DMCA
Quasar: Resource-Efficient and QoS-Aware Cluster Management
| Citations: | 23 - 5 self |
BibTeX
@MISC{Delimitrou_quasar:resource-efficient,
author = {Christina Delimitrou and Christos Kozyrakis},
title = {Quasar: Resource-Efficient and QoS-Aware Cluster Management},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Cloud computing promises flexibility and high performance for users and high cost-efficiency for operators. Neverthe-less, most cloud facilities operate at very low utilization, hurting both cost effectiveness and future scalability. We present Quasar, a cluster management system that increases resource utilization while providing consistently high application performance. Quasar employs three tech-niques. First, it does not rely on resource reservations, which lead to underutilization as users do not necessarily understand workload dynamics and physical resource re-quirements of complex codebases. Instead, users express performance constraints for each workload, letting Quasar determine the right amount of resources to meet these con-straints at any point. Second, Quasar uses classification tech-niques to quickly and accurately determine the impact of the amount of resources (scale-out and scale-up), type of resources, and interference on performance for each work-load and dataset. Third, it uses the classification results to jointly perform resource allocation and assignment, quickly exploring the large space of options for an efficient way to pack workloads on available resources. Quasar monitors workload performance and adjusts resource allocation and assignment when needed. We evaluate Quasar over a wide range of workload scenarios, including combinations of dis-tributed analytics frameworks and low-latency, stateful ser-vices, both on a local cluster and a cluster of dedicated EC2 servers. At steady state, Quasar improves resource utiliza-tion by 47 % in the 200-server EC2 cluster, while meeting performance constraints for workloads of all types.
Keyphrases
qos-aware cluster management performance constraint complex codebases quasar monitor high application performance stateful ser-vices available resource workload dynamic dis-tributed analytics framework dedicated ec2 server large space efficient way cluster management system high performance resource allocation future scalability present quasar high cost-efficiency cloud facility resource reservation right amount classification tech-niques 200-server ec2 cluster low utilization local cluster steady state workload scenario physical resource re-quirements classification result cost effectiveness promise flexibility wide range resource utilization
