BACKGROUND OF THE INVENTION
The present invention relates generally to the management of a network and, more particularly, to a network that connects a variety of elements.
Networks are aggregates of elements that interact with one another. These elements must communicate across the network with each other, which necessitates that they be identified and understood by the network itself. These elements are generally of dissimilar function, design and manufacture, yet must interact seamlessly with one another across the network. To enable such communication, the network must be provided with the information necessary for the recognition and operation of each element.
One approach to this problem is to provide the network itself with the software instructions for the identification and operation of each of the elements on the network. However, for a variety of reasons, these instructions often are not provided at the top network level, i.e., at the level of the network management system (NMS). For example, the devices may be unable to communicate directly with the NMS due to the lack of Ethernet port capability, or the design of the network may be maximized in a manner that (as often is the case) militates against storing this information directly in the NMS. Network designers may forgo providing the NMS with the often substantial firmware necessary to support simple network management protocol (SNMP). The devices may also be unable to communicate directly with the NMS due to incompatibility with SNMP.
These capabilities can be provided by a network management proxy that is employed to monitor the status of a collection of elements or devices that can not communicate directly with the NMS. A proxy is a device (which can also be implemented as software, firmware, or other forms of code) that acts as an intermediary between the networked elements and the NMS. The proxy is capable of communicating with a network management system and also is capable of communicating with the other elements in a network. The proxy (sometimes referred to as the master element) serves as a translator, or bridge, between the network management system and those elements in the network not otherwise capable of communicating directly with the network management system for any reason. The proxy contains all of the information that the NMS needs to manage the element collection, and provides an efficient approach to managing a fixed, unvarying set of networked elements. However, when a new type of element or a new version of an existing element is added to the collection managed by the proxy, the proxy requires an upgrade in its software or firmware in order to monitor the new or changed element. Such upgrades can be costly in terms of time and money, particularly for networks in which elements frequently are added or upgraded.
- SUMMARY OF THE INVENTION
The present invention is directed to the problem of simplifying and reducing the cost of providing a network with information necessary to the proper operation of a networked collection of elements.
The present invention simplifies the task of providing the proxy with current and accurate information concerning the operation of new or upgraded networked elements by placing this information in the elements themselves. Whenever a new element is added to the network, the proxy interrogates it to determine its device type and, in the case of updates to old elements, its revision. The information necessary for the proper operation of the element and its integration with the network is copied from the element to the network proxy.
BRIEF DESCRIPTION OF THE DRAWINGS
Among the advantages of this approach to the management of a network of elements is that it obviates the need to manually update the network proxy with new software to control new or upgraded elements, and thus also eliminates the attendant investment of time, labor and other support costs imposed by prior approaches. The present invention thus assures that all of the elements are automatically properly integrated into the network at the time they are added to the network.
The above-mentioned features and advantages of the invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings, wherein:
FIG. 1 is a schematic illustration of a distributed element network management system constructed according to the principles of the invention.
FIG. 2 illustrates an embodiment of the control module or proxy of a system constructed according to the principles of the invention.
In a distributed element network management system, a master element or proxy is used to handle interactions between the elements and the network. The managed elements are independent devices that are capable of performing all of their necessary functions without the aid of the proxy. (For the purposes of this discussion, the term “element,” as taken in this context, is understood to include any part of a network that is capable of communicating with a network management system, either directly or via a proxy agent. Examples of such elements are nodes, devices, managed devices, modules, and objects.) The proxy is capable of communicating with the network management system over a chosen medium (e.g., coaxial cable, T1 line, telecom network, etc.). The proxy allows the NMS to manage a collection of elements that otherwise can not communicate with the NMS, either because they are not compatible with the network protocol employed or with the transmission medium used.
In a known system, the proxy contains all of the information necessary to monitor status and control the elements in its domain. In the distributed element network management system of this invention, the proxy contains very little information regarding the managed elements, save the proxy-element protocol for interrogating the elements. Instead, the necessary information is contained in the elements themselves, and is shared with the proxy at the time the element is discovered (see FIG. 1). The proxy communicates with all of its managed elements to determine if any new elements have been added, and obtains from each new element the code necessary for the proper integration of that element into the network. (While it is preferred that the transfer of code be under the control of the proxy, according to an alternative embodiment the managed element itself announces its presence to the proxy and manages the transfer of its code to the proxy.)
