/Dialog_Axiata_logo%401x.ce9d6c59728c0c2e8779f62f4e3df25c48fe0e06.png "Dialog Axiata")
Dialog Axiata Migrates to AWS to Optimize Operating Costs by 18% While Meeting Dynamic Consumer Demands
2022
During the 2020 pandemic, telecommunications companies (telcos) faced unprecedented demands on their systems and networks as businesses and schools moved to remote or hybrid working and learning models. Telcos had to adapt to meet customers’ rapidly changing habits and surging connectivity needs. To respond to demand in an agile manner,
more than half of telcos surveyed by McKinsey said they were migrating more assets to the cloud.
One of Sri Lanka’s main telcos,
Dialog Axiata PLC, embarked on a digitization initiative in 2019 and began migrating to the cloud in 2020. Today, Dialog runs more than half of its workloads on the cloud. As part of Dialog’s digitization strategy, the telco adopted a holistic approach to improve its customer experience while considering how its back-office operations could be optimized.
Tasks that used to take two months, such as data preparation and churning out ML models, now take just two weeks on AWS.”
Asela Perera
Chief Information Officer, Dialog Axiata PLC
Chief Information Officer, Dialog Axiata PLC
Enhancing Cloud Capabilities with Partner Support
To provide guidance on migrating enterprise workloads with high availability, Dialog engaged
MillenniumIT ESP (MIT ESP), an
AWS Partner. The telco also relied on Amazon Web Services (AWS) for migration support and for help in training its team members on operating in the cloud. “What I really appreciate is that AWS and Millennium were patient, taking any questions we asked and walking with us on our digital transformation journey to build our capabilities,” says Malik Induruwana, head of IT and enterprise security at Dialog Axiata PLC.
After completing a detailed three-month migration assessment with MIT ESP, Dialog began migrating its Charging Gateway and Order Management (OM) system, which processes all of the company’s quad-play services, to AWS. It also migrated its Campaign Management System (CMS), a critical revenue driver used to manage sales campaigns that are pushed to target audiences in near-real time.
As part of its CMS, Dialog runs daily reports that management and marketing teams use to make strategic revenue decisions. These cumulative reports are run overnight and formerly took until 10 a.m. or later the following day to complete. On AWS, reports are available by 5:30 a.m., so management has all the key campaign information they need at their fingertips when starting the day.
Using Infrastructure as Code to Lower Time to Market
Speed to market has proved a major advantage of operating in the cloud. In its previous on-premises environment, Dialog had to plan for downtime to update features, perform routine maintenance such as patching, and ensure complete data transfer to analytics dashboards. The business now relies on
AWS CloudFormation to provision infrastructure as code and build infrastructure templates it can distribute across environments.
Scaling with Microservices during Big TV Events
Since migrating to the cloud, Dialog has also shifted from a monolithic to a microservices architecture. It now uses containers and autoscaling on AWS for rolling deployments, which improves the customer experience during popular TV viewing events. “Scaling was troublesome in our legacy environment and often led to interruptions or downtime during major cricket matches. We haven’t experienced any downtime during peak events since migrating to AWS,” says Induruwana.
Optimizing Operating Costs with Spot, Reserved, and Graviton Instances
Lockdowns and a downturn in the business environment prompted Dialog to pursue cost optimization as a secondary phase of cloud migration. The enterprise worked with MIT ESP to identify major cost centers, increase costing transparency, and examine utilization patterns for each application. With advice from AWS and MIT ESP’s expertise in FinOps—managing operating expenditures in the cloud—Dialog was then able to optimize the compute and storage resources for each system.
Shahan Kalutanthri, director of cloud technologies at MIT ESP, says, “Dialog has been proactive in adopting the latest technologies, and we’ve been their long-standing tech enablement partner. Still, introducing cloud financial management practice to an enterprise was a unique challenge. Together with the AWS team, we were able to bridge the technology gap and implement the most efficient consumption models using native AWS tools.”
