Understanding Performance Interference of I/O Workload in Virtualized Cloud Environments

Pu, Xing; Liu, Ling; Mei, Yiduo; Sivathanu, Sankaran; Koh, Younggyun; Pu, Calton

doi:10.1109/cloud.2010.65

Cited by 156 publications

(76 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The impact of such contention is well pronounced in cloud datacenters due to resource interference between multiple cloud tenants. The noisy neighbours e↵ect is an analogy for this interference [20]. Several contention scenarios are well known for di↵erent system resources.…”

Section: Resource Contention Bottlenecksmentioning

confidence: 99%

Performance Anomaly Detection and Bottleneck Identification

2015

View full text Add to dashboard Cite

In order to meet stringent performance requirements, system administrators must e↵ectively detect undesirable performance behaviours, identify potential root causes and take adequate corrective measures. The problem of uncovering and understanding performance anomalies and their causes (bottlenecks) in di↵erent system and application domains is well studied. In order to assess progress, research trends and identify open challenges, we have reviewed major contributions in the area and present our findings in this survey. Our approach provides an overview of anomaly detection and bottleneck identification research as it relates to the performance of computing systems. By identifying fundamental elements of the problem, we are able to categorize existing solutions based on multiple factors such as the detection goals, nature of applications and systems, system observability, and detection methods.

show abstract

Section: Resource Contention Bottlenecksmentioning

confidence: 99%

Performance Anomaly Detection and Bottleneck Identification

2015

View full text Add to dashboard Cite

show abstract

“…For example, latency degradation due to resource contention between VMs colocated on the same hardware has been studied by [3][5] [6]. These studies conclude that improvements in VM resource scheduling are needed so that less contention leads to better performance.…”

Section: Related Workmentioning

confidence: 99%

Cloud Support for Latency-Sensitive Telephony Applications

Kim

Schulzrinne

2013

2013 IEEE 5th International Conference on Cloud Computing Technology and Science

View full text Add to dashboard Cite

Abstract-Cloud computing is great for scaling applications but the latency in a guest VM can be unpredictable due to resource contention between neighbors. For telephony applications, which are latency-sensitive, we propose a system to monitor telephony server latencies and adapt the server load based on the measured latencies. We implemented the system and evaluated it on an Amazon EC2 testbed. We show indirectly by comparing our server on EC2 and on a local VM, that there may be contention between EC2 VMs in the wild that leads to higher server latency. While there is some overhead due to constant monitoring of the server, our system manages to lower latency by reducing the load to the server.

show abstract

“…One of the recognized problems is the performance unpredictability in virtualized environments (ranked fifth in the top 10 obstacles for growth of cloud computing [4]). Performance interferences among consolidated applications have been demonstrated for a variety of concrete systems and applications [3], [8], [11], [15].…”

Section: Introductionmentioning

confidence: 99%

Revisiting Performance Interference among Consolidated n-Tier Applications: Sharing is Better Than Isolation

Kanemasa

Wang

et al. 2013

2013 IEEE International Conference on Services Computing

Self Cite

View full text Add to dashboard Cite

Abstract-Performance unpredictability is one of the major concerns slowing down the migration of mission-critical applications into cloud computing infrastructures [4]. An example of non-intuitive result is the measured n-tier application performance in a virtualized environment that showed increasing workload caused a competing, co-located constant workload to decrease its response time [12]. In this paper, we investigate the sensitivity of measured performance in relation to two factors: (1) consolidated server specification of virtual machine resource availability, and (2) burstiness of n-tier application workload. Our first and surprising finding is that specifying a complete isolation, e.g., 50-50 even split of CPU between two co-located virtual machines (VMs) results in significantly lower performance compared to a fully-shared allocation, e.g., up to 100% CPU for both co-located VMs. This happens even at relatively modest resource utilization levels (e.g., 40% CPU in the VMs). Second, we found that an increasingly bursty workload also increases the performance loss among the consolidated servers, even at similarly modest utilization levels (e.g., 70% overall). A potential solution to the first problem (performance loss due to resource allocation) is crosstier-priority scheduling (giving higher priority to shorter jobs), which can reduce the performance loss by a factor of two in our experiments. In contrast, bursty workloads are a more difficult problem: our measurements show they affect both the isolation and sharing strategies in virtual machine resource allocation.

show abstract

Understanding Performance Interference of I/O Workload in Virtualized Cloud Environments

Cited by 156 publications

References 8 publications

Performance Anomaly Detection and Bottleneck Identification

Performance Anomaly Detection and Bottleneck Identification

Cloud Support for Latency-Sensitive Telephony Applications

Revisiting Performance Interference among Consolidated n-Tier Applications: Sharing is Better Than Isolation

Contact Info

Product

Resources

About