When the proxy starts up, it actively attempts to detect the presence of all elements that it is responsible for managing by polling all of its slots. (Alternatively, the proxy may be said to begin by polling all of its ports.) When a new element is discovered, the proxy interrogates the element to determine its device type and revision. In the preferred embodiment, the proxy issues a command that requires the element to respond with identifying information. In this case, the element is required to respond with information that includes the type of element, the revision of its network management process or subagent and its serial number. (The precise mechanism of interrogation may be implemented by any known methods.) If the element is new (either in type or revision), the proxy copies a sub-agent process from the element to itself. The sub-agent process is a piece of software or firmware code originally carried by the device that is intended to execute within the proxy. It contains all of the information and conversion routines necessary to translate commands from the NMS to the native protocol of the element, send the translated request to the element, retrieve a response, build and transmit a response back to the NMS, and otherwise operate as intended on a network. The proxy also uses the sub-agent to monitor the new element, detect the presence of alarm conditions, and generate alarm messages to the NMS (e.g., when the device is removed from the system, when the unit falls, or is otherwise tampered with or altered).
The proxy also transfers all other process agents and files necessary for the control and monitoring of the new element to itself from the element. These agents and files are not limited to network management activities. For example, they may include any menus required for the operation of a front panel control keypad, display, or other interface. Indeed, any file or agent process necessary to operate or interface with the element can be carried and transferred to the proxy as necessary. The agents are designed to run on the proxy; they do not need to be executable on the element in which they are stored.
The code can be stored on the element in a number of ways. For example, it can be stored as binary data in non-volatile memory. The binary data can be the actual binary executable code, an s-record or any other format. Many different known storage formats can be used. The proxy does not need to know where the code is located on the element; it suffices that the element knows where to look when it retrieves the data from memory and sends it to the proxy.
Depending on the amount of memory provided with each element, a number of useful features can be “piggy-backed” onto a given device. For example, a device can be given the code for the operation of devices that have not been designed with this system in mind, so that upon detection of such a device in the network, the proxy could look to a second device for updated code regarding the operation of the first device. In another embodiment, each element or device can be provided with the capacity for carrying two sets of operating firmware, each set of which contains one version of the necessary data and code for network management, each set residing in a bank. When a device leaves the factory, firmware is programmed in the first bank and a second bank is left empty. When an operator upgrades the firmware, the new firmware is stored in the second bank. A third upgrade would be stored in the first bank. The location of each upgrade is always toggled between the two banks, which allows the operator to revert to the last version of the code for any reason (e.g., correcting compatibility problems). Because a new version of code might add a new feature or take one away, each version also preferably carries its own copy of the sub-agent or menu agent or any other files necessary for network management.
FIG. 2 illustrates the use of this invention in the design of the Motorola Corporation's OmniStar® GX2 broadband cable-optics platform, in which the managed elements are individual independent modules or elements, such as application modules and power supplies, that are connected to a proxy by a backplane. Each element contains its own microcontroller and firmware code. The platform is provided with a proxy in the form of a control module (CM), which is a device that is capable of communicating with a NMS over an Ethernet channel via SNMP. The control module monitors each application module and power supply slot to detect the addition of a new module. If a new module is discovered, the control module will examine the new module for its device type and revision. If the control module does not currently possess the necessary agents to manage the new module, the control module commands the new module to transfer the needed agents and files to the control module. In particular, the control module retrieves the version of the sub-agent process from the element during discovery. The CM then compares this identifier to the identifiers of the sub-agent processes that are currently executing in the CM. (These identifiers are kept in a list or table for this purpose.) The control module requests at least two object from the new module: a SNMP sub-agent, and a Menu Agent. Additionally, the CM transfers any other needed files as required, such as bitmaps, Java applets, birth certificates, and manuals.
In the embodiment illustrated in FIG. 2, the SNMP sub-agent is a firmware file written to run on the CM. However, it could be provided in any form of code that can be provided with the element in question. It contains all of the information necessary for the CM to function as a NMS proxy for the application modules. Once it is transferred to the CM, the CM will load the file into its memory and start the sub-agent process. Once it is loaded and running, it will handle communication from the NMS to control or monitor the new module. It also polls each module to detect the presence of any alarm conditions and generate any necessary traps back to the NMS. (Traps are sent back to the NMS to notify it of any alarm conditions or events that occur with or in the managed elements or proxy.)
The Menu Agent is a firmware file used by the CM to provide a menu system so that the operator can control and monitor the device via an optional Shelf Display Unit (SDU). The Menu Agent receives user inputs from the SDU, translates them into a menu structure and feeds back a display message to the operator. The Menu Agent also handles the reading or writing of any data to the module it was loaded from
When a new element is introduced, the network operators need not concern themselves with the task of manually upgrading the proxy to accommodate the new elements, since the invention accomplishes this automatically.
Although various embodiments are specifically illustrated and described herein, it will be appreciated that modifications and variations of the present invention are covered by the above teachings and within the purview of the appended claims without departing from the spirit and scope of the invention. For example, the NMS could dynamically load all of the relevant data and management processes from the proxies or other directly managed elements as described above.