Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04Q—SELECTING
  • H04Q11/00—Selecting arrangements for multiplex systems
  • H04Q11/04—Selecting arrangements for multiplex systems for time-division multiplexing
  • H04Q11/0428—Integrated services digital network, i.e. systems for transmission of different types of digitised signals, e.g. speech, data, telecentral, television signals
  • H04Q11/0478—Provisions for broadband connections
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
  • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
  • H04L41/5003—Managing SLA; Interaction between SLA and QoS
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
  • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
  • H04L41/5003—Managing SLA; Interaction between SLA and QoS
  • H04L41/5009—Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
  • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
  • H04L41/5032—Generating service level reports
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
  • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
  • H04L41/5061—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the interaction between service providers and their network customers, e.g. customer relationship management
  • H04L41/5067—Customer-centric QoS measurements
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L12/00—Data switching networks
  • H04L12/54—Store-and-forward switching systems 
  • H04L12/56—Packet switching systems
  • H04L12/5601—Transfer mode dependent, e.g. ATM
  • H04L2012/5625—Operations, administration and maintenance [OAM]
  • H04L2012/5626—Network management, e.g. Intelligent nets
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
  • H04L12/00—Data switching networks
  • H04L12/54—Store-and-forward switching systems 
  • H04L12/56—Packet switching systems
  • H04L12/5601—Transfer mode dependent, e.g. ATM
  • H04L2012/5629—Admission control
  • H04L2012/5631—Resource management and allocation

Definitions

• the present invention relates to a method of managing a telecommunications network, and in particular to a method of monitoring the compliance with service level agreements. More and more telecommunications services are now becoming available to the consumer. In packet switched networks, unlike circuit switched networks, customers are not given a dedicated circuit; their data is statistically multiplexed with data from other sources. Each customer pays for a particular level of service, and it is therefore important to ensure that the customer is receiving the level of service he has paid for. There is a thus need for a system that manages service level agreements (SLAs) between telecommunications service providers and their business customers. Part of the management process that relates to SLAs is the comparison of the service providers' performance vis-a-vis specific guarantees that it may provide to its customer. Such a system must be capable of handling vast amounts of data. A object of the invention is to provide such a system.
• SLAs service level agreements
• a method managing a telecommunications network comprising the steps of maintaining a database containing data relating to service level agreements with customers using an object model, receiving from the network data relating to the performance of the network, comparing the data received from the network with data stored in said database, and generating a report based on said data showing the performance levels for individual customers in meeting commitments stored in said database containing data relating to service level agreements.
• the method may be implemented, for example, on a Sun Sparc Ultra 2 Unix- based workstation and, for example, work in conjunction with a Newbridge Networks Corporation 46020 network manager.
• the event is generated when the discrepancy between performance levels and commitments exceed a predetermined threshold value.
• a PVC permanent virtual circuit
• the Cell Loss Ratio indicates the quality of a specific service by providing a measurement of the amount of data loss by the service provider's network due to various reasons such as network congestion and network failure.
• the Cell Loss Ratio for an ATM PVC based service is calculated using the following formula:
• CLR (( ⁇ Ra + ⁇ Rz ) - ( ⁇ Ta + ⁇ Tz ))/( ⁇ Ra + ⁇ Rz ) * 100
• Rz number of data cells received by the provider's network from side Z
• Ta number of data cells transmitted by the service provider to side A
• Tz number of data cells transmitted by the service provider to side Z.
• Data which relates to Ra, Rz, Ta, and Tz is typically available from the ATM switching equipment in the form of statistical counters that are regenerated every fifteen minutes.
• Various other statistics, provided by the ATM switching equipment, are required for the verification of other service quality metrics.
• a service provider must measure service performance metrics on 50,000 PVC services. Each PVC service generates 2 statistical reports per 15 minute interval ( 1 for each side of the PVC ). Each statistical report consists of an identifier, a time stamp, and 8 statistical counters. Thus in this scenario, the management system must process 9.6 million records (50,000*2*96 intervals ). Furthermore, raw statistical information must typically be available on-line for up to 60 days. Thus the system must manage ( 9.6 * 60) 576 million records or 54 gigabytes of storage space.
