NGINX is the most popular scale-out web application server. NetCraft ranks it as the leading web server in the world, powering more than 35% of active websites.
In this blog, we demonstrate the performance advantage of Yitian 710 over x86 by comparing NGINX throughput on Alibaba Cloud ECS. We use G8y instances (powered by Yitian 710 processors based on Armv9) and G7 instances (powered by 3rd Generation Intel Xeon Scalable processors) for this comparison.
For client traffic generation, we used the open source benchmarking application wrk.
For the NGINX server, we deployed NGINX on a single Alibaba Cloud ECS instance with 8 workers configured.
Figure 1. NGINX benchmarking topology
The server under test has two ECS instances with the following configurations. The benchmark client used a single G8y.8xlarge instance.
Table 1. Test server configurations
The benchmark tests were performed with the following software versions and test parameters.
The wrk client creates 32 threads with total 1000 connections. To improve performance, we enable XPS (transmit packet steering) on both G8y and G7 instances. We also bind 8 transmit queues to 8 cores separately. We observe up to 32% increased performance running NGINX on Yitian 710 based instances compared to equivalent x86-based instances. The results shown in the following tables are an aggregated result of 20 consecutive test runs.
Let us look at the performance numbers of NGINX on G8y and G7 instances. We compare the throughput (operations/sec) values after multiple test runs.
Table 2. NGINX throughput performance results on G8y vs. G7
Figure 2. Performance gains for G8y vs. G7 instances
NGINX deployed on Yitian 710 based ECS provides up to 32% more throughput in compared to the equivalent x86 based ECS instances. In addition, according to Alibaba, the G8y instance based on Yitian 710 is priced 20% less than the G7 instance.
Please visit this page for details on how to migrate existing applications to Alibaba Cloud Yitian710 instances. For any queries related to your software workloads running on Arm platforms, feel free to reach out to us at sw-ecosystem@arm.com.
More Cloud Workloads on Arm