- Allen Helton, in a recent blog post titled “Serverless Speed Test: Comparing Lambda, Step Functions, App Runner, and Direct Integrations” discusses the different options for serverless development and how they compare in terms of speed. The author found that direct integrations from API Gateway to various AWS services were the fastest, followed by Step Functions and App Runner. Lambda was the slowest option, but it is still a viable choice for certain use cases. Read full review at readysetcloud.io
- AWS Announced Snapchange: an open-source fuzzing framework that uses KVM to create snapshots of the target’s memory. This allows the fuzzer to replay the snapshots and find bugs that would not be found with traditional fuzzing techniques. Snapchange is still under development but has already found several bugs in real-world software.
- Eric Pauley’s article “Farewell to the Era of Cheap EC2 Spot Instances” discusses the changing landscape of AWS spot instances, which have provided users with cost-effective compute resources. However, as macroeconomic conditions shift and businesses look to optimize their cloud bills, the savings obtained from spot instances are diminishing. Spot prices have spiked in recent months, and there has been a significant increase in spot instance demand, leading to higher preemption rates. While diversifying across equivalent instances and leveraging resilient designs can still offer some cost benefits, the erosion of spot discounts suggests a need for organizations to consider alternative approaches, such as Savings Plans, for more guaranteed savings in predictable usage scenarios.
- In their article, “Scaling up the Prime Video audio/video monitoring service and reducing costs by 90%” the team describes how Prime Video’s audio/video monitoring service was originally built using a distributed microservices architecture. However, the team found that this architecture was not scalable and expensive to operate. They decided to switch to a monolithic application running on Amazon EC2 and Amazon ECS. This change allowed them to reduce their infrastructure costs by over 90% while also increasing their scalability. The team also found the monolithic application easier to manage and maintain.