EC2 Performance Analysis for Resource Provisioning of Service-Oriented Applications

Abstract

Abstract. Cloud computing is receiving increasingly attention as it provides infinite resource capacity and "pay-as-you-go" resource usage pattern to hosted applications. To maintain its SLA targets, resource provisioning of service-oriented applications in the cloud requires reliable performance from the cloud resources. In this paper, we study performance behavior of small instances in Amazon EC2. We demonstrate that the performance of virtual instances is relatively stable over time with fluctuations of mean response time within at most 8% of the long-term average. Moreover, we also show that different supposedly identical instances often have very different performance, up to a ratio 4 from each other. We consider this as an important issue that must be addressed, but also as an opportunity as it allows one to assign each instance with a task that matches its own performance profile.

DynamicCloudSim: Simulating heterogeneity in computational clouds

“…Moreover, many data-intensive tasks are neither computational-nor communication-intensive, but primarily I/O-bound. Especially in database applications, a substantial amount of tasks involves reading or writing large amounts of data to local or network storage [7]. …”

“…The effect of these factors of uncertainty and instability is not negligible and has been repeatedly observed to strongly influence the runtime of a given application on commercial clouds such as Amazon's Elastic Compute Cloud (EC2) (e.g., [7][8][9][10][11][12][13]). …”

“…In a performance analysis spanning multiple Amazon EC2 datacenters, Dejun et al [7] observed occasional severe performance drops in virtual machines, which would cause the response time of running tasks to greatly increase. Furthermore, they reported the response time of CPU-and I/Ointensive application to vary by a factor of up to four on VMs of equal configuration.…”

Robust Performance Control for Web Applications in the Cloud

“…2) Resources heterogeneity: The performance of virtual instances provided by current clouds is largely heterogeneous, even among instances of the same type (Dejun et al, 2009). Simple trigger-based provisioning systems do not take this heterogeneity into account, thus providing less efficient resource allocation.…”

Applying Enterprise Architecture to Business Networks

“…Depending on the architectural design of these combinations of atomic services and workflow patterns, different sets of resources might be used or shared to meet the required performance in the service level agreement. For example, in Figure 2.7 (taken from [Dejun et al, 2010]) services 2 and 3 report their performance promise to service 1 but the performance promises of service 2 are an aggregate of services 4, 5, and 6. These resources include humans for manual services to another service for fully automated services.…”