Running Lambda powered by Arm/Graviton2 vs x86_64
AWS Lambda Functions Powered by AWS Graviton2 Processor – Run Your Functions on Arm and Get Up to 34% Better Price Performance
Test case flow
The following test case will be using a lambda to call a dynamoDb table
The following test cases with different runtime environments
For the warm state, it will be sending multiple requests for each lambda one by one (lambda concurrency is 1, all request go to the same lambda instance). And on the slides, you can see the duration of the last request.
NodeJS: ARM is a bit fast
DotNet: ARM is fast
Python: ARM is faster
TeamConnect test case
It is under the TeamConnect account utilizing Lambda, DynamoDB, and EventBridge resources.
Also, it is versioning to be able to use multiple architecture in this case x86 and arm64
Arm64
Vs
x86
Cost
Takeaway
Arm-based systems and instances are available in public clouds such as AWS.
The computational performance of AWS’s Arm EC2 instances is similar to that of the x86_64 instances.
Considering that Arm instances are significantly cheaper, the cost effectiveness of Arm instances are better than x86_64 instances.
Arm instances perform better with “close to metal” applications. We hypothesize that the operating systems have received more engineering efforts to optimize for Arm than high level application frameworks.
Top comments (0)