Posted On: Jan 6, 2022
Starting today, customers can use Amazon EC2 On-Demand Capacity Reservations to reserve capacity for cluster placement groups. With cluster placement groups, customers can launch EC2 instances into logical groups within a segment of the network with high bisection bandwidth, thus getting low latency and high throughput between instances inside the cluster. Cluster placement groups are beneficial for customers with workloads that require tightly coupled node-to-node communication, such as high-performance computing (HPC) workloads or in-memory databases like SAP HANA. With the addition of On-Demand Capacity Reservations for cluster placement groups, customers can get the assurance of reserved capacity as they scale compute resources within their cluster.
To reserve capacity for a cluster placement group, customers can now specify the associated Placement Group ARN while creating an On-Demand Capacity Reservation. When an instance is launched into the placement group, it will be assigned to the Capacity Reservation with matching attributes (instance type, platform, Availability Zone, tenancy, and Placement Group ARN) provided there is unused reserved capacity.
This feature is now available to all Capacity Reservations customers in all commercial Amazon Web Services Regions, including Amazon Web Services China (Beijing) Region, operated by Sinnet, and Amazon Web Services China (Ningxia) Region, operated by NWCD, at no additional cost. You can create and manage Capacity Reservations through the Amazon Web Services Management Console, Amazon SDK, or Amazon CLI. To learn more about this feature, please refer to the Capacity Reservations support for Cluster Placement Groups user guide. To learn more about Capacity Reservations and how to use them, visit FAQs, Linux or Windows user guide. To learn more about EC2 cluster placement groups, visit the Linux or Windows user guide.