On the Cluster Admission Problem for Cloud Computing

Cloud computing providers face the problem of matching heterogeneous customer workloads to resources that will serve them. This is particularly challenging if customers, who are already running a job on a cluster, scale their resource usage up and down over time. The provider therefore has to continuously decide whether she can add additional workloads to a given cluster or if doing so would impact existing workloads’ ability to scale. Currently, this is often done using simple threshold policies to reserve large parts of each cluster, which leads to low efficiency (i.e., low average utilization of the cluster). We propose more sophisticated policies for controlling admission to a cluster and demonstrate that they significantly increase cluster utilization. We first introduce the cluster admission problem and formalize it as a constrained Partially Observable Markov Decision Process (POMDP). As it is infeasible to solve the POMDP optimally, we then systematically design admission policies that estimate moments of each workload’s distribution of future resource usage. Via extensive simulations grounded in a trace from Microsoft Azure, we show that our admission policies lead to a substantial improvement over the simple threshold policy. We then show that substantial further gains are possible if high-quality information is available about arriving workloads. Based on this, we propose an information elicitation approach to incentivize users to provide this information and simulate its effects.

Coordination and Simplexity in Modelling Air Traffic Control as a Service: The Case of NAV Portugal

Copyright © 2017 John Wiley & Sons, Ltd. This study analyses a unique, revelatory case of service modelling in a complex organisation providing air traffic control. The study analyses broad modelling activities, including information elicitation, analysis, and organisation, undertaken by a small team internal to the organisation that so far has spent about 2,400 person-hours of effort in the project. The study follows a qualitative approach in the interpretivist tradition based on interviews and document analysis. The study analyses the project framework, modelling notations, data collection, collaboration, modelling activities, and project outputs. The findings are interpreted in light of two theoretical lenses: coordination and simplexity. The study suggests that simplexity is beneficial for structuring the modelling of complex, knowledge-based services. A pattern was identified combining an initial step promoting simplicity and establishing communication with the stakeholders, followed by a second step acquiring complexity of understanding. Considering the mechanisms defined by coordination theory (flow, share, and fit), the study suggests a predominance of the fit mechanism in modelling knowledge-based services. The paper contributes to a better understanding of the challenges of modelling work through cognitive and knowledge-based lenses and identifies possible strategies to overcome these challenges. The paper also contributes to the emergent literature on simplexity by applying that particular lens to work modelling.

Coordination and Simplexity in Modelling Air Traffic Control as a Service: The Case of NAV Portugal

Copyright © 2017 John Wiley & Sons, Ltd. This study analyses a unique, revelatory case of service modelling in a complex organisation providing air traffic control. The study analyses broad modelling activities, including information elicitation, analysis, and organisation, undertaken by a small team internal to the organisation that so far has spent about 2,400 person-hours of effort in the project. The study follows a qualitative approach in the interpretivist tradition based on interviews and document analysis. The study analyses the project framework, modelling notations, data collection, collaboration, modelling activities, and project outputs. The findings are interpreted in light of two theoretical lenses: coordination and simplexity. The study suggests that simplexity is beneficial for structuring the modelling of complex, knowledge-based services. A pattern was identified combining an initial step promoting simplicity and establishing communication with the stakeholders, followed by a second step acquiring complexity of understanding. Considering the mechanisms defined by coordination theory (flow, share, and fit), the study suggests a predominance of the fit mechanism in modelling knowledge-based services. The paper contributes to a better understanding of the challenges of modelling work through cognitive and knowledge-based lenses and identifies possible strategies to overcome these challenges. The paper also contributes to the emergent literature on simplexity by applying that particular lens to work modelling.

Information Elicitation Mechanisms for Statistical Estimation

We study learning statistical properties from strategic agents with private information. In this problem, agents must be incentivized to truthfully reveal their information even when it cannot be directly verified. Moreover, the information reported by the agents must be aggregated into a statistical estimate. We study two fundamental statistical properties: estimating the mean of an unknown Gaussian, and linear regression with Gaussian error. The information of each agent is one point in a Euclidean space.Our main results are two mechanisms for each of these problems which optimally aggregate the information of agents in the truth-telling equilibrium:• A minimal (non-revelation) mechanism for large populations — agents only need to report one value, but that value need not be their point.• A mechanism for small populations that is non-minimal — agents need to answer more than one question.These mechanisms are “informed truthful” mechanisms where reporting unaltered data (truth-telling) 1) forms a strict Bayesian Nash equilibrium and 2) has strictly higher welfare than any oblivious equilibrium where agents' strategies are independent of their private signals. We also show a minimal revelation mechanism (each agent only reports her signal) for a restricted setting and use an impossibility result to prove the necessity of this restriction.We build upon the peer prediction literature in the single-question setting; however, most previous work in this area focuses on discrete signals, whereas our setting is inherently continuous, and we further simplify the agents' reports.

Development of the Reporting Information about Networks and Groups (RING) task: a method for eliciting information from memory about associates, groups, and networks

Purpose Eliciting detailed and comprehensive information about the structure, organisation and relationships between individuals involved in organised crime gangs, terrorist cells and networks is a challenge in investigations and debriefings. Drawing on memory theory, the purpose of this paper is to develop and test the Reporting Information about Networks and Groups (RING) task, using an innovative piece of information elicitation software. Design/methodology/approach Using an experimental methodology analogous to an intelligence gathering context, participants (n=124) were asked to generate a visual representation of the “network” of individuals attending a recent family event using the RING task. Findings All participants successfully generated visual representations of the relationships between people attending a remembered social event. The groups or networks represented in the RING task output diagrams also reflected effective use of the software functionality with respect to “describing” the nature of the relationships between individuals. Practical implications The authors succeeded in establishing the usability of the RING task software for reporting detailed information about groups of individuals and the relationships between those individuals in a visual format. A number of important limitations and issues for future research to consider are examined. Originality/value The RING task is an innovative development to support the elicitation of targeted information about networks of people and the relationships between them. Given the importance of understanding human networks in order to disrupt criminal activity, the RING task may contribute to intelligence gathering and the investigation of organised crime gangs and terrorist cells and networks.