Impact of user patience on auto-scaling resource capacity for cloud services

Abstract

h i g h l i g h t s• Mechanisms for resource auto-scaling in clouds considering users' patience.• Methods for determining the step size of scaling operations under bound and unbounded maximum capacity. • Users patience model inspired in prospect theory. a b s t r a c tAn important feature of most cloud computing solutions is auto-scaling, an operation that enables dynamic changes on resource capacity. Auto-scaling algorithms generally take into account aspects such as system load and response time to determine when and by how much a resource pool capacity should be extended or shrunk. In this article, we propose a scheduling algorithm and auto-scaling triggering strategies that explore user patience, a metric that estimates the perception end-users have from the Quality of Service (QoS) delivered by a service provider based on the ratio between expected and actual response times for each request. The proposed strategies help reduce costs with resource allocation while maintaining perceived QoS at adequate levels. Results show reductions on resource-hour consumption by up to approximately 9% compared to traditional approaches.

DIaaS: Resource Management System for the Intra-Cloud with On-Premise Desktops

“…VCS simulator [24] Text-based Text User-defined Yes GridSim [25,26] Modified source code Chart None Yes ClusterSim [27] Clustering-based Text User-defined Yes CloudSim [28] Text …”

“…First, to measure the clustering delay time, the number of desktops was increased to 100, 200, 300, 400, 500, 1000, 2000, 3000, 4000, and 5000, as shown in Figure 16. Clustering was performed after the number of clusters was increased to 5,10,15,20,25, and 30 at each increasing point. The clustering speed at each increasing point was determined by averaging the performance speeds obtained from 50 cycles.…”

“…As shown in Figure 15, this process was more efficient and cost effective than the clustering test using 5000 actual desktops and the related cost of labor, electricity, and time. clusters was increased to 5,10,15,20,25, and 30 at each increasing point. The clustering speed at each increasing point was determined by averaging the performance speeds obtained from 50 cycles.…”

“…Existing studies on distributed resource clustering techniques and clustering visualization are outlined in Table 2 [23][24][25][26][27]. Table 2.…”

SLA-aware resource allocation for application service providers in the cloud

“…Moreover, they didn't consider the user behavior. Finally, authors in [23] combined an algorithm for automated scaling based on prediction, resource utilization with user behavior analyzer. Their method assessed the user behavior based on the expected response time and the satisfaction level of users.…”