Skip to content

Whats the impact on latency for DSQL?

1

Just trying to understand how it compares to the current experience through MySQL or similar.

asked 2 years ago905 views

2 Answers
2

As with all questions like this I would encourage you to test DSQL with your workload. Every workload (and in this case every database) is different and that is going potentially be amplified by the region(s) that you choose to run your application(s) in.

EXPERT

answered 2 years ago

EXPERT

reviewed 2 years ago

1

Amazon Aurora DSQL is designed to provide high performance with low latency, but it's important to understand its unique characteristics compared to traditional databases like MySQL.

Aurora DSQL offers virtually unlimited horizontal scaling, allowing it to independently scale reads, writes, compute, and storage. This means it can automatically adapt to meet any workload demand without requiring database sharding or instance upgrades. This scalability can help maintain low latency even as your workload grows.

The active-active distributed architecture of Aurora DSQL is designed for high availability, with 99.99% single-Region and 99.999% multi-Region availability. This architecture, combined with automated failure recovery, helps minimize downtime and maintain consistent performance.

However, it's worth noting that Aurora DSQL uses optimistic concurrency control (OCC), which may have different latency characteristics compared to traditional locking mechanisms used in MySQL. The impact of this on latency would depend on your specific workload and access patterns.

Aurora DSQL's ability to scale independently across different layers (query processing, commit, and storage) allows it to adapt to workloads of various shapes, including different read/write ratios, data sizes, and query complexities. This flexibility can help maintain low latency across diverse workload types.

While specific latency numbers are not provided in the available information, the design of Aurora DSQL suggests that it aims to provide low-latency performance at scale. However, as with any database solution, actual performance would need to be tested with your specific workload to determine the exact impact on latency compared to your current MySQL or similar experience.
Sources
Announcing Amazon Aurora DSQL (Preview) - AWS
Introducing Amazon Aurora DSQL | AWS Database Blog

answered 2 years ago

EXPERT

reviewed 2 years ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.