Amazon EMR Serverless introduces account-level vCPU-based service quota
Amazon EMR Serverless is a serverless option in Amazon EMR that makes it simple for data engineers and data scientists to run open-source big data analytics frameworks without configuring, managing, and scaling clusters or servers. Today we are introducing a new service quota called Max concurrent vCPUs per account. This vCPU-based quota allows you to set the maximum number of aggregate vCPUs your applications are able to scale up to within a Region.
Amazon EMR Serverless is a serverless option in Amazon EMR that makes it simple for data engineers and data scientists to run open-source big data analytics frameworks without configuring, managing, and scaling clusters or servers. Today we are introducing a new service quota called Max concurrent vCPUs per account. This vCPU-based quota allows you to set the maximum number of aggregate vCPUs your applications are able to scale up to within a Region.
With this feature, Amazon EMR Serverless provides you with two controls to help manage costs. You can view and increase this vCPU-based quota in the AWS Service Quotas Management console to set the maximum concurrent active vCPUs for all applications in your account. You can also configure the application property maximum capacity in terms of vCPU, memory (GB), and disk (GB) to define how much an individual application can scale up to. Note that the application-level quota (Maximum active workers) that allowed you to set maximum capacity at an application level is deprecated as the application property maximum capacity allows you to do the same. For e.g. if you have 5 applications and each application can scale up to 1000 vCPUs, set the maximum capacity property to 1000 vCPUs for each application and request the account-level vCPU-based quota to be set to 5 * 1000 = 5000 vCPUs.
This new quota is available in all the regions where EMR Serverless is available. Visit our documentation or read more in the FAQs to learn about this vCPU-based quota.