Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Reexamination Certificate
1997-12-31
2001-07-17
Beausoleil, Robert (Department: 2184)
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
C714S022000, C702S184000
Reexamination Certificate
active
06263458
ABSTRACT:
SPECIFICATION
This invention relates to telecommunications systems and more particularly to a performance monitoring system and method for use on telecommunications systems.
BACKGROUND OF INVENTION
Telecommunication equipment is generally divided into two main categories. One category is the telecommunication switch. The traditional switch routes numbers dialed to a particular circuit which, in turn, is routed to another circuit, continuing in an iterative process. Each stage, to some degree, constitutes a switch. Variations on the switch include the POTS (plain old telephone service) system and newer technology systems such as ATM (asynchronous transfer mode) which dynamically determines on which circuit to route the calls.
The other category of telecommunication equipment is transmission equipment. This includes high bandwidth copper wire or fiberoptic cable intended to transmit data over a long distance or to directly interface with other transmission equipment without additional switching. For example, in the U.S. and Canada, modern transmission equipment generally transmits at STS1, OC3, OC12 and OC48 speeds. Older transmission equipment transmits at DS1 and DS3 speeds. The connections of transmission equipment are not dynamic, but rather are provisioned as fixed ahead of time such as the connections between (or within) telephone companies. In modem telecommunication systems, software is used to establish these permanent connections between two geographical points.
The present invention can be applied to any of these systems including other connectionless networks such as local area networks (LANs) or the Internet, as well as to cable technology.
For most telecommunications equipment that have interfaces, such as switches, the interfaces are composed of copper wires and/or fiberoptic cables. The physical wires or cables (lines) can be defective, suffering from problems such as environmentally induced noise, hardware malfunction or complete severing. Furthermore, the traffic (the one or more paths carried within the line) being transmitted may be defective with respect to bandwidth. For example, problems can occur on a line which carried a particular problematic path which was itself multiplexed together with other paths from their respective lines to form the aggregated traffic contents of a higher bandwidth line.
Performance Monitoring (or Performance Management) (“PM”) provides quality assurance to telecommunication system operators by allowing a piece of telecommunications equipment to assess its own health and the health of the traffic that is flowing into and out of it. PM is accomplished by monitoring and measuring certain kinds of quantitative operational data which reveal the quality of the paths and lines. This operational data can generate “PM data” which can be monitored by employees located at the switch or can be transmitted to a central hub of network operations where other employees oversee the entire network. PM data permits an operator, whether in a central office or worldwide, to diagnose a failure as it occurs or predict the onset of failures (nonfatal increases in the occurrence of a problem over a period of time) and execute steps toward preventative or remedial maintenance of the equipment. Sources of PM data include dedicated hardware application cards located at points of entry of traffic across the copper wire of fiberoptic cable and UNIX processes executed as part of the telecommunications process.
There are approximately 200 identical instances of PM data, including severely errored seconds, defects that have actually occurred, number of failed tests, laser temperature measurement or consumption of pools of resources. The occurrence of any of these instances detected by monitoring operational data creates an event. An event is a defect, failure or anomaly in transmission which causes the occurrence of a severely erred second or a spontaneous condition. The problems could be hardware (e.g., for POTS ) where PM data is focused on software (e.g., for new technology switches) where traffic metering and monitoring (TMM) data is focused. While there are several differences between PM data and TMM data, for the purposes of the present invention, PM data is meant to also encompass TMM data unless stated otherwise.
Because modern telecommunications systems exist in multi-vendor and/or multi-product environments, a telecommunications software system must be designed to operate in such an environment. Consequently, PM system activities and requirements are highly standardized due to the required compatibility between diverse products. Standards organizations such as the International Telecommunication Unit (ITU), the American National Standards Institute (ANSI) and Bell Communications Research (BELCORE) set forth standards which incorporate PM.
PM data is acquired by using several types of functionality. The vast majority of the PM data is acquired by an accumulator function. An accumulator function's value varies unidirectionally. For example, an additive accumulator function begins with a value of zero and is incremented or increased by a value (including zero) at each instance of an event. Conventionally, at each change of value, the accumulator function checks for whether the value has crossed a threshold, after which certain actions may be taken, including the emission of a threshold crossing alert (TCA). A minority of the PM data is acquired by a gauge function. A gauge function's value varies bidirectionally. The value fluctuates straight up and down depending on what is being measured on the transmission line. For example, gauge functions can reflect a laser temperature meter for measuring the power output of a fiberoptic cable. When the value of a meter becomes too high (“too hot”), an overload problem of some kind may be occurring. When the value becomes too low (“too cold”), there may not be enough power to drive the signal through the cable over the distances required. In either case, when a gauge function's value moves outside of the tolerable range, an onset threshold is crossed. When the value of the gauge function returns to the tolerable range, an abatement threshold is crossed. Thus, onset signifies entering some state. Abatement signifies crossing a threshold that returns the value to a previous state. The gauge function inherently checks for these threshold crossings.
These and other PM data acquiring functions are packaged together, conventionally, as one product-dependent software system by PM designers. Furthermore, PM data collection and storage functionality is also bundled into the conventional PM system package. While this satisfies the necessity for custom-tailored solutions, it results in much time and expense in designing, writing and debugging PM software products.
One of the standards for fiberoptic telecommunications is synchronous optical network (SONET). The SONET PM standards (GR-253-CORE SONET Transport Systems: Common Criteria) describe a set of state machines. One of the requirements of the SONET specification, with regard to the accumulator function, is that a network element support performance monitoring which permits the discarding of PM data accumulated in a ten second window of a monitoring interval (“the discard standard”).
There are two approaches which satisfy the discard standard. One is more complex and harder to achieve than the other. The first easier approach actually permits “fudging” or falling short of discarding the unwanted data by clearing certain registers and counts. In essence, the requirement is not fulfilled in this option. It follows that the majority of PM software systems implement the less complex, easier to achieve standard approach. However, this approach results in PM software systems that are less accurate than systems that achieve the higher standard.
The second, more complex approach is dictated by state machines in SONET that imply an undo operation. There are basically two ways of implementing an undo operation. One method incorporates maintaining the previous state
Haney Scott J.
Kadali Venkat R.
Miller Daniel L.
Weldon, Jr. Richard S.
Alcatel USA Sourcing L.P.
Baker & Botts L.L.P.
Beausoleil Robert
Bonzo Bryce P.
LandOfFree
Regulated push method of data collection does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Regulated push method of data collection, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Regulated push method of data collection will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2454597