ZERO TOUCH INVENTORY (ZTI) AS PART OF OPERATION SUPPORT SYSTEM (OSS) IN FUTURE NETWORKS

INVENTAIRE AUTOMATIQUE ZERO TOUCH (ZTI) POUR SYSTEME DE SOUTIEN DES OPERATIONS DE RESEAUX FUTURS

English Abstract

Operation Support System (OSS) is a set of programs that help a communications
service provider
monitor, control, analyze, and manage problems with a telephone or data
network. Inventory Management
Systems such as Granite's Xng or Cramer's Dimension helping service providers
to keep track of their
assets in their network, to improve their service delivery and last but not
least to provision a new service
for customers. Standard products in Inventory space require an
implementation/modeling-phase of an
existing network any where between 6 to 9 months. Also an accuracy of data is
not always guaranteed,
since there is always a discrepancy between the modeling and the planning
organization.
In this invention the inventory module is not to be fed with data from
existing network. It rather scans the
network and poll the information out of the network. Once the information is
gathered from the network,
some smart algorithms needs to be run, in order to put the data together and
get a meaningful inventory
system. Since the data is always updated from network (network driven update
vs. manually), the data
stored is always accurate.

Claims

The embodiments of the invention in which an exclusive properly or privilege
is claimed are defined as
follows:

1. A process for Operation Support System (OSS) auto-reconciliation of network
inventory system
without manual interference of human being for IP and MPLS based networks. In
this process not only the
hardware data can be gathered and modeled, furthermore the logical services
can be inventoried. The
modeling phase does not require an implementation phase (Zero-Touch-Inventory)

2. A process as defined in claim 1, in which the inventory data can be
gathered without manual
interaction, supports for an on-demand-reconciliation of the modeled data and
real data in the network
(network driven vs. modeled).

3. The process as defined in claim 1 will be supported by the Rule Based
Engine. It is an algorithm
based on fuzzy-logic pattern recognition, which is able to automatically
consolidate the network driven
gathered data and build the inventory data.

Description

CA 02451947 2003-12-15
Right now the two biggest OSS inventory leaders of the market Granite and
Cramer need more than 6
months to implement their projects in Tier-1 environment. Most of the time is
being consumed by data
validation and take over of existing data in the new environment. After the
implementation the data needs
to be kept accurate. For this purpose most of the companies have such a
synchronization tool, which tries
to keep the data up to date (ca. 70%). Other companies are able to feed the
inventory system with
accurate data out of network with regards to
Chassis
Cards
Ports
Power supplies
Fans
Hardware, software and firmware
Logical service components .
Hereby the emphasis is still the physical asset management. The support for
logical service is built
rudimentary. In reality such an extended project for auto-reconciliation would
take another 3 months to get
implemented.
The described scenario would apply for old fashioned SDHISONET networks. As
most companies are
moving toward IP based network in a fully dynamic environment, the mentioned
scenario is getting more
complex. The modeling phase is not a one-time approach, rather than ongoing.
The auto-reconciliation is
ideally being run once-a-day.
The Zerp-Touch Inventory is a fully automated, 1009 zero-error-prone approach,
to model the inventory
data based on the data fed directly by the network. There is no human
interaction required in order to
keep the data loaded out of the network(s). The data will be gathered and then
compiled in order to
generate new data e.g. to see who is running VoIP from node A to node Z. This
approach is not only
covering ISO/OSi model Layer1 to Layer 3, but furthermore the application
layer. As situation changes
rapidly in dynamic IP environment, a smart rule based engine needs to be
developed to compile the data
constantly and provide the user with up to date and accurate data. The rule
based engine needs to be self
configurable in order to minimize the maintenance cost.
Zero-Touch inventory ultimately will provide the user with an exact snapshot
of the current configuration of
the network(s).
The sources to load data out of networks) could be the following:
- Telnet module
o Queries data from IOS-based Cisco equipments
- Corba module
o Queries data from Cisco VPN SC
o Receives notifications from Cisco VPN SC
- SNMP module
o Queries data from SNMP-based equipments
o Receives traps from equipments
HP OpenVisw
Once the data are loaded and compiled into useful inventory data, a complete
path consisting of
equipment, links in between can be displayed and visualized. The following
example explains the goal of
the invention. In the frgure 2 a simple network consisting of 4 Cisco 15414
network is displaced.
Furthermore we assume that Cisco 7502 Router is being used to transfer IP
networks directly this
backbone. In this case the Zero-Touch-Inventory module v~rould capture the
equipment with regards to
chassis, cards, ports. The MPLS LSP (Label Switching Paths, the logical
services) are then gathered in
the next step. Once all the information is gathered, the Rule Based Engine can
be run to sort the data and
build the inventory output data. The data flow is being shown in the figure 3.