Posted On: Nov 30, 2016
You can now launch R4 instances, the next generation of Amazon EC2 Memory Optimized instances, featuring a larger instance size, a new Intel Broadwell processor, improved networking and EBS performance, and cheaper on per GiB of RAM compared to R3 instances. R4 instances are well-suited for memory-intensive, latency-sensitive workloads like Business Intelligence (BI), data mining & analysis, in-memory databases, distributed web scale in-memory caching, and applications performing real-time processing of unstructured big data.
R4 instances offer up to 64 vCPUs and 488 GiB of DDR4 memory, and are powered by dual socket Intel® Xeon® E5 Broadwell processors (2.3 GHz). R4 provides up to 20 Gbps of network bandwidth with Elastic Network Adapter (ENA) based Enhanced Networking and 12 Gbps of dedicated throughput to EBS. Customers using Enhanced Networking on R4 instances can benefit from significantly improved network performance. The smaller instance sizes (r4.large, r4.xlarge, r4.2xlarge, and r4.4xlarge) offer up to 10 Gbps of peak network throughput, while customers can drive up to 20 Gbps of traffic in each direction on the largest R4 instance size (r4.16xlarge). All the R4 instance sizes deliver high packet per second performance with consistently low latencies using ENA. We expect these improvements in network performance to benefit a number of workloads including data analytics, batch processing, high performance databases, in-memory analytics and genome assembly and analysis.
R4 instances are offered On-Demand or as Reserved Instances in US East, US West, EU, China, Asia Pacific, South America, and Amazon Web Services GovCloud. To learn more, visit the Amazon EC2 Pricing page or Amazon Web Services China (Beijing) region EC2 Pricing Page for China pricing. The Amazon Web Services China (Beijing) region is operated by SINNET.