We started our AWS Graviton journey four years ago in 2020. With AWS Graviton4-based Amazon EC2 R8g instances, we've now achieved more than double the throughput per vCPU compared to the non-Graviton based instances that we used four years ago.

As the provider of ClickHouse Cloud, a serverless real-time data warehouse, speed is deeply important, as is improving the price performance ratio of our managed service. Since the release of AWS Graviton-based Amazon EC2 instances, we have worked through the transition by building Arm architecture compatibility into our open source packages and ensuring all dependencies were met. We have been running AWS Graviton4-based Amazon EC2 R8g instances in staging environments, internal benchmarks, and as a test platform for applications (ADS-B visualization). We see that R8g instances have the potential to provide up to a 20% better price performance. We expect broad deployment of R8g instances across the control plane and data plane of the ClickHouse Cloud infrastructure.

We regularly experience spikes in traffic, and we are always looking for ways to improve performance while decreasing the cost of running our database workload on MongoDB. We evaluated Amazon EC2 R8g instances in one of our production workloads. The migration to R8g instances from AWS Graviton2-based instances took a few hours, without any code changes. We've found that the CPU load per request decreased by about 50% compared to Graviton2-based instances. This translates to approximately 2x better price performance compared to Graviton2, based on on-demand usage. We plan to expand our usage of Graviton instances in the near future.

At SmugMug and Flickr, we’re committed to delivering fast and efficient services to our photographer customers. As the first adopter of AWS Graviton-based Amazon EC2 instances in 2018, we’ve experienced years of cost savings and performance gains from AWS Graviton. Each generation of Graviton-based instances has unlocked higher levels of price/performance. These generational improvements grew faster than comparable non-Graviton instances, resulting in higher compound growth in our advancements. For image and data compression operations, we’ve measured 20 to 40% performance and capacity improvements using Graviton4-based instances compared to Graviton3-based instances.

At Datadog, we run tens of thousands of nodes, so balancing performance and cost effectiveness is extremely important. That’s why we already run half of our Amazon EC2 fleet on AWS Graviton. Integrating AWS Graviton4-based Amazon EC2 R8g instances into our environment was seamless, and gave us an immediate performance boost out of the box, and we're looking forward to using AWS Graviton4 when it becomes generally available.

AWS Graviton4-based Amazon EC2 R8g instances are the fastest Amazon EC2 instances we've ever tested, and they are delivering outstanding performance across our most competitive and latency sensitive workloads. "We look forward to using Graviton4 to improve player experience and expand what is possible within Fortnite.

In November of last year, we announced that SAP HANA Cloud was live on AWS Graviton-based instances, leading to a 35% price performance improvement for analytics workloads, and an estimated 45% reduction in carbon footprint for SAP HANA Cloud. SAP will be using AWS Graviton4-based Amazon EC2 instances for SAP HANA Cloud and additional SAP applications. Initial results for AWS Graviton4-based Amazon EC2 R8g instances with SAP HANA Cloud show up to 25% performance improvement for analytical and up to 40% improvement for transactional workloads compared to AWS Graviton3-based instances. We have also expanded the use of AWS Graviton-based instances across services in SAP BTP, and based on SAP HANA Cloud using it to power services including SAP Datasphere, and SAP CloudALM.