Posted On: Apr 8, 2020
Starting today, you can launch Amazon EMR clusters with Amazon EC2 G3s, Z1d, M5a, and R5a instances.
Amazon EC2 M5a and R5a instances can now be used with Amazon EMR in the Amazon Web Services China (Beijing) Region, operated by Sinnet and the Amazon Web Services China (Ningxia) Region, operated by NWCD. Both M5a and R5a instances feature 2.5 GHz AMD EPYC 7000 series processors that are variants of Amazon EC2’s general purpose M5 and memory optimized R5 instance families. The AMD-based instances provide additional options for customers who are looking to achieve up to 10% cost savings on their Amazon EC2 compute environment for Amazon EMR workloads.
Amazon EC2 z1d instances can now be used with Amazon EMR in the Amazon Web Services China (Ningxia) Region. Amazon EC2 z1d instances deliver a sustained all core frequency of up to 4.0 GHz. z1d instances feature a custom Intel® Xeon® Scalable processor that can run all cores at up to 4.0 GHz sustained. z1d instances include up to 1.8TB of local instance storage with up to 48 vCPUs and 384 GiB memory.
Amazon EC2 g3s.xlarge instances can now be used with Amazon EMR in the Amazon Web Services China (Beijing) Region. Like other G3 instances, the g3s.xlarge is powered by NVIDIA Tesla M60 GPUs but is designed to be cost-effective for workloads that don’t need the high vCPU and RAM that current larger G3 instance sizes provide.
To learn more, visit the Amazon EC2 Instance Types page and the Amazon Elastic MapReduce Pricing page.