2017 Highlights Series
Nima Kaviani (Cloud Engineer at IBM) and Michael Maximilien (Scientist, Architect, and Engineer at IBM) delivered a talk at Cloud Foundry Summit Silicon Valley in which they shared the results of the multiple experiments they conducted to understand how Cloud Foundry Serverless Computing behaves under different workloads on different platforms.
The duo briefly explained the meaning of Cloud Foundry Serverless and its benefits — including cost cutting.
“If your function runs only occasionally, you’re going to get charged a lot less than having an application up and running for the entire time and having to pay for all the resources [while your] application is sitting idle,” said Kaviani.
The team looked into how different platforms handle memory or CPU intensive functions. Their most interesting finding was the way Cloud Foundry Serverless responds differently on different platforms such as Microsoft Azure, AWS Lambda, IBM’s Openwish and Iron.io.
They also tried to understand how other platforms behave when it comes to creating containers.
“It’s important to understand how these serverless platforms manage containers because container management is a significant part of operating a service system,” said Kaviani.
The team open sourced many of the tools they used for these experiments in order to enable the community to conduct these tests on their own and collect their own data.
Watch the video: