Twitter also used other tools like Netty, designed to help create high-performing protocol servers, and Scalding, which makes it easy to write big data jobs. Twitter hasn't been able to move away from the Ruby on Rails application framework entirely, but these tools, combined with switching its core infrastructure to JVM, have helped the company avoid fail whales and improve performance, Aniszczyk said.
"This pretty much has enabled us to grow as a company," he added. Twitter has over 2,000 employees, of which about half are engineers, he said.
One of the lessons Twitter learned is that basing its infrastructure on open source is a good idea. "That is where you find the best software these days," said Aniszczyk, who added that it is also good to give back to the open source community, which Twitter does at twitter.github.io, which lists 100 public repositories.
Incremental change is also a good idea, he said. "It increases the chance of success if you do things in a small piecemeal fashion," he said, adding that companies should also keep learning from universities.
Twitter's efforts to change its core infrastructure is a good example of how companies can benefit from open source technology, said Jim Zemlin, executive director of the Linux Foundation. "It keeps the cost down and it also allows them to meet very quickly the challenges that they have with the scale of infrastructure," he said.
Many businesses, including large companies such as Google, Amazon and Facebook, are all creating open source hyperscale computing platforms, and they can borrow lessons learned, Zemlin said.
Loek is Amsterdam Correspondent and covers online privacy, intellectual property, open-source and online payment issues for the IDG News Service. Follow him on Twitter at @loekessers or email tips and comments to firstname.lastname@example.org