• a plurality of working table fragments forming part of a fragmented table are created in memory, data is loaded in successive predetermined time periods into successive table fragments in a predetermined sequence, and the data are processed separately when loaded into the table fragments.
• the data are preferably loaded into said table fragments using a round robin table fragmentation strategy.
• the invention also provides a telecommunications network service level manager comprising a database containing data relating to service level agreements with customers using an object model; means for receiving from the network data relating to the performance of the network; means for continually comparing the data received from the network with data stored in said database, and means for generating a report based on said data showing the performance levels for individual customers in meeting commitments stored in said database containing data relating to said service level agreements.
• the invention still further provides a method of controlling a computer in an object-oriented environment wherein descriptors implemented as an object oriented class are used to store meta information on other classes in the system.
• Figure 1 is a software architecture diagram of a system in accordance with the invention
• Figure 2 is a data flow diagram
• Figure 3 is a database entity relationship diagram for the object model
• Figure 4 shows the entities contained in the SMIB
• Figure 5 shows the entities contained in the HIB
• Figure 6 shows the entities contained in the SIB
• Figure 7 is a software Architecture Diagram showing daily data process flow
• Figure 8 shows the system table fragmentation strategy
• Figure 9 shows a fragmented HIB table and related entries in the DBSpace usage table in the SMIB;
• Figure 10 illustrates the identifying and detaching aged data in a HIB table;
• Figure 11 illustrates the attaching of new data to a table in a HIB and updating the
• Figure 12 shows combinations of start and completion times of states, caused by events in the SMIB;
• Figure 13 shows the object models for a telcom information management architecture;
• Figure 14 illustrates the different types of descriptors that are employed in the system
• Figure 15 shows the top layer object model
• Figure 16 shows the admin entity in Figure 13
• FIG 17 shows the service layer object model of Figure 13.
• the invention implements an object model to efficiently construct the management system capable of handling a large volume of information.
• Database Monitor (ckdbmon) 1 exchanges messages with the system databases, namely the service management information database (SMIB) 2, the historical information database (HIB)3, summarized information database (SIB) 4, and Network Interface Systems Director (the Keep Alive Process) 5.
• SIB service management information database
• HOB historical information database
• SIB summarized information database
• Network Interface Systems Director the Keep Alive Process
• the ckdbmon 1 does not interface directly with the system databases, but with a relational database management system (RDBMS) 7 employed by the system, namely Informix Online version 7.13.
• RDBMS relational database management system
• the Data Management Framework consists of a database monitoring tool, load and unload utilities, and several scripts that employ the load and unload utilities in order to migrate and summarize data between the various databases.
• the monitoring component utilizes the Network Interface keep alive process and the monitoring tool's output is logged in a Network Interface Logging Tool compatible format 8.
• the advantages of doing so are that the logging interface is common between the database and network interface frameworks, and the ability to reuse coded and tested tools is provided.
• the load and unload utilities also use the log tool format to post all operational and alert messages, as do the utility scripts.
• the role of the monitoring tool 1 is to ensure that the system databases do not exceed predefined space utilization thresholds, that the System databases remain active and available to the end users, and that the Informix Online specific event log file (typically called online.log) does not grow too large. Should sections of the System databases become too full (exceeding the threshold), a message is posted, via the log tool, to the System Administrator (not shown).
• the monitor 1 will make several attempts to restart Informix Online and will again post an alert to the System Administrator stating that Informix Online is off-line and it (the monitor) is attempting to restart it.
• the monitor 1 When the Informix specific event log file exceeds the predefined size that the monitor is gauging it against, the monitor 1 will remove log file entries (checkpoint notification messages only), starting with the oldest ones, until the log file again fits within a specified size range.
• These monitoring functions provide the System Administrator with more freedom, as less manual checking of the System Databases status is necessary. Additionally, the monitor promotes greater System database availability as the most common database operation- stopping difficulty, namely running out of space, is monitored and alerts are sent in anticipation of a problem occurring, not just in response to one.
• the Director 5 ensures that the database monitor (ckdbmon) 1 is started and remains active.
