The adoption of computing resources oversubscription in cloud environments is conventionally limited to a restricted subset of Virtual Machines (VMs) within the providers’ offerings, primarily driven by performance considerations. So far, VMs schedulers mostly implement all-or-nothing oversubscription strategies, wherein all VM resources are either oversubscribed or remain unaltered. While the former strategy offers higher consolidation rates, the latter delivers better performance guarantees.In this paper, we conducted an empirical study of the individual usage of virtual CPUs (vCPUs) in the OVHCloud production environment and we demonstrate that, as they are not uniformly utilized, the current holistic approach may not be appropriate. Based on these observations, we introduce a novel approach, named SweetspotVM, where oversubscription ratios are applied at the granularity of individual vCPU, instead of the whole VMs. This novel paradigm unlocks a more flexible oversubscription management strategy, pinning oversubscription ratios per vCPU within VMs. We present a prototype of SweetspotVM to illustrate the feasibility of accommodating multiple oversubscription levels within a single host and assigning them to individual vCPU.We assess the viability of our approach on a physical platform, demonstrating the possibility of dividing the cost of hosting VMs by 3, while maintaining the VMs performance at the level of non-oversubscribed platforms. We, therefore, believe that SweetspotVM opens new avenues to boost the consolidation of VMs on a reduced number of servers, with positive impacts on the environmental footprint of cloud computing.

Cloud providers generally expose a large catalog of Virtual Machine (VM) offers, some being categorized as premium-guaranteeing dedicated resources-and others being hosted in oversubscribed environments, where virtual resources can exceed the physical capabilities of Physical Machines (PMs). The latter strategy is often employed to increase platform utilization, as hosted VMs are unlikely to fully utilize all their allocated resources simultaneously [1]. However, managing multiple oversubscribed VM levels introduces an additional layer of complexity for Cloud providers, often leading them to provision isolated clusters of PMs for each category of offers. In this paper, we introduce SLACKVM, a novel Cloud-shared architecture wherein VMs from various oversubscription levels coexist on the same cluster of PMs. In particular, we demonstrate that oversubscription levels can be complementary, meaning they do not saturate the same resource components. By leveraging this complementarity, Cloud providers can couple multiple levels to better consolidate VM offers onto PMs, and reduce the size of their clusters by up to 9.6%. These resource savings result in both an operational cost reduction and a reduced ecological footprint for Cloud infrastructures, with a limited impact on the Quality of Service (QoS).

Due to the rapid growth of Cloud services, data centers have become major energy consumers, resulting in significant CO2emissions. Several infrastructure-focused strategies, such as resource consolidation, have been used to reduce the carbon footprint of Cloud infrastructure. However, end-users are often left out of the picture. Since they are the primary target of Cloud applications, it would be beneficial to actively involve them in reducing the carbon footprint of Cloud applications. In this paper, we offer end-users a way to influence the carbon footprint of Cloud applications they use. To this end, we ask end-users to select a high-level mode to control the carbon footprint of a Cloud application. We then design a dynamic adaptation algorithm that determines an appropriate configuration for the application for each request, based on the end-user mode and on the carbon intensity of the infrastructure energy sources. We implement and evaluate our system on a simple image-resizing application. We run experiments on SeDuCe, a Cloud infrastructure testbed partially powered by solar panels. Our results show that we save energy consumption by up to $84 \%$ when all end-users agree to degrade the quality of the application’s output, and we provide a good quality-energy trade-off when end-users make heterogeneous choices. In addition, we are able to improve quality by leveraging the available green energy budget.

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 asoversubscription—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, namedScroogeVM, 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 ofScroogeVMwith state-of-the-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.

Cloud infrastructures are large-scale and complex platforms designed to host a wide diversity of applications and workloads. Given these complexity and scale factors, simulators and benchmarks are broadly adopted in vitro to study their behaviors, prototype new software components and heuristics, and evaluate their effective performances.However, both state-of-the-art simulations and benchmarks may suffer from a representativeness problem, as the reported results can vary depending on their input workloads. For example, a Infrastructure-as-a-Service (IaaS) platform aims to host Virtual Machines (VMs), whose characteristics (resource configurations, workload intensity, arrival/departure rate, etc.) can greatly differ depending on Cloud providers and public/private deployments. Addressing this IaaS representativeness thus requires Cloud providers to share production-scale datasets, which might be considered sensitive. Moreover, Simulations and benchmarks require a specific experiment scenario that cannot be easily generated from Cloud providers characteristics.To address these issues, this paper introduces CloudFactory, a IaaS workload generator. Our contribution is first composed of a library that can be used by Cloud providers to share IaaS statistics, instead of raw datasets. Then, we introduce a generator designed to produce realistic VM workloads that match these statistics. CloudFactory is made available as open-source software that can be adopted by Cloud providers and researchers to foster the evaluation of new contributions.As an example, we perform an analysis on scheduling evolution for different IaaS workload intensity of two different Cloud providers: Microsoft Azure and Chameleon. We also report on OVHcloud statistics computed from CloudFactory and compare them to other Cloud providers.

