|Client||Largest telecom provider in Sri Lanka|
Cloud Deployment for Sri Lanka’s Largest Telecom Provider
The objective was to build a comprehensive cloud services platform to offer IaaS, PaaS, SaaS, DaaS and a plethora of other services.
Our broad delivery scope consisted of the following:
- Create cloud architecture and design based on the client’s business objectives. This consisted of both network and system design.
- Deployment of management & control servers.
- Deployment of hosts in production and staging environment.
- Integration of Citrix-ready solutions with the cloud.
- Development of connectors for other ISVs for integration with Citrix SSP.
- Design and development of web front-end portal.
- Integration with the client’s existing OSS/BSS, CRM for user data and billing & payment transaction portability.
- Migration of existing VPS environment into the cloud.
- Training and knowledge transfer.
- Limited awareness of the cloud business model with the client – Since this was the client’s first initiative in cloud business, there were a lot of misconceptions on how the cloud should function. We, therefore, took special care and a significant amount of time and effort to bring the customer’s expectations to realistic levels.
- Dated IT systems at the client’s end – Since the client’s core applications were a few years old, most of them did not have a web services layer, making API-based integrations extremely difficult. We, therefore, created an external mapping database and synchronised all transactions through SQL reads/writes. Though not elegant, this workaround served the purpose of seamless synchronisation between both platforms.
- Poorly written APIs for Citrix SSP – The BSS API library of Citrix CPBM application was poorly written and documented, leading to issues with integration with the front-end portal and the client’s IT systems. We created custom scripts to extract information from the applications database and present it to the other layers.
Since the client’s major target segment is of large enterprises and the Government sector, we followed the primary objective of high resiliency. We opted for a complete HA architecture with no single point of failure, right down to the network interface level on the servers. Also, we created a separate zone for the staging environment to test all new releases before deploying on the cloud.