British charity Comic Relief has revealed that it used open source NoSQL database MongoDB to create a flexible and scalable platform that was vital in handling this year's record breaking £75 million event.
In previous years, Comic Relief had been delivered via a Java and relational database model through 12 different partners, which meant there was a lengthy feedback cycle.
This year, Comic Relief enlisted Armakuni, the cloud-service technologies provider, to create a platform-as-a-service solution that could handle 10,000 concurrent call centre operators and a peak of 500 donations per second.
For Comic Relief, fifty minutes of downtime during peak fundraising hours could result in half their annual online donations income being lost.
"During a national live TV show, the culmination of a 10 week national media onslaught, there is absolutely no second chance to get it right," said Tim Savage, co-founder of Armakuni.
MongoDB was chosen as a key component of Armakuni's platform because it was known to be flexible, reliable and scalable enough to deal with huge spikes in activity. It also worked seamlessly with other solutions needed for the project, such as Cloud Foundry.
The Armakuni system dealt with 600,000 transactions during the seven hour prime time TV slot, with donations coming in from from web, mobile and call centre users.
The database was also accessible to 10,000 call centre operators who were encouraged to compete against their peers for the most donations.
In the end the platform helped raise more than in any other single night in Comic Relief's 28-year history with a total of £75,107,851 - passing the previous high of £74,300,000.
"This year's Red Nose Day was a great success. The scalability, flexibility and resilience of the platform meant that, for the first time, the technology was not a limiting factor in raising as much money as possible." said Phil Latham at Comic Relief.
Armakuni is now preparing for next year's event, where it hopes to offer organisers real time analytics.
"The next challenge is crunching data in real time and using the information in even more clever ways," said Savage.