• Informix may only be brought off-line by the Administrator explicitly issuing the valid Informix commands (onmode -uy, followed by onmode -ky). Informix is immune to the Director's shutdown commands, and cannot be brought off-line by ckdbmon, so that the System databases may remain on-line, and available, even if the Director or ckdbmon should experience difficulties and shutdown.
• Figure 3 illustrates an example of an object model in accordance with the invention.
• object entities are defined as follows. Service Entities Customer
• the customer is a legally identified organization that is contracting for the supply of one or more services from one or more service providers.
• the contract is a legal administrative and technical document describing what will be provided to the Customer, how and when it will be provided and the terms and conditions under which it will be provided. It also describes the obligations placed upon the Customer.
• Contract Thresholds are SLA thresholds that are associated with a Contract. Child of: Contract via Contract id number.
• a service is anything that the service provider determines that customers wish to purchase and that the service provider is willing to supply.
• a Current Service record contains information on a currently provided service.
• a service is anything that the service provider determines that customers wish to purchase and that the service provider is willing to supply.
• a Historical Service record contains information that was current until the associated customer, contract, or service profile changed.
• a Contact unambiguously identifies a person who carries out a role associated with a specific service entity (Customer, Contract,
• a service component is any network entity that is used within a service offered to a customer.
• Each service can consist of one or more service components.
• the service component will not be related to any service. It may be assigned to a service at a later time.
• a service profile describes the characteristics of a specific service or group of services. Parent to: Current Service, Historical Service, Valid Service
• Service Profile Thresholds are SLA thresholds that are associated with a Service Profile.
• a role identifies all the actions, with respect to the System, that a user in a specific job function is permitted to perform. Parent to: User, Entity Access via Role id number.
• a user is a communicating entity which is registered in the Resolve
• This entry contains references to the user, the action performed, the type of entity operated on, and the time the operation occurred.
• Entity/Operation number Entity Access Control
• This entity is used to define what operations, on what entity, are permitted for a given role.
• One instance of this record is created for each entity to which a specific role has privileges on.
• Valid Entity Operations This entity is used to define valid operation and entity combinations for the Resolve System. These combinations are then used to assign Entity Access to Roles, and to create Audit information. Parent to: Audit Entry, Entity Access Control via Entity/Operation number. Network Entities FR PVC
• Frame Relay Permanent Virtual Circuit Parent to: FR NP, FR CTP via Network Entity id number.
• a FR PVC will have two or more (two in this release) Termination Points. Future services, such as multicast connections will have multiple CTP's.
• Frame Relay Network Performance This entity consists of network performance statistics collected from each FR PVC path end. Parent to: FR PVC Daily NP, FR PVC Monthly NP via FR PVC id number, and Path End id number. Child of: FR PVC via FR PVC id number. FR PVC Daily NP
• FR PVC Daily Network Performance Daily summarization of FR PVC Network Performance.
• FR Monthly Network Performance Monthly summarization of FR PVC Network Performance. Child of: FR NP via FR PVC id number, and Path End id number.
• Asynchronous Transfer Mode Permanent Virtual Circuit Parent to: ATM NP, ATM CTP via Network Entity id number. Child of: Service Component via Network Entity id number.
• ATM CTP Asynchronous Transfer Mode Circuit Termination Point.
• Asynchronous Transfer Mode Network Performance This entity consists of network performance statistics collected from each
• ATM PVC Daily Network Performance Daily summarization of ATM PVC Network Performance.
• ATM NP via ATM PVC id number, and Path End id number.
• ATM PVC Monthly NP
• TDM Nl via Network Entity id number.
• An Event is information describing an occurrence on the network entity for which a report is required.
• FR PVC Daily Quality of Service This entity describes the quality of service provided, for each Frame Relay PVC Network Entity, with respect to availability time, outage time, etc. on a daily basis.
• the QOS statistics are derived from the data contained in the Old State entity.
• FR PVC Monthly QOS FR PVC Monthly Quality of Service. This entity describes the quality of service provided, for each Frame Relay PVC Network Entity, with respect to availability time, outage time, etc. on a monthly basis.
• the QOS statistics are derived from the data contained in the Old State entity. Child of: Old State via Network Entity id number.
• ATM PVC Daily Quality of Service This entity describes the quality of service provided, for each Asynchronous Transfer Mode PVC Network Entity, with respect to availability time, outage time, etc. on a daily basis.
