Customer Data Management

What Angry Birds Can Teach Enterprises About Critical Data

It’s true: that addictive little app you play in the palm of your hand can teach us a lot about managing massive amounts of critical data. What’s the connection between a mobile gaming app and a robust and effective enterprise infrastructure? Countless (and growing) amounts of critical data and the pressure to constantly meet high customer experience expectations.

Rovio (creators of Angry Birds) deals with more data than many enterprises. As of last year, Angry Birds experienced over two billion downloads, 263 million monthly users, and was downloaded on 50% of new mobile devices. Now, consider all the data these users bring (personal information, player stats, etc.). The numbers get big fast.

Gamers are arguably the most challenging audience - with extremely high expectations that demand low latency gameplay and 24/7 access to their accounts on all of their devices. Gaming companies like Rovio have nailed down critical data by focusing on the key factors—performance and availability—that affect business revenue, attrition rates, and user experiences.

The number one factor in gaming success is performance. A few seconds of downtime or latency can be disastrous: Do you think a gamer who has great skills but can’t shoot fast enough due to game latency will hang around? Not a chance. Gamers also expect that their data (scores, stats, user profiles, etc.) will always be accessible and never lost. Nearly all of the successful gaming companies have implemented a distributed database to better serve their users, and the benefits below can also easily be translated to the enterprise:

  1. Cater to high user expectations. Users have high expectations about what their experience should look like - regardless of industry. Companies like Rovio, Amazon, and Facebook have set the bar high and they all utilize distributed systems to ensure high availability, zero data loss, and low-latency. Distributed databases are the only way these companies can keep up with critical data while keeping their customers happy.
  2. Flexible storage. Unlike relational databases, Riak, the NoSQL database Rovio uses, has a schema-less design and zero restrictions on data formats. This means any data that’s critical to users (such as user and profile information, game statistics, social information, and session data) can easily be stored, managed, and delivered, regardless of its format. But most importantly, that data is readily available when gamers need them.
  3. Low-latency design. If an app is unreliable or experiences lags, it will be unceremoniously dropped. In a distributed environment, data is replicated across multiple nodes, which provides a high tolerance for hardware failure. This ensures the system will stay up and running, without losing critical data.
  4. Global data locality. For gaming companies, it is essential to keep critical data as close as possible to users to deliver optimal performance. Multi-datacenter replication establishes a global presence to keep data close to users, while providing the same high level of service, no matter where they are in the world.
  5. Cost-effective management of rapid growth. In December 2012, Rovio had 263 million monthly active users. By March 2013, its games soared to 1.7 billion downloads and counting. This translates to more users, each bringing hundreds of pieces of data to the system and making scalability crucial to success. Rovio’s infrastructure allowed them to easily add and remove nodes, quickly scaling from tens to hundreds of servers based on immediate customer demand to save money and resources.

Rovio is not the only gaming company that is successfully utilizing data to positively impact their bottom line. Having previously worked with a relational database that required costly and time-consuming manual sharding to scale and deliver data, Hibernum Creations deployed a distributed NoSQL database architecture to more effectively store user game information and serve thousands of requests per second for over a million monthly active users. Additionally, Mochi Media, a flash-based gaming platform, uses a distributed architecture to store social graph data for Mochi Social and MochiGames, serving data quickly and reliably. Lastly, Kiip builds rewards and achievements into games using a distributed database to achieve low read/write latencies and horizontal scalability for session and device data.

Gaming companies of all sizes have realized the benefits of distributed NoSQL systems and have built successful businesses by understanding a few key rules:

  1. Keep data close to your customers - to deliver low latency experiences
  2. Be flexible - utilize a schema-less design to easily manage and deliver multiple data types and provide a consistent user experience across various channels
  3. Never go down - with today’s technology, failure and downtime is simply not an option
  4. Be predictable - there are enough surprises in business - architecture costs, scale, and user experience don’t need to be one
  5. Pay-as-you-go - systems that allow you to easily scale up or down make it simple to only pay for what you need at the time

Many companies have spent the last few years learning to capture big data, but now they need to learn how to leverage it. These ever-growing data sets can drastically improve the bottom line of various businesses. And no one understands that better than the gaming community.


Andy Gross, Chief Architect, Basho



« Why 2014 Could be the Right Time to Move Out East


News Roundup: Secret IPOs, Tech Philanthropy and Bitcoin Psychology »
Andy Gross

Andy Gross, Chief Architect, Basho

  • Mail


Do you think your smartphone is making you a workaholic?