method and apparatus that provides a display of the current operational status of all the data sources connected to a computer network.
(2) Description of the Prior Art
A common data processing system design utilizes a distributed computing environment where data is managed and disseminated over two or more interconnected computers. Such an interconnection of computers with the ability to communicate information between the computers is known as a computer network, or simply a network. Such networks may include a large number of components, including various types of computers and peripheral devices, which may be configured in a variety of ways and may be characterized as “Local Area Networks” (LANs) or “Wide Area Networks” (WANs) by the geographic area over which the components are distributed. LANs and WANs often employ several common configurations or architectures. For example, in a generic client/server configuration a certain number of the interconnected computers may function as clients while others may function as servers that provide services to the clients. Such a client/server configuration is a common example of the several available configurations of distributed computing environments (LANs and WANs) and is well known by those skilled in the art.
One manner of implementing software applications to run on a LAN or WAN is to use a vendor-independent network software architecture and infrastructure that various heterogeneous software applications can use to work together over the network. Such an implementation can be achieved using the “Common Object Request Broker Architecture” (CORBA) Specification. CORBA is a vendor independent specification for an architecture and infrastructure that promotes interoperability. It integrates computers from different vendors ranging in size from mainframes to desktops. CORBA provides a software bus that enables system applications to exchange and communicate information where such applications typically are distributed across a LAN or WAN. A CORBA based system relies on data abstraction to permit software applications running on the system to function unconstrained by the underlying network details, such as the types of workstations, the types of operating systems, and/or the programming languages of other application implementations. One of CORBA's most important, as well as most frequent uses is in network server applications that must handle a large number of clients, at high hit rates with high reliability. Applications utilizing CORBA are typically implemented and configured as either providers of data (servers and suppliers) or users of data (clients and consumers).
A large-scale distributed software system utilizing CORBA based architecture could have hundreds of servers/suppliers. In a large-scale system it may not be immediately obvious to a system manager whether all of the servers/suppliers are functioning properly or if a single server/supplier out of hundreds has failed. A number of software tools presently exist for gathering a variety of data from servers and/or clients in a distributed software system.
For example, U.S. Pat. No. 5,226,120 to Brown et al. for “Apparatus and Method of Monitoring the Status of a Local Area Network” (issued Jul. 6, 1993) teaches a monitoring and status displaying apparatus which determines the topology of a LAN by identifying and isolating each hub in the LAN and identifying the types of modules and the physical locations of the modules in the hub so that an image of the actual hub can be displayed on the video screen of the control console. The actual hub image shows the location and types of modules installed in the hub. However, this apparatus is oriented towards the managing and ascertaining of the specific physical layout of the hardware that composes the LAN. It does not monitor the processes running on the separate data terminal devices to determine if they are providing data appropriately nor does it display this information as status.
U.S. Pat. No. 5,675,798 to Chang for “System and Method for Selectively and Contemporaneously Monitoring Processes in a Multiprocessing Server” (issued Oct. 7, 1997) teaches a system for displaying the status of specific client applications running on a single multiprocessing server. The invention, by isolating and focusing on a specific client process, provides information of a granularity to identify processes which are hung up on semaphores, messages, queues or the like. The system employs a monitoring computer program running on the multiprocessing server which links control blocks to every process and displays information regarding the processes on a video display. Although this invention can provide a display of the status of a particular client process running on the server, it does not provide a system wide display of all servers and suppliers at once. There is no opportunity to obtain a complete system wide picture of which servers are operating and which are not. Also, this invention does not employ existing CORBA resources to achieve its results, but rather needs to employ additional performance monitoring code.
U.S. Pat. No. 5,939,999 to Ohgaki for “Polling Method and Apparatus for a Digital Processing System” (issued Aug. 17, 1999) teaches a digital processing system, such as transmission apparatus used in a digital communication system which employs a supervisory control unit and a plurality of controlled units each performing various processes. The supervisory control unit must perform a polling operation to each controlled unit to collect status information and determine whether any one of the controlled units is not mounted or has failed. Although the Ohgaki apparatus permits a considerable reduction of polling time over the prior art between the supervisory control unit and the plurality of controlled units, there is no display of such information to any type of graphical user interface.
U.S. Pat. No. 6,070,190 to Reps et al. for “Client-Based Application Availability and Response Monitoring and Reporting for Distributed Computing Environments” (issued May 30, 2000) teaches a method, system and program product for monitoring, from a client computer system, the performance of an application program residing on a server computer system connected to a LAN or WAN. A probe program residing at the client computer generates requests for the services of the application program and records transaction records based upon service responses from the server. The transaction records contain information such as whether a service response was made, the length of time to receive a response and other desired performance metrics. This information is stored in a database for statistical analysis and is also displayed on a graphical user interface. The goal is to monitor desired performance metrics of software applications running on servers from an end-user's vantage point. Although this invention can provide a display of the status of a particular process running on a server, it does not provide a system wide look at the status of all servers. Furthermore, it employs additional performance monitoring code in the form of a probe program incorporated into each client computer.
The above-described patents, while offering valuable information, do not by themselves provide the system manager of a large-scale distributed software system with a method and apparatus for easily and continuously ascertaining the system wide state of all of the servers/suppliers in real time. With such a method and apparatus, a system user might avoid a situation where he/she is unaware that a server or supplier is experiencing a loss of functionality until an undesired event occurs. System testers and maintenance personnel would also benefit from such a method and apparatus by quickly ascertaining the baseline status of the system and then quickly detecting and localizing a system problem to a particular server/supplier. In addition, it would be advantageous to provide such a method and apparatus without requiring additional performance monitoring code to be incorporated into each server/supplier application and without requiring a Simple Network Management Protocol (SNMP) agent to be running on all applicable hosts and an available system manager to oversee the system wide monitoring.