Fog Computing is a paradigm aiming to decentralize the Cloud by geographically distributing away computation, storage and network resources as well as related services. This notably reduces bottlenecks and data movement. However, managing Fog resources is a major challenge because the targeted systems are large, geographically distributed, unreliable and very dynamic. Cloud systems are generally managed via centralized autonomic controllers automatically optimizing both application QoS and resource usage. To leverage the self-management of Fog resources, we propose to orchestrate a fleet of autonomic controllers in a decentralized manner, each with a local view of its own resources. In this paper, we present our SeMaFoR (Self-Management of Fog Resources) vision that aims at collaboratively operating Fog resources. SeMaFoR is a generic approach made of three cornerstones: an Architecture Description Language for the Fog, a collaborative and consensual decision-making process, and an automatic coordination mechanism for reconfiguration.

Smart contracts, scripts at the heart of blockchain-based applications, are meant to be available forever once deployed. However, this property has a price. The amount of space required to store new contracts keeps increasing. This increase impacts each participating node's performance and makes it inconvenient for low-end devices to participate in the network. Among all contracts deployed in the blockchain, a vast majority will lead to little if any usage. We demonstrate that, in the course of one year, 70 % of deployed contracts lead to no use. Unfortunately, unused contracts keep occupying space on the blockchain. To tackle this issue, we propose a new protocol to identify and delete unused contracts. Through simulation, based on Ethereum historical data, we show that deletion of smart contracts after an inactivity period of 90 days could lead to a 66 % reduction in the number of contracts stored over a year.

Over the last years, research activities on blockchain technologies have fairly increased. Firstly introduced with Bitcoin, some projects have since emerged to create or improve blockchain features like privacy while others propose to overcome technical limitations such as scalability and energy consumption. New proposals are often evaluated with ad hoc tools and experimental environments. Reproducibility and comparison of these new contributions with the state of the art of the blockchain technologies are therefore complicated. To the best of our knowledge, only a few tools partially address the design of a generic benchmarking of blockchain technologies (e.g., load generation). In this paper, we introduce BCTMark, a generic framework for benchmarking blockchain technologies on an emulated network in a reproducible way. To illustrate the portability of experiments using BCTMark, we have conducted some experiments on two different testbeds: a cluster of Dell PowerEdge R630 servers (Grid'5000) and one of Raspberry Pi 3+. Experiments have been conducted on three different blockchain systems (Ethereum Clique/Ethash and Hyperledger Fabric) to measure their CPU consumption and energy footprint for different numbers of clients.

Best Paper Award

While the proliferation of cloud services have greatly impacted our society, how green are these services is yet to be answered. Although, demand escalation for green services has grown due to societal awareness, the approaches to provide green services and establish Green SLAs remain oblivious for cloud or infrastructure providers. The main challenge for cloud provider is to manage Green SLAs with their customers while satisfying their business objectives, such as maximizing profits by lowering expenditure for green energy. Since, Green SLA needs to be proposed based on the presence of green energy, the intermittent nature of renewable sources makes it difficult to be achieved. In response, this paper presents a scheme for green energy management in the presence of explicit and implicit integration of renewable energy in data center. More specifically we propose three contributions: i) we introduce the concept of virtualization of green energy to address the uncertainty of green energy availability, ii) we extend the Cloud Service Level Agreement (CSLA) language to support Green SLA by introducing two new threshold parameters and iii) we introduce greenSLA algorithm which leverages the concept of virtualization of green energy to provide per interval specific Green SLA. Experiments were conducted with real workload profile from PlanetLab and server power model from SPECpower to demonstrate that, Green SLA can be successfully established and satisfied without incurring higher cost.

With the ever growing demand and popularity of cloud based services, data centers have to urgently face energy consumption issue. Similar to other large consumers of power, data centers find themselves increasingly pressured to reduce their carbon footprint. In response, cloud providers have started to set sustainability goals to reduce carbon emissions by using renewable sources to their services. Traditionally, batch processing cloud applications are deadline oriented, hence they can be easily adapted with the different green energy profile. Whereas, interactive cloud applications are imposed with several performance criteria. This paper, the first of its kind, investigates a thorough analysis of energy consumption and performance trade-off by allowing smart usage of green energy for interactive cloud application. Moreover, we propose an auto-scaler, named SaaScaler, that implements several control loop based application controllers to satisfy different performance (i.e., response time, availability, and user experience) and resource aware metrics (i.e., quality of energy). Based on extensive experiments with RUBiS benchmark and real workload traces using single compute node in Openstack/Grid'5000, results suggest that 13 percent brown energy consumption can be reduced without deprovisioning any physical or virtual resources at IaaS layer while 29 percent more users can access the application by dynamically adjusting capacity requirements. Furthermore, our investigation verifies that the energy consumption deviates as little as .07 percent when our approach is scaled using several physical nodes.