• the QOS statistics are derived from the data contained in the Old State entity. Child of: Old State via Network Entity id number.
• ATM PVC Monthly QOS ATM PVC Monthly Quality of Service. This entity describes the quality of service provided, for each Asynchronous Transfer Mode PVC Network Entity, with respect to availability time, outage time, etc. on a monthly basis.
• the QOS statistics are derived from the data contained in the Old State entity.
• TDM Daily Quality of Service This entity describes the quality of service provided, for each Time Division Multiplexing Circuit Network Entity, with respect to availability time, outage time, etc. on a daily basis.
• the QOS statistics are derived from the data contained in the Old State entity. Child of: Old State via Network Entity id number.
• TDM Monthly QOS TDM Monthly Quality of Service This entity describes the quality of service provided, for each Time Division Multiplexing Circuit Network Entity, with respect to availability time, outage time, etc. on a monthly basis.
• the QOS statistics are derived from the data contained in the Old State entity. Child of: Old State via Network Entity id number.
• This entity is used as an attach point for addition description information relating to the Network Entities. Child of: Service Component via Network Entity id number, Info
• This entity is used to specify the type of service information that can be associated with each Network Entity. Parent to: Service Info via Info Type id number.
• Event Translation This entity maps 46020 events to Resolve events. 46020 Stat Translation
• This entity maps 46020 statistics to Resolve statistics and indicates which statistics should be gathered for the Resolve Databases. 46020 CallAtt Translation
• This entity maps 46020 objects to Resolve Network Entities. This includes the ability to map more than one Network Management System's objects.
• Table Version Info This entity is used to track the version of each physical table in the
• This entity tracks all Resolve archives, both full database backups of the SMIB and SIB, and daily table backups within the HIB.
• This entity keeps track of all the dbspaces available and in use in the Resolve Databases. This entity is used to maintain the large inflow and outflow of data to the HIB.
• the Physical Database Design is the physical, or actual, representation of the Object Model and Logical Database Design. In most instances, the physical design maps quite closely to the logical design, but some deviations may be to achieve greater response performance, or to take advantage of additional features of the RDBMS employed, or to accommodate a lack of required features in the RDBMS.
• the SMIB 2 is an operational data store. It contains both 'soft data' - data (customer, contract, SLA) that can be derived from other Service Provider systems, and data that is in a constant state of flux - Service and Service Component data.
• the SMIB 2 is the definitive source from which to derive inventory and status reports on the Networks, the impact on Service Provider Customers, and the appropriate individuals to contact with respect to Network events.
• the SMIB is enabled as a transaction logging database. That is, any changes made to the SMIB are not only stored in the database, but also recorded in transaction logs that can be replayed in the event that disaster recovery is necessitated, thus the SMIB can be recovered up to its most recent update
• the data contained in the Events, Current State, and Old State entities is only a single days worth. This data is migrated, nightly, to the HIB.
• the SMIB is shown in
• the HIB 3 shown in Figure 5 is a very large store of data. It contains Network Events, the corresponding Network Entity states, and the Network Performance Statistics for all the Network Entities that are currently being tracked (as indicated in the SMIB). By volume of data, the HIB is approximately 40 to 50 times larger.
• the HIB is a data warehouse. It contains very large volumes of data, covering the same Network Entities over a period of time (60 days, in the case of Resolve 1.0), and the data is never updated by end users, or by connecting systems.
• the HIB is NOT a data warehouse from the view that it does not contain data brought together from multiple heterogeneous data sources, but this is a discussion that is of little relevance to this document. Suffice to say, that the HIB contains a very large volume of data that is quite static in nature. The daily Events, Current States, and Old States are migrated to the HIB from the SMIB nightly, and the Network Performance Statistics are loaded, from flat files (created by the
• the HIB 3 does not employ transaction logging, meaning that the HIB cannot be recovered to the most recent point in time. Recovery to the most recent point in time, however, is not necessary as the HIB does not permit user updates against it. Since the only updates are performed by nightly processes, the new data added to the HIB daily is archived to tape by one of these processes. Thus, any disaster recovery may be performed by the Resolve Administrator using the daily data that has been archived to tape.