There is currently no method and apparatus which can provide a “quick look” status of all of the data sources of a data processing system utilizing CORBA architecture on a LAN which avoids incorporating performance monitoring code into each server/supplier application, avoids having an SNMP agent running on every host, enables a system operator or manager to determine the availability of system data, allows testers to determine the configuration of the system for a particular test and provides maintenance personnel a debugging capability to assist in troubleshooting of problems. What is needed is a status display tool for a multi-unit computer based network hosting a large-scale distributed software system.
It is a general purpose and object of the present invention to provide a method and apparatus that provides a display of the current operational status of all the system data sources in a computer network to a network user.
It is a further object that the display graphically depict the server/supplier availability for the entire system in real time.
It is still a further object to provide a display of the current operational status of all the system data sources through a graphical user interface viewable at any network video connection.
It is yet a further object that the operational status data be color coded to depict various states where, for example, one color represents the “data available” state and another represents the “invalid reference state.”
Another object is that an audio signal be generated when a server/supplier transitions from an available state to an unavailable state.
Still another object is that the operational status information be periodically stored in a database for long-term statistical analysis.
These objects are accomplished with the present invention by creating generic client/consumer CORBA Interface Definition Language (IDL) interfaces to emulate all of the consumers or clients on the network. The generic consumer interface determines the state of suppliers that generate consumer specific periodic data by simply verifying that said data is being received at the specified periodicity. For the purposes of this invention, suppliers are considered to include the implementation of the standard Object Management Group (OMG) CORBA Object Services (Cos) Event Service. With aperiodic data suppliers an embedded “heartbeat” signal is detected in order to determine supplier availability. Similarly, the generic client interfaces periodically invoke an operation on their respective servers to ensure their availability. Based on the operation of the generic interfaces, the status of each server and supplier is displayed in the appropriate format through a graphical user interface.
A more complete understanding of the invention and many of the attendant advantages thereto will be readily appreciated as the same becomes better understood by reference to the following detailed description when considered in conjunction with the accompanying drawings wherein:
The present invention is designed for use in a client-server consumer-supplier network such as the one depicted in
A high-level block diagram of the status display tool is shown in
Associated with each indicator is a “detailed status” text area 410 that adds amplifying detail on the exact status such as “data available,” “data out of range,” “no object reference in the Name Server,” etc. For example, if the server/supplier is yellow, the detailed status text area could state “Server # xxx data is out of range”. In addition, an optional audio signal is generated when a server/supplier transitions from an available state to an unavailable state.
The control/analysis algorithm 110b for suppliers is depicted in
In order to verify data timing and range, the tool utilizes system design data that has been predetermined and stored in the systems configuration files 130. Depending on what data can be verified, the algorithm 110a determines the appropriate color for each supplier.
In the same manner as with the suppliers, the control/analysis algorithm for servers 110a implements a corresponding client interface 50a for each server 80 in the system and performs the same analysis. However, in the case of servers, the timing data verified is the amount of time it takes the client interface 50a to complete an invocation for services and then receive the service.
The advantages of the present invention over the prior art are that the status display tool provides a novel approach for determining the status of all CORBA data sources on a network in an automated and time saving fashion, rather than having a network manager individually manually test hundreds of servers and suppliers one at a time. The method and apparatus provides significant advantages over prior art, in that the desired status information is determined unobtrusively without having the servers or suppliers know that the information is being obtained. The method and apparatus does not use any external probes or monitoring process because but rather employs the existing available CORBA resources making it simpler and more efficient than the prior art.
What has thus been described is a method and apparatus that provides a display of the current operational status of all the data sources in a computer network to a network user. The display graphically depict the server/supplier availability for the entire system in real time through a graphical user interface viewable at any network video connection where the operational status data is color coded to depict various states. The operational status information is periodically stored in a database for long-term statistical analysis.
Obviously many modifications and variations of the present invention may become apparent in light of the above teachings. For example, implementation and use of the invention could be tailored to a closed network as on a ship, or a widely disbursed network like the Internet. The colors on the GUI display may vary, as may the detailed status messages. The status data provided by the invention can then be provided to a system manager and integrated in with network status to provide a complete picture of both the system hardware and software.
In light of the above, it is therefore understood that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described.
| Number | Name | Date | Kind |
|---|---|---|---|
| 5226120 | Brown et al. | Jul 1993 | A |
| 5652839 | Giorgio et al. | Jul 1997 | A |
| 5675798 | Chang | Oct 1997 | A |
| 5939999 | Ohgaki | Aug 1999 | A |
| 6070190 | Reps et al. | May 2000 | A |
| 6799147 | Balasubramanian et al. | Sep 2004 | B1 |
| 7113934 | Levesque et al. | Sep 2006 | B2 |
| 7185075 | Mishra et al. | Feb 2007 | B1 |
| 7260818 | Iterum et al. | Aug 2007 | B1 |
| 20030133556 | Naik et al. | Jul 2003 | A1 |
| 20040064820 | Bussiere et al. | Apr 2004 | A1 |
| 20040085345 | Galou et al. | May 2004 | A1 |
| 20040250260 | Pioso | Dec 2004 | A1 |