Telecom Case Studies
Confronted With Latency Issues Within Their Key Infrastructure
Client: A large Telecom group with more than 170 million customers in 5 continents
The challenge was to engineer a scalable Data Warehouse solution in a way that set the standard for all future Data Engineering initiatives and to boost the client’s ability to deliver business intelligence into the organisation.
The client mobile Data Warehouse was a brownfield site with a complex technical and organisational environment. Elait set up the architecture framework, the process framework, and an internal team of application specialists who helped create templates upfront to enable rapid development of ETL code for the warehouse application.
Elait rebuilt the interfaces, set up standards and the structure for application maintenance in terms of development processes, testing processes, standards, governance around the tools, and change control procedures.
The scalable Data Warehouse architecture was organised to enable the customer to quickly and efficiently expand by adding more hardware.
Elait helped totally transform the speed of moving data around. A day’s worth of data now runs in a quarter of the time it used to take.
Track Customer Connections & Derive Business Intelligence For Campaigns
Client: Large Telecom corporation with more than 26 million customers
A greenfield project where terabytes of data were migrated from one Data Warehouse to another. Re-architect data stored in earlier systems to new Data Warehouse to enable a customer-centric model.
Elait setup a new Data Warehouse, migrated data from the inherited Data Warehouse to the new one, and upgraded infrastructure and application from legacy estates to contemporary systems.
Migration Of Current Data Integration (Primarily ETL) Code From Windows To UNIX
Client: A Telecom major from the Middle East
Migrating large amount of data from multiple data sources, optimise memory utilisation and implementation of the appropriate aggregation approach according to complexity and data size.
The proposed architecture allowed for the recycling or exclusion of data that was deemed to be invalid or failed to meet data quality measures.
At strategic control points throughout the ETL process, audit data such as record input count the output count, and rejected count was captured and stored in audit tables to facilitate operational route cause analysis, reconciliation, and trend analysis. The Data Integration ETL collected Log information across every flow and transformed process to ensure all the graph level operational statistics are collated.
Error handling was designed to categorically explain the cause of data rejection, expedite problem resolution via comprehensive, unambiguous, and consistent information, in addition to the ability to raise alerts according to error severity.
The architecture reduced development time and cost, had reusable components and processes that allowed the client to build once and reuse many times. This also resulted in reduced effort, decreased risk, minimal defects & incidents, and significantly fewer maintenance incidents.