Over time, we've seen QlikView increasingly embraced by large enterprises with thousands or even tens of thousands of employees. We also see it deployed on public-facing Web sites. The ability to scale and support thousands of users has become an increasingly important requirement for business analysis software. We faced this requirement as we began to use our demo site heavily to support corporate marketing efforts.
We were successfully able to handle a peak load of 8,000 distinct anonymous Kick It and Qlik It users a day by:
Utilizing the cloud. We moved our QlikView servers from our data center in Sweden to a US data center in the Amazon EC2 cloud. This shift to an elastic, cloud-based resource enabled us to meet spikes in demand affordably.
Adding geographic locations. We put servers in three geographic locations. By putting the servers closer to the people, we were able to reduce network lag and improve response times.
Adding server capacity. We bumped up from "large" to "extra large" servers and increased the minimum number of servers in each geographic location from three to five. This enabled us to support a larger number of users simultaneously.
This story is one example of how we live our customers' experience ourselves. For more details about how we did it, click here to download the white paper, "How We Scaled to Support Thousands of Users Daily on a Global Scale."