Instaclustr Now Offering AWS Graviton2 Instances With gp3 volumes for Apache Cassandra®

Instaclustr has released AWS Graviton2 instances and gp3 volumes for Apache Cassandra®, providing customers with both performance and financial benefits.

Instaclustr is pleased to announce the general availability of AWS instances on Graviton2 processors, combined with gp3 volumes for Instaclustr managed Apache Cassandra. AWS Graviton2 infrastructure has better compute performance than other equivalent instances using the same number of vCPUs and the same amount of RAM, so provides notable performance improvement at a reduced cost.

Instaclustr managed Apache Cassandra on AWS Graviton2 has also incorporated the new generation of AWS Elastic Block Storage (EBS) gp3 disks that have a higher IO throughput compared to EBS gp2 disks. Previous IOPS constraints on smaller instances have been eliminated with the new generation of AWS gp3 volumes that come with at least 3000 IOPS no matter the disk size, as well as the ability to specify additional IOPS and throughput independent of disk size. This is good news for our customers that previously had to provision larger than necessary disks for the purpose of achieving sufficient throughput. Through load testing, Instaclustr has matched the provisioned IOPS and throughput of the gp3 disks with the capabilities of the associated instance to achieve an overall balanced infrastructure configuration for optimum price/performance.

Instaclustr has conducted benchmarking with each of our Apache Cassandra production node sizes to compare the performance between AWS EC2 instances running x86 processors with gp2 EBS volumes against AWS EC2 instances running Graviton2 ARM-based processors with gp3 EBS volumes.The performance test results show that AWS Graviton2 instances with gp3 volumes outperforms the AWS x86 instances and gp2 volumes in the majority of tests.

In conjunction with the release of AWS Graviton2 instances with gp3 volumes, Instaclustr has also released a managed Cassandra feature that allows customers to add more disk space, or scale up and down processing capacity on-demand without downtime to their services and without intervention by Instaclustr. This is known as Customer Initiated Resize and can be performed through the Instaclustr Console, the API, or Terraform provider. This function builds on an existing feature known as Dynamic Resizing and will replace that feature and the special class of resizeable nodes that it depended on.

Customer Initiated Resize gives significant flexibility to customers through the Console, API, or Terraform provider to select any larger production disk or instance size when scaling up, and for instance size, also the ability to scale down. This is feature is very valuable in situations where you have predictable changes in load (such as a planned sale or event) as you can scale up or down processing capacity of a cluster in under an hour compared to many hours or even days for horizontal scaling, allowing you to optimize your provisioned infrastructure to meet your expected load.  

Customer Initiated Resize for Cassandra offers these key features:

• Easy access to temporary or permanently increased processing capacity.
• Safety checks conducted as a prerequisite to the resize operation.
• Resizing without downtime or data loss during the process.
• On-demand customer access to resizing for time sensitive operations without needing intervention by Instaclustr.
• Flexible resize methods with Instaclustr Console, API, or Terraform provider.
• Initial availability on AWS, with GCP and Azure cloud service providers to follow.
• Customer Initiated Resize supporting migration from Intel to Graviton2 based nodes.

This release adds to our existing Customer Initiated Resize features for Managed Apache Kafka®, OpenSearch®, and Redis® for greater on-demand functionality for our customers without the support overhead.

To start using Instaclustr Managed Cassandra on AWS Graviton2 instances with gp3 volumes, sign up for a free trial on our Console today. Or read the Instaclustr Blog post ‘Horizontal vs Vertical Scaling: Which Scaling Strategy is Right For Me?‘ to discover the scaling strategy best suited to you.