• the SIB 4 contains the end product of all the data collection and processing efforts. It is here that the end users of Resolve 1.0 may most easily extract meaningful information.
• the processes to create SIB information may be customized to suit a particular Service
• the information in the SIB is static in nature as it is not updated or modified by users or processes. Because of its condensed nature (a single days worth of statistics for one service component equates to 96 records in the HIB, but only one record in the SIB), the SIB can present information covering a broader time period (180 days).
• the Quality of Service entities are derived from Old State data in the HIB, and the
• Network Performance Statistics are summarizations of Network Performance Statistics in the HIB.
• the SIB 4 does not employ transaction logging. It does however, have regular backups of the entire SIB made. In the event of a disaster, the Resolve Administrator could restore the SIB back to its current state by restoring the most recent backup.
• the SIB 4 shown in Figure 6 contains the end product of all the data collection and processing efforts. It is here that the end users of may most easily extract meaningful information. All the information in the SIB is summarized and processed data extracted from the HIB. The processes to create SIB information may be customized to suit a particular Service Provider.
• the information in the SIB is static in nature as it is not updated or modified by users or processes. Because of its condensed nature (a single days worth of statistics for one service component equates to 96 records in the HIB, but only one record in the SIB), the SIB can present information covering a broader time period (180 days).
• the Quality of Service entities are derived from Old State data in the HIB, and the Network Performance Statistics are summarizations of Network Performance Statistics in the HIB.
• the SIB does not employ transaction logging. It does however, have regular backups of the entire SIB made. In the event of a disaster, the Resolve Administrator could restore the SIB back to its current state by restoring the most recent backup.
• the instance of the Director on the server starts a ckdbmon process for each instance of Informix Online that exists on that server. That is, if two Informix Online servers are running on the same workstation (this is a distinct possibility), two instances of ckdbmon will be started. Each instance of ckdbmon will start its own instance of the ckdbmon log tool 8 (ckdblog). Additionally, the cron table is set so that regularly scheduled database jobs are initiated.
• Figure 4 illustrates the use of table fragmentation as it is employed by the System Database Management Framework.
• Data is loaded, from an ASCII delimited flat file into the temporary working table, that has been created in its own dbspace (1).
• One days worth of data is loaded into each dbspace.
• advantage can be taken of its smaller size, relative to the larger HIB data table (1 million rows vs. 600 million rows for stats), and the fact that it contains a single full days worth of data (a day is the smallest unit that the SIB data is summarized on), to perform any summarizations or quality of service (QoS) derivations to be loaded into the SIB on this work table.
• QoS quality of service
• ASCII delimited file is archived in the AIB (2), and this fact recorded in the SMIB, so that this data can be re-examined even after it has been aged and purged from the HIB and SIB.
• this ASCII file contains one full days worth of data.
• the aged data that currently exists in the HIB can be easily removed by simply detaching the dbspace that contains that particular days worth of data from the main table and then deleting that single portion (5). This reduces the time that would be required if the data were to be deleted directly from the main table (the dbspace that has just been cleared of aged data can then be reused as the work dbspace).
• Table fragmentation that is the actual data in a fragmented table can be separated, using one of three methods: 1.
• Expression Based Fragmentation - all data is placed in a dbspace based on a expression involving a column or columns within the table (e.g. a table containing phone numbers may be fragmented using expressions based on ranges of area codes).
• Hash Fragmentation - a unique, or key column in the fragmented table is put through a hashing formula to determine which dbspace that particular record should reside in. 3.
• Round Robin all data that is inserted into the fragmented table is distributed evenly across all dbspaces in the fragmented table.
• An interesting characteristic about the Round Robin method is that any data that is in a dbspace that is attached to a fragmented table (vs. being inserted directly into the fragmented table), is not redistributed to other dbspaces within the fragmented table. This means that as long as the System utilities do not insert data directly into the main, fragmented tables in the HIB, but instead, create work tables and load them and attach them to the main fragmented tables, it is possible to know exactly which dbspace contains a particular days worth of data.
• dbspace usage is a table in the SMIB (the actual name is txd_dbspaceusage).
