A Synthetic Workload Generation Technique for Stress Testing Session-Based Systems

Krishnamurthy, Diwakar; Rolia, Jerry; Majumdar, Shikharesh

doi:10.1109/tse.2006.106

Cited by 84 publications

(83 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, in a real system we may see a transition from the "delete item" state to the "purchase" state, based on the relatively common scenario in which a customer places several potential buys in the cart while browsing, and then makes the final selection among them when checking out. But a simple Markov chain cannot capture such dependencies, and might erroneously create a sequence where a single item is placed in the cart and deleted, but the client nevertheless proceeds to the checkout [416]. An alternative that does not fall into this trap is to base the model on "sessionlets" -excerpts of real sessions that contain legitimate sequences that can then be stringed together.…”

Section: Sequences Of Operationsmentioning

confidence: 99%

Workload Modeling for Computer Systems Performance Evaluation

Feitelson

2015

191

131

View full text Add to dashboard Cite

Reliable performance evaluations require the use of representative workloads. This is no easy task since modern computer systems and their workloads are complex, with many interrelated attributes and complicated structures. Experts often use sophisticated mathematics to analyze and describe workload models, making these models difficult for practitioners to grasp. This book aims to close this gap by emphasizing the intuition and the reasoning behind the definitions and derivations related to the workload models. It provides numerous examples from real production systems, with hundreds of graphs. Using this book, readers will be able to analyze collected workload data and clean it if necessary, derive statistical models that include skewed marginal distributions and correlations, and consider the need for generative models and feedback from the system. The descriptive statistics techniques covered are also useful for other domains.

show abstract

Section: Sequences Of Operationsmentioning

confidence: 99%

Workload Modeling for Computer Systems Performance Evaluation

Feitelson

2015

191

131

View full text Add to dashboard Cite

show abstract

“…Web application performance modeling uses the method [13] based on session description. A session is a collection of multiple transactions; for example, TPC-W contains 14 basic transaction types [14].…”

Section: Web Application Performance Modeling Based On Queuingmentioning

confidence: 99%

Resource Allocation Algorithm Based on Profit Maximization for Crowd Sensing

Gao

Wang

Xin-wu

2015

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

The key to realizing the crowd sensing network is to overcome the resource restrictions of energy, bandwidth, computing, and so on. First of all, due to the number of users and sensor availability will be dynamic change over time, crowd sensing system is difficult to accurately predict and allocate resource to accomplish a specific task. Secondly, there is a need to consider how to choose an effective subset from a large number of users with different sensing ability, so as to allocate the sensing devices in communication resources under the constraint conditions. This paper proposes a profit maximization algorithm for resource allocation component in crowd sensing environment. The proposed algorithm not only considers the current profit of crowd sensing service request but also considers the long-term expected profits, so as to ensure long-term maximum profit. The objective function is no longer to minimize the completion time but rather to achieve the target profit maximization. The experimental results show that the new algorithm is feasible and superior to the traditional algorithms.

show abstract

“…This general technique has been adapted to many domains [22,24,27]. While workload generation is useful for stress testing software and for achieving high path coverage, it has not been shown suitable for the creation of indicative workloads.…”

Section: Context and Motivationmentioning

confidence: 99%

The road not taken: Estimating path execution frequency statically

Buse

Weimer

2009

2009 IEEE 31st International Conference on Software Engineering

View full text Add to dashboard Cite

A variety of compilers, static analyses, and testing frameworks rely heavily on path frequency information. Uses for such information range from optimizing transformations to bug finding. Path frequencies are typically obtained through profiling, but that approach is severely restricted: it requires running programs in an indicative environment, and on indicative test inputs.We present a descriptive statistical model of path frequency based on features that can be readily obtained from a program's source code. Our model is over 90% accurate with respect to several benchmarks, and is sufficient for selecting the 5% of paths that account for over half of a program's total runtime. We demonstrate our technique's robustness by measuring its performance as a static branch predictor, finding it to be more accurate than previous approaches on average. Finally, our qualitative analysis of the model provides insight into which source-level features indicate "hot paths."

show abstract

A Synthetic Workload Generation Technique for Stress Testing Session-Based Systems

Cited by 84 publications

References 14 publications

Workload Modeling for Computer Systems Performance Evaluation

Workload Modeling for Computer Systems Performance Evaluation

Resource Allocation Algorithm Based on Profit Maximization for Crowd Sensing

The road not taken: Estimating path execution frequency statically

Contact Info

Product

Resources

About