With the proliferation of Cloud computing, data centers have to urgently face energy consumption issues. Although recent efforts such as the integration of renewable energy to data centers or energy efficient techniques in (virtual) machines contribute to the reduction of carbon footprint, creating green energy awareness around Interactive Cloud Applications by smartly using the presence of green energy has not been yet addressed. By awareness, we mean the inherited capability of Software-as-a-Service applications to dynamically adapt with the availability of green energy and to reduce energy consumption while green energy is scarce or absent. In this paper, we present two application controllers based on different metrics (e.g., availability of green energy, response time, user experience level). Based on %and we have done extensive experiments with a real application benchmark and workloads in Grid'5000, results suggest that providers revenue can be increased as high as 64%, while 13% brown energy can be reduced without deprovisioning any physical or virtual resources at IaaS layer and 17 fold increment of performance can be guaranteed.

Cloud Computing promises to completely revolutionize the capacity management of resources. The elasticity and the economy of scale are the intrinsic elements that dierentiate it from traditional computing paradigm. A good capacity planning method is a necessary factor but not sucient to fully exploit Cloud elasticity. This paper proposes innovative policies for resource management to achieve the optimal balance between capacity and quality of Cloud services while supporting Cloud technical and conceptual limitations. The main idea is to control finely the scalability and the termination of virtual machines in regards of several criteria such as the lifecycle of the instances (e.g. Initialization time) or their cost. The approach was evaluated with a real infrastructure (Amazon EC2) and an application test bed. Experimental results illustrate the soundness of the proposed approach and the impact of scalability/termination resource policies. Using Delta Scaling, the cost saving of as much as 30% can be achieved while causing the minimum number of violations, as small as 1%.

Elasticity is the intrinsic element that differentiates Cloud computing from traditional computing paradigm, since it allows service providers to rapidly adjust their needs for resources to absorb the demand and hence guarantee a minimum level of Quality of Service (QoS) that respects the Service Level Agreements (SLAs) previously defined with their clients. However, due to non-negligible resource initiation time, network fluctuations or unpredictable workload, it becomes hard to guarantee QoS levels and SLA violations may occur. This paper proposes a language support for Cloud elasticity management that relies on CSLA (Cloud Service Level Agreement). CSLA offers new features such as QoS/functionality degradation and an advanced penalty model that allow providers to finely express contracts so that services self-adaptation capabilities are improved and SLA violations minimized. The approach was evaluated with a real infrastructure and application test bed. Experimental results show that the use of CSLA makes Cloud services capable of absorbing more peaks and oscillations by trading-off the QoS levels and costs due to penalties.

Cloud Computing provides a convenient means of remote on-demand and pay-per-use access to computing resources. However, its ad hoc management of quality-of-service and SLA poses significant challenges to the performance, dependability and costs of online cloud services. The paper precisely addresses this issue and makes a threefold contribution. First, it introduces a new cloud model, the SLAaaS (SLA aware Service) model. SLAaaS enables a systematic integration of QoS levels and SLA into the cloud. It is orthogonal to other cloud models such as SaaS or PaaS, and may apply to any of them. Second, the paper introduces CSLA, a novel language to describe QoS-oriented SLA associated with cloud services. Third, the paper presents a control theoretic approach to provide performance, dependability and cost guarantees for online cloud services, with time-varying workloads. The proposed approach is validated through case studies and extensive experiments with online services hosted in clouds such as Amazon EC2. The case studies illustrate SLA guarantees for various services such as a MapReduce service, a cluster-based multi-tier e-commerce service, and a low-level locking service.

Cloud computing paradigm has become the solution to provide good service quality and exploit economies of scale. However, the management of such elastic resources, with different Quality-of-Service (QoS) combined with on-demand self-service, is a complex issue. This paper proposes an approach driven by Service Level Agreement (SLA) for optimizing the capacity planning for Cloud applications. The main challenge for a service provider is to determine the best trade-off between profit and customer satisfaction. In order to address this issue, we follow a queueing network proposal and present an analytical performance model to predict Cloud service performance. Based on a utility function and a capacity planning method, our solution calculates the optimal configuration of a Cloud application. We rely on autonomic computing to adjust continuously the configuration. Simulation experiments indicate that our model i) faithfully captures the performance of Cloud applications for a number of workloads and configurations and ii) successfully keeps the best trade-off.

In a system where classes are treated as first class objects, classes are defined as instances of other classes called metaclasses. An important benefit of using metaclasses is the ability to assign properties to classes (e.g. being abstract, being final, tracing particular messages, supporting multiple inheritance), independently from the base-level code. However, when both inheritance and instantiation are explicitly and simultaneously involved, communication between classes and their instances raises the metaclass compatibility issue. Some languages (such as SMALLTALK) address this issue but do not easily allow the assignment of specific properties to classes. In contrast, other languages (such as CLOS) allow the assignment of specific properties to classes but do not tackle the compatibility issue well.In this paper, we describe a new model of metalevel organization, called the compatibility model, which overcomes this difficulty. It allows safe metaclass programming since it makes it possible to assign specific properties to classes while ensuring metaclass compatibility. Therefore, we can take advantage of the expressive power of metaclasses to build reliable software. We extend this compatibility model in order to enable safe reuse and composition of class specific properties. This extension is implemented in NEOCLASSTALK, a fully reflective SMALLTALK.