• This table is used by the utilities loadhib event, loadhib_ostate, loadhib_frnp, and loadhib_atmnp, and it allows these utilities to identify the dbspace that contains data that is a particular age (in the example, 60 days old), to remove that data by detaching the dbspace, and then load the new data into that dbspace, and update the dbspace usage table (Figure 3-4).
• the SMIB is the operational data store of the system Databases. This means that the data contained in it is timely and can change frequently.
• the operational and administrative states of each path, as well as the events that caused a path to be in that state are recorded in the SMIB.
• Event data is stored in the tey_event table, while the current state of each path is stored in the tcs_currstate table. Querying the current state table will allow the user to build up-to-the-minute inventory reports of available paths, and operational reports of path status'.
• the non-current, or 'old' state data is also saved and maintained (this is stored in the tos_oldstate table).
• a database trigger (add_state_rec) creates a new record in the old state table indicating the old states of the path, the time that it originally entered that state, and the amount of time (in seconds) it remained in that state.
• a suspend_collection signal is sent to the events collectors, via the Network Interface Director's interface process called RCI.
• This suspend_collection signal will cause the events collectors to stop inserting new events into the SMIB and instead, store those events in buffers until the previous days worth of events and old state records can be removed from the SMIB.
• a resume_collection signal is sent, via RCI to the Director and, to the events collectors, and normal processing resumes, with any buffered events being processed first.
• Figure 11 is a simple timeline demonstrating the various combinations of events and states (with respect to start and completion times) that may exist in the events table in the SMIB.
• Events 0 and 1 occurred prior to the day that is being processed (October 10, 1996), but the state that events 0 and 1 placed paths 0 and 1 in, respectively, continued into the processed day.
• Event 2 placed path 2 in a state that both started and completed (at Event 2') within the processing day.
• Event 3 occurred in the processing day, but the state of path 3, caused by event 3, continued into the next (the current) day, as did the state that event 0 placed path 0 in.
• the strategy, as implemented in the System Database script unload_smib, is to add a column onto the end of each old state record in the old state tables in both the SMIB and HIB.
• This column called the partial column, contains a 0 if the record has been completed prior to the end of the previous processing day (midnight) - the states caused by events 1 and 2, in Figure 8, fall into this category if processing for October 10, 1996 is being performed (as expected) in the early morning of October 11, 1996.
• the Events table will contain all events that occurred either during the hours of the processing day - Oct. 10 (events 2, 3, 1', 2') or those which were current events as of 12:00am of the processing day (events 0, 1).
• the old state table will only contain records of previous states that were in effect at some time during the processing day. This is why there are old state records relating to events pre2 and pre3 - the states pre2-2 and pre3-3 were in effect as of 12:00am, Oct. 10.
• the following table illustrates the changes that occur in the old state table in the SMIB during nightly processing, as performed by unload_smib.sh.
• the current states of the paths are written to the old state table as partial records with duration's calculated to midnight of the processing day.
• Each table type in the HIB (currently Events, Old States, FR Network Performance, and ATM Network Performance) has a limited number (typically 15) spare dbspaces reserved for aged data. This is data that is older that what is considered active data in the HIB (older than 60 days in this example).
• the following table illustrates how the data in the dbspace usage table is applied to locate a dbspace of type event that is not currently used and is available (indicated by a NULL value in the last usage date column, and the 1 in the available column).
• dbspace usage table available dbspace highlighted
• the actual aged data ASCII delimited files are tracked by another storage management table in the SMIB call the archive information table. This table gets updated every time an archive of HIB data occurs. The type of data (events, old state, etc), the volume label of the tape or file that the ASCII file is stored on, the date of the data, and the version of the table structure (for future use) is recorded.
• a request for the data is made to the System Administrator.
• the System Administrator can then use the restore script to first, select the particular data that was requested, then restore it to the HIB.
• the dbspace usage table is updated to indicate that this data is now in the HIB and the dbspace used is not available for other restores until this data has been removed again from the HIB.
• purge_hib will scan the date values in the SIB to build of list of days of aged data. This is a slower process, but due to the lessor volume of data in the SIB (relative to the HIB) performance should not be an issue. As with purge_hib, purge_sib will delete any data that is as old or older than the days old value passed to it.
