Specialization of datacenter resources brings performance and energy improvements in response to the growing scale and diversity of cloud applications. Yet heterogeneous hardware adds complexity and volatility to latency-sensitive applications. A resource allocation mechanism that leverages architectural principles can overcome both of these obstacles.We integrate research in heterogeneous architectures with recent advances in multi-agent systems. Embedding architectural insight into proxies that bid on behalf of applications, a market effectively allocates hardware to applications with diverse preferences and valuations. Exploring a space of heterogeneous datacenter configurations, which mix server-class Xeon and mobile-class Atom processors, we find an optimal heterogeneous balance that improves both welfare and energyefficiency. We further design and evaluate twelve design points along the Xeon-to-Atom spectrum, and find that a mix of three processor architectures achieves a 12× reduction in response time violations relative to equal-power homogeneous systems. Heterogeneity -Principles and StrategiesThe largest datacenters today are equipped with highperformance processors. Despite diversity due to process technology or generations, these cores all reside at the highperformance end of the design spectrum. Thus, we refer to the processors in state-of-the-art datacenters as homogeneous by design. While such homogeneity can provide near-uniform performance, it also keeps datacenters from exploiting recent advances in energy-efficient hardware. For example, small processor cores are far more power efficient than conventional, high-performance ones. Since only certain tasks are amenable to small core execution, big cores must also remain as guarantors of service quality. Heterogeneity as a Design SpaceServer heterogeneity is efficient but requires sophisticated resource managers to balance performance risk and reward. This balance requires a novel type of design space exploration to survey and appraise a variety of datacenter configurations. To illustrate the challenge, Figure 1 depicts the design space for two core types: a high-performance, server-class core and its low-power, mobile-class counterpart. Combinations of these two processor types fall into three regions shown in the Venn diagram. Two regions represent homogeneous configurations, where the datacenter is comprised of only server or mobile cores. Heterogeneous mixes lie in the third region, the intersection of the sets.The colorbar shows the percentage of allocation intervals that suffered a quality-of-service degradation for a pair of task streams; this data is collected through simulation with parameters found in §4. For the workloads in this experiment, the two homogeneous configurations violate quality-of-service agreements at least 6% of the time. 1 As some high-performance, power-hungry nodes are replaced by a larger number of lowpower processors, datacenter heterogeneity improves qualityof-service and reduces the frequency of violations to < 1%....
Combinatorial auctions (CAs) are used to allocate multiple items among bidders with complex valuations. Since the value space grows exponentially in the number of items, it is impossible for bidders to report their full value function even in medium-sized settings. Prior work has shown that current designs often fail to elicit the most relevant values of the bidders, thus leading to inefficiencies. We address this problem by introducing a machine learning-based elicitation algorithm to identify which values to query from the bidders. Based on this elicitation paradigm we design a new CA mechanism we call PVM, where payments are determined so that bidders’ incentives are aligned with allocative efficiency. We validate PVM experimentally in several spectrum auction domains, and we show that it achieves high allocative efficiency even when only few values are elicited from the bidders.
In mechanism design it is typical to impose incentive compatibility and then derive an optimal mechanism subject to this constraint. By replacing the incentive compatibility requirement with the goal of minimizing expected ex post regret, we are able to adapt statistical machine learning techniques to the design of payment rules. This computational approach to mechanism design is applicable to domains with multidimensional types and situations where computational efficiency is a concern. Specifically, given an outcome rule and access to a type distribution, we train a support vector machine with a special discriminant function structure such that it implicitly establishes a payment rule with desirable incentive properties. We discuss applications to a multi-minded combinatorial auction with a greedy winner-determination algorithm and to an assignment problem with egalitarian outcome rule. Experimental results demonstrate both that the construction produces payment rules with low ex post regret, and that penalizing classification errors is effective in preventing failures of ex post individual rationality.
In this paper, we present a machine learning-powered iterative combinatorial auction (CA). The main goal of integrating machine learning (ML) into the auction is to improve preference elicitation, which is a major challenge in large CAs. In contrast to prior work, our auction design uses value queries instead of prices to drive the auction. The ML algorithm is used to help the auction decide which value queries to ask in every iteration. While using ML inside an auction introduces new challenges, we demonstrate how we obtain a design that is individually rational, has good incentives, and is computationally practical. We benchmark our new auction against the well-known combinatorial clock auction (CCA). Our results indicate that, especially in large domains, our ML-powered auction can achieve higher allocative e ciency than the CCA, even with only a small number of value queries.
Heterogeneous design presents an opportunity to improve energy efficiency but raises a challenge in resource management. Prior design methodologies aim for performance and efficiency, yet a deployed system may miss these targets due to run-time effects, which we denote as risk. We propose design strategies that explicitly aim to mitigate risk. We introduce new processor selection criteria, such as the coefficient of variation in performance, to produce heterogeneous configurations that balance performance risks and efficiency rewards. Out of the tens of strategies we consider, risk-aware approaches account for more than 70% of the strategies that produce systems with the best service quality. Applying these risk-mitigating strategies to heterogeneous datacenter design can produce a system that violates response time targets 50% less often.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.