Laboratoires Spiral (France)

Despite continuous improvements, cloud physical resources remain underused, hence severely impacting the efficiency of these infrastructures at large. To overcome this inefficiency, Infrastructure-as-a-Service (IaaS) providers usually compensate for oversized Virtual Machines (VMs) by offering more virtual resources than are physically available on a host. However, this technique-known as oversubscription-may hinder performances when a statically-defined oversubscription ratio results in resource contention of hosted VMs. Therefore, instead of setting a static and cluster-wide ratio, this article studies how a greedy increase of the oversubscription ratio per Physical Machine (PM) and resources type can preserve performance goals. Keeping performance unchanged allows our contribution to be more realistically adopted by production-scale IaaS infrastructures. This contribution, named SCROOGEVM, leverages the detection of PM stability to carefully increase the associated oversubscription ratios. Based on metrics shared by public cloud providers, we investigate the impact of resource oversubscription on performance degradation. Subsequently, we conduct a comparative analysis of SCROOGEVM with state-ofthe-art oversubscription computations. The results demonstrate that our approach outperforms existing methods by leveraging the presence of long-lasting VMs, while avoiding live migration penalties and performance impacts for stakeholders.

The feedback control of High-Performance Computing (HPC) has been explored as an application area of Control Theory, because of the high variability involved in their resource management. A regulation mechanism can allow to soundly automate the injection of small flexible jobs in a cluster. A trade-off is needed, to fill up the cluster's computing capacity while avoiding overload of e.g., the file server. In this work, we describe new results in this context, where the overload avoidance controller is made adaptive to the jobs' size, that is a time-varying unknown parameter. To do so, the original PI controller is enhanced with an online estimation algorithm that allows the controller to adapt to various working conditions, to avoid performance degradation. Parallel and robust estimation algorithms are designed, tackling the challenges of bursting and noise in the system. Validation and evaluation of the adaptive controller are performed on a large-scale experimental HPC platform, showing higher robustness than the state-of-the-art in highly varying conditions. Reproducible analysis are available at doi:10.5281/zenodo.11961696.

Growing concern over digital privacy has led to the widespread use of tracking restriction tools, such as ad blockers, Virtual Private Networks (VPN), and privacy-focused web browsers. All major browser vendors have also deprecated, or plan to deprecate, third-party cookies to reduce tracking. Despite these efforts, advertising companies continuously innovate to overcome these restrictions. Recently, advertising platforms, like Meta, have been promoting server-side tracking solutions to bypass traditional browser-based tracking restrictions. This paper explores how server-side tracking technologies can link website visitors with their user accounts on Meta products. The goal is to assess the effectiveness and accuracy of employing this technology, as well as the effect of tracking restrictions on online tracking. Our methodology involves a series of experiments where we integrate Meta's client-side tracker (the Meta Pixel) and server-side technology (the Conversions API) on different web pages. We then drive traffic to these pages and evaluate the success rate of linking website visitors to their profiles on Meta products. Our findings show that Meta's server-side technology can match between 34% and 51% of website visitors to user profiles on Meta products using basic information like the visitor's IP address, user agent, and location data. This is comparable to Pixel-based user matching in optimal conditions (i.e., in the absence of tracking restrictions), which links between 42% and 61% of user profiles. Nevertheless, we see a considerable difference in accuracy: while the Pixel-based tracking achieves 100% accuracy, less than 65% of the profiles matched by server-side tracking are accurate.

We study the problem of enumerating the satisfying assignments for circuit classes from knowledge compilation, where assignments are ranked in a specific order. In particular, we show how this problem can be used to efficiently perform ranked enumeration of the answers to MSO queries over trees, with the order being given by a ranking function satisfying a subset-monotonicity property. Assuming that the number of variables is constant, we show that we can enumerate the satisfying assignments in ranked order for so-called multivalued circuits that are smooth, decomposable, and in negation normal form (smooth multivalued DNNF). There is no preprocessing and the enumeration delay is linear in the size of the circuit times the number of values, plus a logarithmic term in the number of assignments produced so far. If we further assume that the circuit is deterministic (smooth multivalued d-DNNF), we can achieve linear-time preprocessing in the circuit, and the delay only features the logarithmic term.

With the advent of cloud computing, different cloud providers with heterogeneous cloud services (compute, storage, network, applications, etc.) and their related Application Programming Interfaces (APIs) have emerged. This heterogeneity complicates the implementation of an interoperable cloud system. Several standards have been proposed to address this challenge and provide a unified interface to cloud resources. The Open Cloud Computing Interface (OCCI) thereby focuses on the standardization of a common API for Infrastructure-as-a-Service (IaaS) providers, while the Topology and Orchestration Specification for Cloud Applications (TOSCA) focuses on the standardization of a template language to enable the proper definition of the topology of cloud applications and their orchestrations on top of a cloud system. TOSCA thereby does not define how the application topologies are created on the cloud. Therefore, we analyze the conceptual similarities between the two approaches and we study how we can integrate them to obtain a complete standard-based approach to manage both Cloud Infrastructure and Cloud application layers. We propose an automated extensive mapping between the concepts of the two standards, and we provide TOSCA Studio, a model-driven tool chain for TOSCA that conforms to OCCI. TOSCA Studio allows to graphically design cloud applications as well as to deploy and manage them at runtime using a fully model-driven cloud orchestrator based on the two standards. Our contribution is validated by successfully transforming and deploying three cloud applications: WordPress, Node Cellar and Multi-Tier.

We reconsider the problem of containment of monadic datalog (MDL) queries in unions of conjunctive queries (UCQs). Prior work has dealt with special cases of the problem but has left the precise complexity characterization open. In addition, the complexity of one important special case, that of containment under access patterns, was not known before. We start by revisiting the connection between MDL/UCQ containment and containment problems involving regular tree languages. We then present a general approach for getting tighter bounds on the complexity of query containment, based on analysis of the number of mappings of queries into tree-like instances. We give two applications of the machinery. We first give an important special case of the MDL/UCQ containment problem that is in EXPTIME, and we use this bound to show an EXPTIME bound on containment under access patterns. Second, we show that the same technique can be used to get a new tight upper bound for containment of tree automata in UCQs. We finally show that the new MDL/UCQ upper bounds are tight. We establish a 2EXPTIME lower bound on the MDL/UCQ containment problem, resolving an open problem from the early 1990s. This bound holds for the MDL/CQ containment problem as well. We also show that changes to the conditions given in our special cases can not be eliminated, and that in particular slight variations of the problem of containment under access patterns become 2EXPTIME-complete.

Verification and Evaluation of Computer and Communication Systems - 12th International Conference, VECoS 2018, Grenoble, France, September 26-28, 2018, Proceedings

update for oadoi on Nov 02 2018