Case study

AWS Migration Services for Seamless Big Data Analytics in Telecom

We’ve designed conceptual solutions to verify the AWS migration strategy in terms of infrastructure cost efficiencies and advanced data analysis

AI & MLCloud & DevOpsData & AnalyticsTelecom & Media

Project highlights

  • Design an architecture to migrate data processing and analytics from on-prem to the AWS cloud
  • Benchmark performance to optimize cloud infrastructure and use of services
  • Estimate infrastructure costs at scale to ensure seamless and efficient business growth
Team size:
4 engineers
2 months

Business challenge

Our client is a major national telecommunications company providing mobile and broadband data transmission services to over 23 million subscribers. In search of ways to scale their business using modern solutions but at optimal cost, they needed to move away from on-premises servers and decided to evaluate the concept of AWS cloud migration. By migrating their existing data warehouse to AWS, our client envisioned streamlining their data processing and business intelligence (BI) systems. For this, they required a cloud technology partner offering AWS migration services to build a proof of concept (PoC) to verify the efficiency of such a transition.

To advance their AWS migration strategy, our client launched a tender to find the best IT service provider for this mission. Recognized for skilled AWS-certified engineers, many successful projects, and profound experience in building telecom solutions, Intellias remarkably stood out among the competitors. Also, our AWS partner status attested to the fact that we meet strict requirements for AWS migration services in terms of expertise, capabilities, and engagement in the AWS cloud ecosystem.

Within the two-month duration of this PoC, we nimbly designed a customized cloud solution architecture, assessed resource requirements, measured the performance of the developed systems, and calculated expected infrastructure costs. This allowed our client to validate their ideas and estimate implementation costs quickly and with minimal effort to further inform their decisions and help them plan properly for the actual migration to AWS.AWS Migration Services for Seamless Big Data Analytics in Telecom

Solution delivered

Holistic vision

Like many telecommunications providers, our client has been relying on an on-premises database for their data processing and BI tools. Their existing data warehouse had extended exponentially, skyrocketing the cost of its maintenance and scalability due to more and more vendor licenses and hardware. The telecom provider thus decided to explore migration to AWS as a cost-saving and resource-efficient measure.

Along with cost and resource optimization, our client’s motivation behind AWS cloud migration was to accelerate their growth and innovation and thus future-proof their position in the telecom industry. That’s why we also worked to increase our client’s long-term business value by upgrading their underlying big data and analytics with the latest technologies such as data lakes and fast data.

Multipronged approach

In terms of our AWS migration services, we had to implement three dataset scenarios for our client and evaluate them based on cost and performance efficiency:

  • Aggregation batch processing (for handling accounts)
  • Predictive model building (to anticipate customer behaviors)
  • Stream data indexing (for fast search of the latest activity)

We came up with two architecture designs for each scenario. One uses custom jobs that explicitly replicate the functionality of the on-premises data warehouse in the cloud, and the other is entirely based on out-of-the-box AWS capabilities. As part of our AWS data migration service, we also estimated monthly expenses for cloud infrastructure based on the volume of data handled by the telecom provider in real life. This flexible approach allowed our client to understand available options and make well-grounded choices based on the benchmark comparisons we provided.

To effectively implement batch processing during the migration to AWS, we first measured relevant performance metrics for the existing on-premises data warehouse. After that, we transferred the data model to cloud components and designed an efficient extract–transform–load (ETL) solution that scored much better than the existing system. As most of the processing happened at night (when it’s less expensive), only some critical jobs needed to run during the more expensive daytime hours. Therefore, we saved a lot of resources by not needing to keep all the infrastructure constantly engaged.

To enable predictive model building when migrating to AWS, we prepared comprehensive data lake-based solutions, unlocking a full range of ML/AI capabilities for our client’s data scientists and BI specialists. We focused on ensuring a seamless workflow so end users could build and run their models with improved speed and flexibility.

For stream data indexing, we designed another ETL solution to implement fast searching by customer ID to instantly find information on the latest activity. As with batch processing, we first broke down the existing on-premises workflow into elements, measured the performance of each, and then decomposed the data model in AWS to achieve an optimized implementation. In this way, we effectively used fast data analytics to promptly get actionable insights from an enormous data pool.

Expert synergy

This business-critical project involved three stakeholders — our client (the largest national telecommunications provider), AWS (as a cloud platform provider), and Intellias (as an AWS data migration service consultant). We orchestrated our collaboration to make it mutually beneficial. A manager, solution architect, DevOps engineer, and data engineer from Intellias extended our client’s team by providing technology skills in migrating to AWS.

The Intellias–AWS partnership has allowed us to have on-demand technical sessions and workshops with AWS subject matter experts to advise on best practices and eliminate risks. An excellent example of the data migration service AWS and Intellias have provided in tandem is effectively handling credit activation for our client.

Intellias took an active part in setting up the AWS cloud migration environment jointly with our client’s team and helped them solve other technical issues such as adjusting the VPN transfer rate. We also conducted a detailed cybersecurity analysis as part of our AWS migration services to ensure reliable protection against virtual threats for a safe cloud environment. The suggested approach would provide security across all network layers for our client — an essential requirement in today’s telecom industry.

Business outcome

Working alongside our telecom client and in partnership with AWS, we successfully delivered architecture designs for cloud solutions that effectively solved issues of vendor lock-in such as costly on-premises data servers. To verify the telecom provider’s AWS migration strategy and to estimate future infrastructure costs, Intellias engineers provided performance benchmarking and gathered resource statistics for a comprehensive comparison.

When transferred to AWS, data processing will open broad opportunities for the telecom provider to optimize resource costs, while moving analytics and BI to the cloud will improve customer retention by offering subscribers more personalized services. After migrating to AWS, our client expects to considerably cut infrastructure expenses, increase revenue, and easily scale their business.

This pivotal prototyping project accomplished by Intellias for a giant telecommunications company in a tight timeframe but with due dedication and professional responsibility laid the foundation for mutually beneficial cooperation in the future. As a continuation of our AWS data migration service, we keep on working on new proposals and delivering high-end solutions to our telecom client.

How can we help you?

Get in touch with us. We'd love to hear from you.

We use cookies to bring best personalized experience for you.
By clicking “Accept” below, you agree to our use of cookies as described in the Cookie Policy

Thank you for your message.
We will get back to you shortly.