Akita: A CPU scheduler for virtualized Clouds

TL;DR

Akita is a hypervisor CPU scheduler that ensures predictable performance for high-criticality VMs while maintaining high utilization by co-locating them with low-criticality VMs through dynamic slowdown mechanisms.

Contribution

Akita introduces a novel CPU scheduling approach that guarantees SLOs for high-criticality VMs in virtualized clouds, enabling co-location with low-criticality VMs without sacrificing predictability.

Findings

High-criticality Memcached VMs maintain predictable performance.

Akita achieves higher resource utilization compared to traditional schedulers.

Prototype implementation demonstrates practical effectiveness.

Abstract

Clouds inherit CPU scheduling policies of operating systems. These policies enforce fairness while leveraging best-effort mechanisms to enhance responsiveness of all schedulable entities, irrespective of their service level objectives (SLOs). This leads to unpredictable performance that forces cloud providers to enforce strict reservation and isolation policies to prevent high-criticality services (e.g., Memcached) from being impacted by low-criticality ones (e.g., logging), which results in low utilization. In this paper, we present Akita, a hypervisor CPU scheduler that delivers predictable performance at high utilization. Akita allows virtual machines (VMs) to be categorized into high- and low-criticality VMs. Akita provides strong guarantees on the ability of cloud providers to meet SLOs of high-criticality VMs, by temporarily slowing down low-criticality VMs if necessary. Akita, therefore, allows the co-existence of high and low-criticality VMs on the same physical machine, leading to higher utilization. The effectiveness of Akita is demonstrated by a prototype implementation in the Xen hypervisor. We present experimental results that show the many advantages of adopting Akita as the hypervisor CPU scheduler. In particular, we show that high-criticality Memcached VMs are able to deliver predictable performance despite being co-located with low-criticality CPU-bound VMs.