Abstract

In cloud computing, sole tenant nodes dedicate hardware to a single customer, enabling the customer to tightly optimize utilization and performance to match their load profile. However, with sole tenancy, the burdens of capacity management, risk management, task scheduling, triggering live-migration, etc., fall on the customer. This disclosure describes techniques to enable customers to run their workloads in standard, multitenant cloud computing environments while still being able to customize the experience to levels comparable to those offered via dedicated hardware or sole tenant hosts. The high level of customization in a multitenant environment is enabled even as the customer uses the familiar API set of the public cloud. Benefits of single tenancy such as load profile optimized overcommit, custom placement, autoscaling, etc., accrue without the burdens of capacity management, task scheduling, risk management, etc. as the workloads enjoy the better statistical multiplexing offered by the standard, multitenant public cloud.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.

Share

COinS