Dialog now uses
Amazon Elastic Cloud Compute (Amazon EC2) Spot Instances for non-production workloads running in microservices/Kubernetes environments. As of May 2022, through the use of Spot Instances, Dialog has saved $127,000. Additional cost-saving mechanisms include the adoption of AWS
Savings Plans for fixed workloads in Amazon EC2 and Reserved Instances for static database workloads in
Amazon Relational Database Service (Amazon RDS). The adoption of Savings Plans alone have saved the company $490,000.
The business also switched from Intel to
AWS Graviton Processors and is using Graviton instances to optimize costs on core database and analytics workloads. Dialog plans to increase Graviton adoption in the coming months to notch further savings. Finally, the business migrated from gp2 to gp3 volumes on
Amazon Elastic Block Store (Amazon EBS), which offers better performance at the same or lower price. These changes, plus auto-scheduling workloads to shut down when not in use, have resulted in cost optimizations of 18 percent.
Overcoming Scaling Limitations with Cloud-Native Data Lake
Amazon QuickSight is another tool that has proven invaluable for Dialog. In addition to building custom dashboards in Amazon QuickSight to ensure cost transparency for business owners, teams use the service to support Dialog’s analytics pipeline. The telco is currently migrating its data lake to
Amazon Simple Storage Service (Amazon S3) and its data warehouse to AWS.
“We realized the scaling limitations we had with our on-premises data setup, which caused delays in extract, transform, and load processes. The elements of our data pipeline that we’ve migrated to AWS thus far have experienced little to no disruption,” Induruwana says.
Data processing is not only more reliable and scalable in the AWS Cloud, but also faster. “Tasks that used to take two months, such as data preparation and churning out machine learning models, now take just two weeks on AWS,” says Asela Perera, chief information officer at Dialog Axiata.
Creating AI- and ML-Driven Use Cases
Dialog is actively building advanced analytics use cases that incorporate ML through
Amazon SageMaker. It is in the process of creating an AI Factory within the company, with more than 30 data scientists, analysts, and ML engineers who were hired to standardize and democratize within the organization. The AI Factory has churned out five use cases thus far. One use case is focused on ways to distribute stock of mobile phones and accessories in an optimal way to Dialog’s 50,000 retail points throughout the country.
Induruwana concludes, “AWS helped us set up our AI and ML practice in a structured way. We’ve seen that AWS is always there for us, in times of difficulty and in times of innovation.”
Learn More
To learn more, visit aws.amazon.com/cloud-migration
About Dialog Axiata PLC
Dialog Axiata is among Sri Lanka’s leading telecommunications companies, offering subscribers mobile, wireless, broadband, and TV services. Its vision is to be the undisputed leader in providing multi-sensory connectivity. Dialog’s parent company, Axiata Group, is one of the largest telco providers in Asia.
Benefits of AWS
- Optimizes cloud operating costs by 18%
- Receives training for 300+ team members across 22 courses
- Runs key overnight reports at least 3.5 hours faster
- Reduces time taken for ML data preparation from 2 months to 2 weeks
- Eliminates downtime during major TV events and data transfers
- Uses AI and ML to optimize deliveries and identify fraud
AWS Services Used
Amazon EC2 Spot Instances
Amazon EC2 Spot Instances let you take advantage of unused EC2 capacity in the AWS cloud.
Amazon SageMaker
Prepare, build, train, and deploy high-quality machine learning models quickly by bringing together a broad set of capabilities purpose-built for machine learning.
Amazon QuickSight
Amazon QuickSight allows everyone in your organization to understand your data by asking questions in natural language, exploring through interactive dashboards, or automatically looking for patterns and outliers powered by machine learning.
Amazon Elastic Block Store
Amazon Elastic Block Store (Amazon EBS) is an easy-to-use, scalable, high-performance block-storage service designed for Amazon Elastic Compute Cloud (Amazon EC2).
Get Started
Organizations of all sizes across all industries are transforming their businesses and delivering on their missions every day using AWS. Contact our experts and start your own AWS journey today.