Abstract: Techniques for spending limits in a cloud computing network are disclosed. The system receives a usage request in a cloud computing network for a user. The user having a customer tenancy in the cloud computing network including cloud computing resources. The system determines a cloud computing usage increase value relative to a time frame (such as a day) using at least the usage request. The system determines whether the cloud computing usage increase value exceeds a threshold. Based on determining that the cloud computing usage increase value exceeds the threshold, the system performs a remediation action. The remediation action comprises at least one of rejecting the usage request and bypassing the threshold.

Abstract: Various embodiments of a system and method for decentralized performance monitoring of host systems are described. Embodiments may include one or more client systems, each of which may monitor the performance of one or more host systems. In some embodiments, at least some of the host systems may be members of a cloud computing environment. A given client system may collect metrics from multiple different host systems and aggregate such metrics on the client system. In various embodiments, metrics may be collected by the client system directly from the multiple different host systems without utilizing a centralized repository of metrics. In various embodiments, the given client's receipt of the metrics from the multiple different hosts systems may be an initial aggregation of those metrics together on the same computer system. The client system may generate a graphical representation of the metrics collected from multiple hosts systems.