The Context
dbSeer takes pride in helping organizations with demanding IT workloads maximize their ROI through strategic cloud analytics engineering services. Sustainable scalability is a common concern many businesses are facing, yet many are unaware it does not require more investment in costly hardware. When TOMIA, the world’s leading provider of B2B and OSS/BSS management solutions, turned to dbSeer with a pressing need for cost-effective and efficient scalability, our team of seasoned engineers knew exactly what to do.
The Challenge
TOMIA processes hundreds of billions of call detail records to provide actionable business analytics to facilitate near real-time optimized trading, billing, and routing that is compliant with agreements between telecom companies. The key dilemma TOMIA faced was that although the maximum workload only took place once or twice per month, they were paying for full capacity at all times. Because TOMIA used commodity servers to do extensive processing, this required increasing hardware and corresponding costs to yield faster speed.
The lift and shift approach of migrating their event-processing engine to AWS EC2, while providing a quick way to move from on-prem infrastructure, did not offer TOMIA the best architecture for the cloud in the areas of scalability, performance, and cost. Hence, this was just the first step for TOMIA. The next step is where dbSeer’s expertise came into use – building a scalable model that could reap AWS’ pay-what-you-use benefit to maximize savings.
The Solution
dbSeer believes clients should only pay for what they use and enjoy the freedom to scale up without steep costs. Our team leveraged AWS’ near-infinite elastic resources and open-source technologies to cut TOMIA’s analytics platforms’ cost. In turn, the time to produce results is heavily reduced. This enabled TOMIA to enjoy unlimited market expansion at a fraction of the cost and serve their customers efficiently.
The Technicalities
dbSeer re-architected TOMIA’s AWS infrastructure to evolve from a SQL process running on Microsoft SQL Server into a Python-based process built on native AWS services and open source libraries to perform the same tasks.
dbSeer’s use of AWS native services such as AWS SWF, AWS SQS, AWS Glue, AWS Redshift, and AWS S3 creates a powerful platform that decouples compute resources from data storage; therefore, the ability to scale and pay for what you use is enabled.
The Numbers – 60x Faster at 10% the Cost
The new AWS processing engine dbSeer built for TOMIA was thoroughly tested on a variety of server configurations and volumes. It demonstrated the following five highlights –
- Linear Scalability – The processing speed increased proportionally with the volume of servers added.
- Reduced Costs – The processing cost per million CDR records could be reduced by up to 90% because Linux-based instances are at a low cost and AWS offers a volume-based pricing model.
- Increased Speed – Workflow could be distributed across hundreds of low-cost, efficient machines, resulting in a processing speed up to 60x faster than TOMIA’s previous architecture.
- Improved Durability – With S3, Amazon guarantees 99.99% data durability. This is achieved by replicating the data across multiple data centers, providing significantly safer storage than relying on an individual server instance.
- Easy Deployment & Modification – By treating TOMIA’s infrastructure as a code and using AWS cloud formation, dbSeer simplified the deployment and changes of processes, making it easier to track and also more efficient.
Final Thoughts
TOMIA witnessed excellent results in the areas of efficiency, elasticity, and ROI thanks to the power of elastic cloud infrastructure and dbSeer’s robust analytics consultation for re-architecting infrastructure to utilize AWS’ pay-as-you-go principle fully. dbSeer’s bespoke solution did not require TOMIA to over-invest in hardware to handle expected, but infrequent periods of high traffic. Resources are now allocated dynamically for less server idle time, increased speed, and decreased costs.
Let’s Talk!
Wondering how dbSeer can customize the right infrastructure for your business so scaling can be sustainable and cost-effective? We invite you to schedule a free, no-strings-attached thirty-minute custom data analytics engineering consultation. Click here for more details.
