XenonStack - A Stack Innovator

Analytic solutions

Analytic solutions for Telecom

The Problem

Migrated from traditional to BigData Solution

BigDataAnalytics

Challenges

The speed of I/O was too high along with huge amout of data.

Our Solution

Telecom Industry brought up an opportunity to work on data; big in terms of volume and velocity. A huge amount of data is produced by telecom industry at a very high velocity. This data is increasing rapidly as telecommunication is spreading rapidly. We are serving telecom companies form different parts of China and Japan. These companies are vendors of IP based telecommunication services.

Each of these client produces 12 – 14 TeraBytes of data per month. While serving more than 30 clients, we retain 90 days of data for analytic purpose and this data is replicated to create a total of 3.6 Peta Bytes of data on the cluster.Considering the velocity of data that reaches 10,000 writes per second for single client during peak hours, we needed an architecture which could handle parallel heavy IO operations and robust enough to work in worst scenario of network bottlenecks. We found our solution in a Cassandra release to build this solution.

As Cassandra architecture is peer to peer in nature, it has no single point of failure. Cassandra is designed to work on in-memory operations and hence is fast enough to handle such velocity. Still, this efficiency depends on the cluster configuration. While working with data in motion, we have build an interactive interface in javascript backed by NodeJS to serve analytics in real-time.

As Cassandra architecture is peer to peer in nature, it has no single point of failure. Cassandra is designed to work on in-memory operations and hence is fast enough to handle such velocity. Still, this efficiency depends on the cluster configuration. While working with data in motion, we have build an interactive interface in javascript backed by NodeJS to serve analytics in real-time.

Disclaimer

Clients’ identity and research methodology shared with us has been kept confidential in this document in lieu of confidentiality agreement with client and our policy. If you would like to know more about this research, please write to us at business@xenonstack.com.

Transforming to a Data-Driven Enterprise

Talk to Experts for Assessment on Infrastructure Automation,  
DevOps Intelligence, Big Data Engineering and Decision Science

REACH US
Contact Us

Chat With Our Experts

Your Information Submitted Successfully!