Abstract: Presenting workload performance in a storage system, including: receiving, via a user interface, information describing a potential change to an execution environment of the storage system; and displaying, via the user interface and in dependence upon a load model and predicted characteristics of one or more workloads executing on the storage system, predicted performance load on the storage system that would result from implementing the potential change.

Abstract: Technologies are described herein for deploying an update to hosts in a heterogeneous host fleet. One or more relevant host attributes for hosts in the host fleet are identified. Hosts in the fleet of hosts are classified into groups based on the identified relevant host attributes. A pilot host set is built by selecting one or more hosts from one or more of the groups according to a pilot host selection criteria. The update is deployed to the hosts included in the pilot host set as a test deployment. If the test deployment is successful, the update is deployed to the remaining hosts in the fleet of hosts; otherwise, the deployment is modified to continue the deployment to a subset of the host or to cancel the deployment and to restore the hosts in the pilot host set to their previous states. An analysis may be conducted based on the deployment data to provide recommendations for future deployments.

Abstract: Deployment feedback for updates to resources implemented in a private network may be implemented. Feedback codes may be generated and included in deployments sent to a private network for deployment at resources implemented in the private network. One or more of the included feedback codes may be selected based on the performance of the deployment and provided via a feedback communication channel that is disconnected and distinct from the private network. Once received, a current status of the deployment may be determined based on the one or more feedback codes provided via the feedback communication channel.

Abstract: Customers wanting to deploy software packages, or updates to those packages, across a group of servers or other computing resources can rely upon a component such as a resource manager to manage the deployment. The resource manager can utilize a data structure that stores deployment information by Revision number, and merges information for each verified deployment into a Mainline for those resources. Each Deployment can involve an Individual Release or a Baseline Release, and the importance of those Releases can be determined with respect to a current snapshot of the Mainline. Such an approach enables important Release and Deployment information to be quickly determined and obtained, which can help with configuring and scheduling future Deployments.

Abstract: A distributed execution environment includes various resources, such as instances of computing resources, hardware resources, and software resources. Values for attributes of the resources are collected. The collected attribute values for the resources are utilized to compute minhash values for the resources that describe the similarity between the resources. The computed minhash values are then utilized to generate a resource similarity visualization that provides a visual indication of the similarity between the resources.