Published on 15 Mar 2012
T-Mobile USA, Inc. is a communications provider (CSP) that offers wireless voice, messaging and data services to more than 293 million customers. Headquartered in Bellevue, WA, the company's services are available in more than 9,200 public-access locations nationwide, including roaming locations. In this video, you will hear from Christine Twiford, Network Technology Solutions Manager, discuss how T-Mobile increased its storage capacity from 100TB to 2 PB and, as a result, network availability and overall efficiency when it upgrades its data warehouse environment with an IBM Netezza 1000.
Prior to IBM Netezza implementation, T-Mobile relied on a 40 TB Oracle data warehouse for all of its data-storage needs. However, the company needed a faster, more efficient data warehouse with additional capacity. The company invested in an IBM Netezza 100 appliance, with a capacity of 100 TB, to store clickstreams: records of each website a user viewed or used and for how long. As clickstream data became more abundant, the Netezza 100 appliance reached its capacity. T-Mobile needed a solution that provides longer retention and more meaningful results for clickstream data.
Being familiar and satisfied with Netezza products, T-Mobile upgraded to an IBM Netezza 1000 appliance to support the increased volume of clickstream data from extensive smartphone use nationwide. The client uses the Netezza 1000 appliance to provide end-to-end extraction for downstream applications including revenue assurance, billing and customer care. They have even begun an initiative with social networking analytics, where they have been able access in real-time, which can help to prevent churn.
- Reduce tax and call-routing fees by using the data stored to defend against false claims - Acquire dropped-call and churn-reduction analysis capabilities - Increase network availability by identifying and fixing any network "holes" - Measure and correct handset performance to help ensure product performance
In this video, Christine Twiford, T-Mobile's Manager of Network Technology Solutions, explains how and why T-Mobile replaced a 40 terabyte (TB) Oracle data warehouse with an IBM Netezza appliance. She discusses the various business units served by the IBM Netezza appliance today and the evolution of that platform which today holds 2 petabytes (PB) of data.
Christine Twiford, T-Mobile's Manager of Network Technology Solutions
Our primary mission with the Netezza project is to serve the network engineering side of the house and to make sure that our customers are having a quality experience with our network. We also do serve the rest of the organization and so we’ve been able to send data to revenue assurance, to billing, to marketing. We’ve even begun an initiative with social networking analytics and we’re able to work with marketing and customer care to work on some fairly real-time interventions to try and prevent churn which is kind of an exciting initiative.
Everything that we do with Netezza was pretty much impossible before. It was a complete change in our experience of data warehousing when we moved to Netezza. We were not able to do nationwide searches before or any kind of aggregations at a whole network level. That was just inconceivable. We’re able to do that now in just seconds with Netezza and that’s changed everything for us.
We’re replacing a 40 TB oracle system and we replaced it with a 100 TB 10800.
As we scaled from 100 TB to 200 TB to 500 TB to 1 PB to now 2 PB we still have the same team. There’s just a principle engineer, a lead software developer -- or a lead SQL developer -- and a product manager who interfaces with the business and tries to understand those needs and get those needs met.
We load 17 billion records a day and we have 150,000 ETL processes – or ETL processes -- each day serving 1300 customers with direct access through a web interface and all of these things are maintained by a very tiny team within a department within an organization under the umbrella of T-Mobile.
There was a family stranded off the coast of Miami and they were just out there, their boat had capsized and in the process of capsizing of the boat, they got one phone call out. But you know, when your phone is wet, it powers off, there’s nothing you can do. We were able to assist in tracing the last tower that that call was made from and able to give an area where that boat was capsized, otherwise the search could have taken many more hours. They were able to find them I think within just an hour or so.
IBM products and services that were used in this case study.
IBM Netezza 1000