Case study · Telecommunications

Celine – unified inventory for Orange networks

Four Orange affiliates kept their inventory data in different tools, from NetCracker to Excel spreadsheets. CELINE brings every source into one common data model for GNOC, the Global Network Operation Center.

The same approach fits: Network and telecom projects Critical infrastructure Photovoltaics and energy Automated convenience retail Logistics and distributed operations

1Different tools, different formats

Inventory tool
NE_NAME    NE_TYPE   LOCATION
BA-RTR-014ROUTERSK-BA-142
BRU-FW-002FIREWALLBE-BRU-07
KE-RTR-003ROUTERSK-KE-031
Excel spreadsheet
ABC
1DeviceCityState
2Switch 07Warszawaok
3Switch 11Krakówok
4OLT 2Gdańskcheck
NMS export
{"host":"mad-olt-03","kind":"olt","site":"MAD-04"}
{"host":"bcn-rtr-21","kind":"router","site":"BCN-01"}
{"host":"lgg-sw-05","kind":"switch","site":"LGG-03"}

2Data pumps

3One common data model

CELINE
CountrySiteElementTypeStatus
PLGDN-05OLT-002OLTMaintenance
SKKE-031RTR-003RouterMaintenance
ESBCN-01RTR-021RouterActive
PLKRK-02SW-011SwitchActive
BEBRU-07FW-002FirewallActive
ESMAD-04OLT-003OLTActive
PLWAW-12SW-007SwitchActive
SKBA-142RTR-014RouterActive
Illustrative records
  • 4Orange affiliates’ inventories in one system: Slovakia, Poland, Spain and Belgium
  • 1common data model able to describe any inventory situation
  • BeforeEach country kept its own tool and its own format.
  • AfterOne common data model: the one place GNOC looks for inventory data.
About the project

GNOC had to work with whatever tools each country offered.

Koderia (formerly Objectify) has been part of CELINE from the very beginning. Our first contribution was to define the problem GNOC was facing.

CELINE was designed for GNOC, which maintains the network infrastructure of Orange affiliates. To do that, access to information about the network infrastructure is paramount.

Originally, GNOC worked with whatever tools each country had in place. It was clear early on that this was not efficient and that GNOC needed a single inventory solution integrating all available data sources.

Client
Orange – GNOC (Global Network Operation Center)
Industry
Telecommunications
Delivered
Umbrella inventory management system
Competencies
Data integration, data modeling, microservices, distributed applications, high data volume processing
Technologies
JavaSpring StackPostgreSQLApache LuceneVueVuetify

Challenge · Information scattered across many different data sources

  1. Data diversity

    Sources ranged from huge inventory solutions like NetCracker to Excel spreadsheets.

  2. Data synchronisation

    Data from each source had to be collected and transformed into one common model.

  3. Quality and usable information

    Users must always see correct data, and get the information they need, not just raw data.

Solution

A common data model and a data pump for every source

Koderia had worked on telecom inventory management long before CELINE and brought that know-how into the design from day one.

Data sources

Each affiliate kept its own inventory tools, and they could not be replaced: local engineers use them every day and they are wired into other IT systems. CELINE had to work with all of them.

Map of Europe: Orange Belgium, Orange Spain, Orange Slovakia and Orange Poland each connected by an arrow to CELINE
4 affiliates → 1 central inventory in Poland

Data pumps

Specialised data pumps collect data, transform it into the Common data model and, after each synchronisation, send the aggregated view to a centralised inventory in Poland.

Each source gets its own connection and transformation. All downstream calculations and algorithms stay generic and independent of the source.

  1. Source system× 4
  2. Data pumpconnect · transform
  3. Common data model1
  4. Central inventory · PolandSK · PL · ES · BE

Common data model

The model starts with very general terms and moves to more specific types. That is why it can describe any inventory situation, including sources added later.

  1. Inventory object
  2. Network element
  3. Router · Switch · OLT · Firewall

Illustrative hierarchy

Quality assurance

Mechanisms identify data inconsistencies such as broken reference integrity or violated data invariants.

Users can override data coming from a source system, and those corrections stay consistently visible across the whole application.

CELINE network element detail screen with general information, attributes and circuits, overridden values highlighted
Corrections stay visible across the application

Search and reports

Search across all inventory objects and attributes, plus integration with our DMS for work instructions, manuals, common problems and solutions.

Custom reports tailor-made for the daily needs of GNOC engineers aggregate data from various sources.

Want one view of your own assets?Network data scattered by country?

CELINE report listing network elements with building, name, usage description, platform and NOC owner columns
One search across all sources
Results

One primary source of inventory data for GNOC

What changed for GNOC, and where it led.

One inventory solution

CELINE integrates the available data sources into a single inventory solution for GNOC.

Extensible data model

The Common data model can document inventory situations from sources not yet identified when it was designed.

Foundation for SIA

CELINE caught the attention of the whole Orange group and became the foundation for SIA (Service Impact Analysis) at Orange Slovakia.

Where it led

  1. CELINEUnified inventory · Orange GNOCThis case study
  2. SIAService Impact Analysis · Orange SlovakiaRead the case study
  3. RCARoot Cause Analysis · extension of SIARead the case study
Is this for you?

Your assets are spread out. Your data doesn’t have to be.

CELINE was built for Orange networks, but the same problem appears wherever assets are spread across many locations, owners and tools.

  • Network and telecom projects
  • Critical infrastructure
  • Photovoltaics and energy
  • Automated convenience retail
  • Logistics and distributed operations
Sounds familiar?

0 of 5 sound familiar

How we would start

  1. Define the problem togetherWe map your data sources, who uses them and what they need to know. On CELINE, this was our first contribution.
  2. Design one common data modelA generic model that can describe any situation in your inventory, including sources you add later.
  3. Connect one source at a timeEach system gets its own data pump. Existing tools stay in place, so teams keep working as before.
Network data scattered by country?

Tell us
about your project

Describe what you want to build. We’ll get back to you within 24 hours with a proposal for next steps.

Need more info?

Reach out directly to our project managers, who will be happy to tell you more.

Romana Kovalová