• Informix Online provides support for Referential Integrity in the form of Referential Constraints (foreign keys).
• the Rl constraints ensure that a child record cannot be added if the relating value does not exist in the parent table, and conversely, that a parent record cannot be deleted if a child record relating to it exists.
• links. sh - this file creates the symbolic links for the database that are used by Informix Online for dbspace files. This file will require that the devices linked to, and the chunks linked to each device are defined. Additionally, the symbolic links to the tape devices required by the System Databases (two devices - one for log files, the other for archives) must be defined.
• onspace.sh this file creates the dbspaces on top of the symbolic links created by links.
• the onspace commands specify not only the dbspace name and the path of the chunk supporting the dbspace, but also the bytes offset and the size of the dbspace.
• onspace.sh must match the links. sh file. If it does not, the installation will proceed, but the System System will not function as some dbspaces will not be brought on-line due to incorrect space allocation (incorrect offsets).
• 3. dbspaceusage.sql - this file calls the stored procedure, init_dbspaceusage, which seeds the dbspace usage table for storage management. If the system you are installing will store 60 days of active HIB data, this file does not need to be changed. However, if you plan on storing more days or less days of HIB data, the first parameter in each execute procedure statement must be updated to reflect this. 4.
• this file indicates where files are located, names of system tables, and the number of days of HIB and SIB data to be stored. If the number of days of HIB data stored is not 60, or the number of days of SIB data to be stored is not 180, this file must be updated to reflect this.
• the System Database When the System Database is first installed, it is prepared, and expects, to accept data immediately. Conversely, if data is not submitted to the HIB processes for a period of time immediately after the system has been installed, The Data Management Framework processes will automatically perform the necessary alterations to ensure smooth operation.
• Figure 13 is an overview of the object models for a TIM (Telecom Information
• Figure 14 shows the types of descriptors that are provided in a typical system for managing service level agreements. There is a top level descriptor, and derived descriptors relating to various aspects of the system.
• the top level descriptor stores meta information on entities.
• the base class provides a template where unique Ids, names and descriptions are stored.
• the derived classes define additional qualities for specific descriptors.
• the descriptor uniquely identifies the descriptor record attached to a specific class Name:char[200]
• the descriptor name is the unique name of a descriptor Description :char[200]
• This class is used to store meta information on entities.
• the base class provides a template where unique IDs, names and descriptions are stored.
• Derived classes are used to define additional qualities for specific descriptors. The intent behind the Descriptor concept is to minimize the impact of adding new functionality on critical parts of the system. For example, new SLA thresholds can be added to the system without modifying the Service Profile code.
• the descriptorld uniquely identifies the descriptor record attached to a specific class. name : char[80]
• the descriptorName is the unique name of a descriptor, description : char[200]
• Figure 16 also illustrates how descriptors are used to dynamically add new capabilities to the system. As new entities or operations on entities are defined and implemented, new entity operation descriptors are defined and added. When these descriptors are added, the user management module becomes aware of these capabilities. The user management module is used to give/deny access to specific parts of the system.
• Each instance of a descriptor sources its information from a specific row in a relational database table.
• the following table illustrates the instance data associated with the service component descriptors for use in a system for managing service level agreements in a telecommunications network.
• ATM VCC A virtual channel connection. A collection of connections that form an end-to-end path through a network
• VPC virtual path connection
• TDM NI A TDM Network Interface 6 FR CTP Frame Relay Circuit Termination Point.
• New descriptors are added for SLA thresholds, and statistics.
• New valid operations instances are created for ATM UNI. No software is required to change in the user / security management module.
• New service profiles can be created using ATM UNI without any modifications to the software in this critical area.
• New services can be configured with ATM UNI service components assigned to them. Within the configurator, the only software modifications required are new screens to view details associated with the ATM UNI.
• the service definition layer is shown in Figure 17.
• This class is used to store meta information on entities.
• the base class provides a template where unique IDs, names and descriptions are stored.
• Derived classes are used to define additional qualities for specific descriptors. The intent behind the Descriptor concept is to minimize the impact of adding new functionality on critical parts of the system. For example, new SLA thresholds can be added to the system without modifying the Service Profile code.
