Agile dynamic provisioning of multi-tier Internet applications

Abstract

-Dynamic capacity provisioning is a useful technique for handling the multi-time-scale variations seen in Internet workloads. In this paper, we propose a novel dynamic provisioning technique for multi-tier Internet applications that employs (i) a flexible queuing model to determine how much resources to allocate to each tier of the application, and (ii) a combination of predictive and reactive methods that determine when to provision these resources, both at large and small time scales. We propose a novel data center architecture based on virtual machine monitors to reduce provisioning overheads. Our experiments on a forty-machine Xen/Linux-based hosting platform demonstrate the responsiveness of our technique in handling dynamic workloads. In one scenario where a flash crowd caused the workload of a three-tier application to double, our technique was able to double the application capacity within five minutes, thus maintaining response time targets. Our technique also reduced the overhead of switching servers across applications from several minutes to less than a second, while meeting the performance targets of residual sessions.

Resource pool management: Reactive versus proactive or let’s be friends

“…Commonly available resource demand traces are the basis for our management system. However, there are many research papers, e. g., [36,37], which design dynamic provisioning systems for targeted classes of applications, e. g., multi-tier applications. There are excellent earlier works on load sharing systems that support batch-like workloads [38,39].…”

Robust Performance Control for Web Applications in the Cloud

“…(Urgaonkar et al, 2008) designed and imple- mented a predictive and reactive mechanism using a queuing model to decide the number of resources to be provisioned, and an admission control mechanism to face extreme workload variations. Even though the use of admission mechanisms enforce the performance requirements, it reduces the QoS of the service, and therefore affects user experience.…”

“…Faced with this difficult scientific challenge, the academic community has proposed a wide range of sophisticated resource provisioning algorithms (Dejun et al, 2011;Muppala et al, 2012;Urgaonkar et al, 2008;Vasić et al, 2012). However, we observe a wide discrepancy between these academic propositions and the very simple mechanisms that are currently available to cloud customers.…”

A Pluggable Autoscaling Service for Open Cloud PaaS Systems

“…[1] and [10] focus on developing new elasticity techniques for web applications. Two main differences exist between these works and the pluggable autoscaling system proposed here.…”

“…While this work focuses on using hot spares to decrease the startup time for new nodes, [1] uses hot spares to mitigate potential SLA violations that can occur if their adaptive autoscaler puts too much workload on their own nodes. In contrast, [10] targets machines running over the Xen hypervisor, and does not use it as a research platform, opting instead to focus on elasticity algorithms customized for the three-tier web deployment strategy. Second, these efforts do not seek to provide a pluggable autoscaling solution for researchers to experiment with.…”