Amazon EC2 I8g Instances

Aerospike customers run real-time workloads at scale where milliseconds matter. By leveraging Amazon EC2 I8g instances powered by AWS Graviton4 processors and 3rd generation AWS Nitro SSDs, we’ve achieved groundbreaking performance improvements—delivering up to 6X more throughput and 90% faster tail latencies compared to I4g instances. This not only drives significant cost savings but also opens up new revenue opportunities by ensuring critical transactions are processed on time. We’re excited to see our customers harness these advances to elevate their real-time applications to new heights.

At Elastic, we are driving innovation and cost-efficiency by enabling customers to leverage our Search AI-powered observability, security, and search solutions on Arm-based architecture. We tested the Amazon EC2 I8g instances, the latest generation storage-optimized instances, powered by AWS Graviton4 processors and the 3rd generation AWS Nitro SSDs. Across the benchmarks that represent the performance of common use-case patterns of customers using Elasticsearch, we saw up to 30% better price-performance compared to the previous generation Amazon EC2 I4g instances. We are excited about the introduction of I8g instances which can help us realize more cost-efficiency on AWS Graviton platform.

HubSpot uses Elasticsearch clusters running on Amazon EC2 instances to handle hundreds of millions of search and index requests daily. We are transitioning to OpenSearch and migrating our infrastructure to AWS Graviton-based Amazon EC2 instances for better cost efficiency. In our performance tests, the new cluster setup with Amazon EC2 I8g instances showed significant improvements over previous generation instances: up to 20% lower CPU usage, 20% better p99 search and index latencies, 40% better bulk index latencies, and 30% better delete latencies. These results translate to 100ms faster client search response times compared to the previous generation instances.

Elastic is committed to driving user innovation and cost-efficiency with the Search AI Platform that powers observability, security and search solutions on Arm-based architectures. In our testing, Amazon EC2 i8ge instances provide up to 21% better indexing throughput compared to the Amazon EC2 I7ie instances. We are looking forward to i8ge instances, which promise to provide improved performance and cost-efficiency for Elasticsearch users deploying on AWS.

Yuvraj Gupta, Principal Product Manager, Elastic

Our transition to AWS Graviton4-based Amazon EC2 instances has delivered impressive results, particularly in addressing our performance needs. With the Amazon EC2 I8ge instances preview, we're seeing approximately 40% reduction in I/O wait times. The migration from previous generation Graviton-based instances was straightforward, taking just a few days, and the price-performance benefits have made it a worthwhile investment for Sprinklr's unified customer experience management platform. AWS's support throughout this process, including early access to preview instance types, has been instrumental in our successful adoption.

Nitin Goyal, VP Engineering, Sprinklr

In the AI era, infrastructure must be both powerful and cost-effective. Our evaluation of the new AWS Graviton4-based Amazon EC2 I8ge instances shows that they deliver on both fronts. The transition was remarkably simple, and with the new instances, NeuralMesh will be able to handle over 20% more concurrent AI jobs and deliver 25% greater throughput on the same infrastructure. This promises to help our customers achieve new levels of efficiency and extract more value from their investment, directly lowering the cost of innovation. This compelling price performance makes enterprise AI more accessible and sustainable, and we are excited to build on this foundation.

Meltwater's main Elasticsearch cluster indexes billions of new documents and serves millions of search requests every day to help our customers track conversations as they happen. Storage optimized Amazon EC2 I8ge instances represent the single biggest performance boost across search, indexing, and analytics that we have seen in the last decade — more than a 2x performance compared to Amazon EC2 I3en instances. Switching to AWS Graviton4 enables our customers to enjoy an even snappier search experience and allows us to further increase the freshness and breadth of data available for instant analysis. We are equally excited about the potential for accelerating our vector search applications by switching from Amazon EC2 M7gd instances to Amazon EC2 I8g instances, especially for our multi-turn agent elastic cluster.

Karl Neyvaldt, Director of Search, Meltwater

By migrating to AWS Graviton4-based Amazon EC2 instances, we have transformed our operational performance. For example, after upgrading to Amazon EC2 I8g instances, we see 133% better performance in the MongoDB database tasked with 1 billion daily record updates. A critical server infrastructure that manages more than 300,000 concurrent websocket connections experienced similar gains: a 66% performance improvement after transitioning from Amazon EC2 M6g instances to Amazon EC2 M8g instances. These enhancements have not only reduced our infrastructure costs but also enabled us to deliver faster, more reliable data services to our clients, strengthening our position as the world's leading web data platform.