• the descriptorld uniquely identifies the descriptor record attached to a specific class. name : char[80]
• the descriptorName is the unique name of a descriptor.
• description char[200] The description is a brief summary of the purpose of the descriptor.
• Service entities are the base components of service model, representing the customer, the contracts owned by that customer, the services contained within the contract, and the service profiles desribing the services.
• Entity Rolesf Associations responsibleFor in association Contact_ServiceEntity collectedBy in association ServiceEntityJUserLogEntry
• the deletion of a service entity will be tracked as a modification with the result of the delete flag being set.
• Threshold object A template for a Threshold object that defines the characteristics and description but not the specific value of the threshold.
• An SLA threshold descriptor must be created for each new threshold type that the system will support.
• Attributes defaultValue : double A default value for the threshold. The value can be customized in each instance of the
• This class is used to enumerate the descriptions of functions or job roles that can be associated with a contact person within the system.
• a Function descriptor must be created for each new function that the system will support.
• a Contact unambiguously identifies a person who carries out a role associated with a specific Service Entity.
• the class provides the necessary information to contact that person. * * definition from NM Forum - SMART Performance Reporting White Paper,
• serviceEntityType short
• Contract Thresholds are SLA Thresholds that are associated with a Contract.
• the SLA is a set of technical, administrative and management parameters that the service provider can report against. They typically are based on objective measures and have a high correlation with the users' perception of the quality of service. Each parameter is typically a threshold that, when surpassed, means that the quality test in question has failed. * * definition from NM Forum - SMART Performance Reporting White Paper, September,
• contractType short
• val double The numeric value of the threshold.
• the Customer is a legally identified organization that is contracting for the supply of one or more services from one or more service providers. * * definition from NM Forum -
• Roles/Associations belongsTo in association Customer_Contract ownedBy in association Service_Customer
• a street address for the customer A street address for the customer.
• the Contract is a legal administrative and technical document describing what will be provided to the Customer, how and when it will be provided and the terms and condition under which it will be provided. It also describes the obligations placed upon the
• Roles/Associations maintains in association Customer_Contract isContainedBy in association Contract_Service usedln in association ContractThreshold_Contract Attributes: customerld : integer
• a contract cannot be deleted while services are associated with it. All contact information associated with the contract is deleted when the contract is deleted.
• the relation to a customer is set at creation time and cannot be changed subsequently.
• a Service is anything that the service provider determines that Customers wish to purchase and that the service provider is willing to supply. * More specifically (within the context of Resolve), a Service is a telecommunications product sold to a Customer by a service provider which is managed by the service provider. Services include transmission facilities and associated applications. A Service can consist of multiple Service
• Roles/Associations contains in association Contract_Service owns in association Service_Customer describes in association ServiceProfile_Service containedBy in association ServiceComponent_Service
• customerld integer Foreign key to the customer class. References the customer who owns this service.
• customerType short
• Deletion of this class follows the rules of deletion for the entity class, and leads to deletion of all related contact information.
• This class is used to track current services. Current services become historical services everytime a modification is made to their associations with a customer, contract or service profile record.
• This class is used to track services that are no longer in use. This information is necessary for historical reporting. Everytime the associated customer, contract or service profile for a current service is changed, a historical service is created.
• This class contains pertinent information on a specific individual.
• the described system provides an efficient method of managing service level management agreements in a packet switched network that is capable of handling vast quantities of data in a flexible manner.

Landscapes

• Engineering & Computer Science (AREA)
• Computer Networks & Wireless Communication (AREA)
• Signal Processing (AREA)
• Business, Economics & Management (AREA)
• General Business, Economics & Management (AREA)
• Data Exchanges In Wide-Area Networks (AREA)

Applications Claiming Priority (7)

Publications (1)

Family

ID=27170307

Family Applications (1)

Country Status (3)

Families Citing this family (24)

Family Cites Families (2)

• 1998
  • 1998-03-16 EP EP98912167A patent/EP0968588A1/de not_active Withdrawn
  • 1998-03-16 WO PCT/CA1998/000232 patent/WO1998042102A1/en not_active Application Discontinuation
  • 1998-03-16 AU AU67144/98A patent/AU6714498A/en not_active Abandoned

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events