Talks
Speakers
Events
Topics
Sign in
Home
Talks
Speakers
Events
Topics
Leaderboard
Use
Analytics
Sign in
Suggest modification to this talk
Title
Description
@nodunayo What does haggling at a garage sale have to do with load balancing in distributed systems? How does bidding in an art auction relate to cloud service orchestration? Familiarity with the ideas and technologies involved in cloud computing is becoming ever more important for developers. This talk will demonstrate how you can use game theory - the study of strategic decision making - to design more efficient, and innovative, distributed systems. "Talk given at GORUCO 2015: http://goruco.com"
Date
Summarized using AI?
If this talk's summary was generated by AI, please check this box. A "Summarized using AI" badge will be displayed in the summary tab to indicate that the summary was generated using AI.
Show "Summarized using AI" badge on summary page
Summary
Markdown supported
In her keynote presentation at GoRuCo 2015, Nadia Odunayo delves into the significance of game theory in enhancing the design and efficiency of distributed systems, particularly within the realm of cloud computing. The talk begins with an illustrative story about two university students needing to devise a convincing excuse for missing their chemistry exam, leading into an exploration of strategic decision-making principles. Key Points Discussed: - Game theory is introduced as the study of strategic interactions among rational agents, with applications in various fields, including economics and computer science. - The presentation outlines a simple game using bargaining theory to distribute resources efficiently in a distributed system, particularly addressing load balancing between computers. - Odunayo presents a scenario involving two computers, A and B, to explain how job allocation can be modeled as a bargaining game, demonstrating how the Nash bargaining solution helps determine optimal job distribution. - A second example employs auction theory, illustrating a second-price auction setup. The concept of dominant strategies is explained, highlighting the benefit to participants of truthfully reporting their bids based on their intrinsic valuation of the items being auctioned, in this case, a bag of sweets. - The application of auction theory extends to load allocation in distributed systems, where self-interested agents (computers) must provide accurate bids about their resource capabilities, ultimately ensuring efficient task allocation. - Real-world examples and algorithms demonstrate how game theory informs the orchestration of services within platforms like Cloud Foundry. An algorithm based on the Nash bargaining solution is discussed, showing its efficiency compared to other allocation methods. - The concept of antifragility is introduced, highlighting the ability of systems to learn and adapt through feedback in challenging situations. Conclusions and Takeaways: - Game theory provides valuable tools for understanding and solving complex problems in load distribution and resource allocation within distributed systems. - The talk emphasizes that effective strategies derived from game theory can enhance the performance of these systems by modeling component interactions and ensuring honest reporting of capabilities. - Lastly, Odunayo encourages developers to explore various aspects of game theory further, suggesting practical applications in a music of computer systems and inviting them to approach future projects with these frameworks in mind.
Suggest modifications
Cancel
