Abstract

Applications executing in virtualized environments experience a lack of performance isolation from other services executing on shared resources. Latency sensitive applications running in the cloud are therefore sometimes affected by highly-variable response times, yet understanding the causes of such variations at the level of individual requests is an unsolved problem.

We present Soroban, a framework for attributing latency to either the cloud provider or their customer. Soroban allows developers to instrument applications, such as web servers to determine, for each request, how much of the latency is due to the cloud provider, and how much is due to the consumer’s application or service. With this support Soroban enables cloud-providers to provision based on acceptable-latencies, adopt fine-grained charging levels that reflect latency demands of users and attribute performance anomalies to either the cloud provider or their consumer. We apply Soroban to a HTTP server and show that it identifies when the cause of latency is due to a provider-induced activity, such as underprovisioning a host, or due to the software run by the customer.

For more details, please visit https://www.cl.cam.ac.uk/research/dtg/rscfl/soroban.html

This is a practice talk for HotCloud’15