Building Reusable Repertoires for Stochastic Self-* Planners

Kinneer, Cody; Tonder, Rijnard van; Garlan, David; Goues, Claire Le

doi:10.1109/acsos49614.2020.00045

Cited by 6 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [31], authors perform an adaptation of the search space granularity to have a trade-off between precise decisions and performance, using a constructivist approach. Other approaches use plannings and re-use of sequences of actions to guide decisions [32]- [34]. While these works map control properties to software qualities in the face of uncertainty, they do not perform evaluations with large dimensions.…”

Section: Related Workmentioning

confidence: 99%

Towards a Methodology for Building Dynamic Urgent Applications on Continuum Computing Platforms

Balouek-Thomert

Lefèvre

Caron

et al. 2022

2022 First Combined International Workshop on Interactive Urgent Supercomputing (CIW-IUS)

View full text Add to dashboard Cite

Advanced cyberinfrastructure aims at making the use of streaming data a common practice in the scientific community. They offer an ecosystem that links data, compute, network, and users to deliver knowledge obtained from multiple data sources using large-scale computational models. However, integrating this heterogeneous data with time-sensitive systems is difficult due to a lack of programming abstractions that can allow data-driven reactive behaviors throughout the edge-tocloud/HPC computing continuum. Here we present a methodology for incorporating contextual information into the application logic while taking into consideration the heterogeneity of the underlying platform and the unpredictability of the data. A fire science scenario that includes sensors at the network's edge for smoke detection and computational models launched in the cloud for wildfire simulation and air quality assessment serves as the inspiration for this method. We then discuss research directions for tackling similar scenarios with a particular focus on resource management and programming models.

show abstract

Section: Related Workmentioning

confidence: 99%

Towards a Methodology for Building Dynamic Urgent Applications on Continuum Computing Platforms

Balouek-Thomert

Lefèvre

Caron

et al. 2022

2022 First Combined International Workshop on Interactive Urgent Supercomputing (CIW-IUS)

View full text Add to dashboard Cite

show abstract

“…Note that in this work, we do not address the important issue of how to select plans for reuse, and focus attention on reusing a single plan at a time. We consider how a self-* planner can reuse a repertoire of saved plans that are amenable to reuse and likely to generalize to future situations in other work [30]. To generate the starting plan, we generate a plan for the baseline scenario using the GP planner initialized with a population of randomly generated plans, rather than seeding the population with prior plans.…”

Section: Plan Reusementioning

confidence: 99%

“…In our other work [30], we apply plan reuse to repertoires of plans, while in this work we focus on reusing a single plan at a time. Reusing repertoires of plans raises additional challenges to successful plan reuse, since reusing multiple large preexisting plans introduces even more evaluation overhead.…”

Section: Related Work Planningmentioning

confidence: 99%

Information Reuse and Stochastic Search

Kinneer

Garlan

Goues

2020

ACM Trans. Auton. Adapt. Syst.

Self Cite

View full text Add to dashboard Cite

Many software systems operate in environments of change and uncertainty. Techniques for self-adaptation allow these systems to automatically respond to environmental changes, yet they do not handle changes to the adaptive system itself, such as the addition or removal of adaptation tactics. Instead, changes in a self-adaptive system often require a human planner to redo an expensive planning process to allow the system to continue satisfying its quality requirements under different conditions; automated techniques must replan from scratch. We propose to address this problem by reusing prior planning knowledge to adapt to unexpected situations. We present a planner based on genetic programming that reuses existing plans and evaluate this planner on two case-study systems: a cloud-based web server and a team of autonomous aircraft. While reusing material in genetic algorithms has been recently applied successfully in the area of automated program repair, we find that naively reusing existing plans for self- * planning can actually result in a utility loss. Furthermore, we propose a series of techniques to lower the costs of reuse, allowing genetic techniques to leverage existing information to improve utility when replanning for unexpected changes, and we find that coarsely shaped search-spaces present profitable opportunities for reuse.

show abstract

“…Traditionally, the analysis of the data monitored by an autonomic system is based on a combination of modeling, thresholds and forecasting methods to predict the future system state and define the most suitable system configuration to reach the predefined goals. [4][5][6] One of the main research trends of autonomic computing application is self-adaptive Cloud systems. [7][8][9][10][11][12][13][14] Indeed, within large datacenters running some large-scale applications, the autonomicity of the underlying systems is a key feature to meet the performance, availability and service level agreements (SLAs) while ensuring a cost effectiveness of offered services.…”

Section: Introductionmentioning

confidence: 99%

Designing a methodological framework for modeling and performance forecasting of self‐adaptive cloud systems

Sliem

Salmi

Ioualalen

2021

Concurrency and Computation

View full text Add to dashboard Cite

Autonomic computing got an increasing attention during past years in the context of Cloud computing in order to optimize the Cloud systems performance and increase their cost-effectiveness. In that direction, a broad range of various self-adaptability strategies have been proposed to overcome uncertainties related to Cloud systems like workload variations, hardware failures and malicious attacks. A special attention in the literature has been paid to resource allocation for cost and performance optimization, and self-healing for better faut-tolerance. However, while a lot of progress has been done on these topics, there is still a lack of a standardized methodology, easily reproducible and allowing validation and comparison of a self-adaptability strategies.This article proposes a methodology allowing to forecast the autonomic Cloud systems performances. The idea is to compare different self-adaptability approaches at the design stage, and identify the most adequate and optimized configuration. This methodology is based on a new architecture description language and a stochastic Petri nets-based modeling and simulation approach. We illustrate our methodology in this article through a running example and a set of experiments.

show abstract

Building Reusable Repertoires for Stochastic Self-* Planners

Cited by 6 publications

References 24 publications

Towards a Methodology for Building Dynamic Urgent Applications on Continuum Computing Platforms

Towards a Methodology for Building Dynamic Urgent Applications on Continuum Computing Platforms

Information Reuse and Stochastic Search

Designing a methodological framework for modeling and performance forecasting of self‐adaptive cloud systems

Contact Info

Product

Resources

About