FIESTA: A Sample-Balanced Multi-Program Workload Methodology
نویسندگان
چکیده
Workload construction methodologies for multiprogram experiments are more complicated than those for single-program experiments. Fixed-workload methodologies pre-select samples from each program and use these in every experiment. They enable direct comparisons between experiments, but may also yield runs of which significant portions are spent executing only the slowest program(s). Variable-workload methodologies eliminate this load imbalance by using the multi-program run to define the workload, normalizing performance to the performance of the resulting individual program regions. However, they make direct comparisons difficult and tend to produce workloads that over-estimate throughput and speedup. We propose a multi-program workload methodology called FIESTA which is based on the observation that there are two kinds of load imbalance. Sample imbalance is due to differences in standalone program running times. Schedule imbalance is due to asymmetric contention during multi-program execution. Sample imbalance is harmful because it dilutes multi-program behaviors. Schedule imbalance is a characteristic of concurrent execution that should be preserved and measured. Traditional fixed-workload methodologies admit both kinds of imbalance. Variable-workload methodologies eliminate both kinds of imbalance. FIESTA is a fixed-workload methodology that eliminates only sample imbalance. It does so by pre-selecting program regions for equal standalone running times rather than for equal
منابع مشابه
Multi-Objective Heuristics for the Vehicle Routing Problem
The distribution service of commodities has tremendous impacts on cost-effective performance and level of customer satisfaction for third party logistics. One of main concerns is how to balance workloads among vehicles and ensure delivery time for each vehicle within the required duration. In this study, a multi-objective mixed zero-one integer programming model for the vehicle routing problem...
متن کاملThe relationship between visual website complexity and a user’s mental workload: A NeuroIS perspective
I report results from an experiment on the relationship between visual website complexity and users' mental workload. Applying a pupillary based workload assessment as a NeuroIS methodology, I found indications that a balanced level of navigation complexity, i.e., the number of (sub)menus, in combination with a balanced level of information complexity, is the best choice from a user’s mental wo...
متن کاملA New Multi-Objective Model for Dynamic Cell Formation Problem with Fuzzy Parameters
This paper proposes a comprehensive, multi-objective, mixed-integer, nonlinear programming (MINLP) model for a cell formation problem (CFP) under fuzzy and dynamic conditions aiming at: (1) minimizing the total cost which consists of the costs of intercellular movements and subcontracting parts as well as the cost of purchasing, operation, maintenance and reconfiguration of machines, (2) maximi...
متن کاملAssessment of Hospital Performance using Balanced Scorecard Model and Program Chain Pattern
Introduction: A Balanced Scorecard model is a performance evaluation methodology that provides a broad perspective on identifying the weaknesses and strengths of an organization's activities. The purpose of this study was to assess the performance of the hospital using a Balanced Scorecard Model and a Program Chain Pattern. Methods: This descriptive study was conducted at Imam Khomeini Educatio...
متن کاملWorkload Reduction for Multi-Input Profile-Directed Optimization
Profile-directed optimization is an effective technique to improve program performance, but it may result in program performance and compiler behavior that is sensitive to both the selection of inputs used for training and the actual input in each run of the program. Cross-validation over a workload of inputs can address the input-sensitivity problem, but introduces the need to select a represe...
متن کامل