Bottlenecks and Their Performance Implications in E-commerce Systems

Zhang, Qi; Riska, Alma; Riedel, Erik; Smirni, Evgenia

doi:10.1007/978-3-540-30471-5_21

Cited by 16 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Consequently, one can only resort to synthetic workload generators to study such systems, the most prominent being the TPC-W benchmark [15]. Studies based on the TPC-W and its variants focus on bottleneck identification [1], [26], [20], and there is consensus that the CPU of the database server is almost always the bottleneck.…”

Section: Related Workmentioning

confidence: 99%

Workload-aware load balancing for clustered Web servers

Zhang

Riska

Sun

et al. 2005

IEEE Trans. Parallel Distrib. Syst.

104

View full text Add to dashboard Cite

Abstract-We focus on load balancing policies for homogeneous clustered Web servers that tune their parameters on-the-fly to adapt to changes in the arrival rates and service times of incoming requests. The proposed scheduling policy, ADAPTLOAD, monitors the incoming workload and self-adjusts its balancing parameters according to changes in the operational environment such as rapid fluctuations in the arrival rates or document popularity. Using actual traces from the 1998 World Cup Web site, we conduct a detailed characterization of the workload demands and demonstrate how online workload monitoring can play a significant part in meeting the performance challenges of robust policy design. We show that the proposed load balancing policy based on statistical information derived from recent workload history provides similar performance benefits as locality-aware allocation schemes, without requiring locality data. Extensive experimentation indicates that ADAPTLOAD results in an effective scheme, even when servers must support both static and dynamic Web pages.

show abstract

Section: Related Workmentioning

confidence: 99%

Workload-aware load balancing for clustered Web servers

Zhang

Riska

Sun

et al. 2005

IEEE Trans. Parallel Distrib. Syst.

104

View full text Add to dashboard Cite

show abstract

“…Another disadvantage is, that both the backend and the shop require several implementation changes for a secure integration: An interface for backend access and a mechanism for integration of cloud-based instances in the load-balancing need to be provided. According to Zhang et al [11] the database is the bottleneck of E-commerce applications. A further distribution of business logic could temporarily increase the workload on the backend systems on startup.…”

Section: Infrastructurementioning

confidence: 99%

Hybrid cloud architectures for the online commerce

Lackermair

2011

Procedia Computer Science

View full text Add to dashboard Cite

Cloud Computing is a relatively new computing approach which promises enhanced scalability, flexibility and cost-efficiency. In practice, however, there are many uncertainties about the usage of this way of provisioning IT resources. Concerns about evolving dependencies and security issues have arisen. As online commerce is dependent on a reliable, secure online stores, it is vital to take those concerns into account. The hybrid architecture combines traditional IT service provisioning with the usage of internet-based services. This approach would help to harness the strengths of Cloud Computing, while minimizing the impacts of its potential risks. This article discusses hybrid cloud architectures with the focus on realizing shop functionalities. Advantages and drawbacks of integration for each service type in the Cloud Computing stack are given, from which possible problems of hybrid clouds are derived. The conclusions in this article provide a conceptual basis for further research and evaluation of the use of net-based services in the field of online commerce.

show abstract

“…Consequently, one can only resort to synthetic workload generators to study such systems, the most prominent being the TPC-W benchmark for an e-commerce site [20]. Studies based on TPC focus on bottleneck identification and find a variety of causes [8,2,24,13,27]. Note that the TPC benchmarks implement stationary arrivals only, which gives a restricted view of the system: observing the system in steady state does not tell us how the systems behaves in transient overload.…”

Section: Related Workmentioning

confidence: 99%

Workload propagation - overload in bursty servers

Zhang

Riska

Riedel

2005

Second International Conference on the Quantitative Evaluation of Systems (QEST'05)

View full text Add to dashboard Cite

Internet servers are developing into complex but central components in the information infrastructure and are accessed by an ever-increasing and diversified user population. As such, they are susceptible to unpredicted and transient overloads, making highly available, yet cost-effective service a critical challenge. System responsiveness during overload periods is at least as important as system behavior under average, steady state load. Using measurements from a 3-tier e-commerce server, we show how normal and overload conditions propagate through the system hierarchy, i.e., from the front-end server to the back-end database, focusing on CPU, memory, and I/O performance. We find that overload conditions at lower system levels occur not only when the number of users in the system increases, but also during rapid changes in the nature of the work requested by the current users. These observation advocate the development of system mechanisms at the lower levels that can detect overload and self-adapt their configuration parameters in order to ensure high service availability and graceful performance degradation. We illustrate the effectiveness of a scheduling mechanism at the disk and provide a proof of concept that such self-adaptive scheduling mechanisms at the lower levels are a step toward faster system recovery under conditions of transient overload, and can thus complement admission control at the front end.

show abstract

Bottlenecks and Their Performance Implications in E-commerce Systems

Cited by 16 publications

References 12 publications

Workload-aware load balancing for clustered Web servers

Workload-aware load balancing for clustered Web servers

Hybrid cloud architectures for the online commerce

Workload propagation - overload in bursty servers

Contact Info

Product

Resources

About