arXiv subjects

Francesco De Pellegrini

Publications and source records attributed to Francesco De Pellegrini.

At least 19 recordsLinked to original sources

arXiv2026

Learning in Proportional Allocation Auctions Games

The Kelly or proportional allocation mechanism is a simple and efficient auction-based scheme that distributes an infinitely divisible resource proportionally to the agents bids. When agents are aware of the allocation rule, their interactions form a game extensively studied in the literature. This paper examines the less explored repeated Kelly game, focusing mainly on utilities that are logarithmic in the allocated resource fraction. We first derive this logarithmic form from fairness-throughput trade-offs in wireless network slicing, and then prove that the induced stage game admits a unique Nash equilibrium NE. For the repeated play, we prove convergence to this NE under three behavioral models: (i) all agents use Online Gradient Descent (OGD), (ii) all agents use Dual Averaging with a quadratic regularizer (DAQ) (a variant of the Follow-the-Regularized leader algorithm), and (iii) all agents play myopic best responses (BR). Our convergence results hold even when agents use personalized learning rates in OGD and DAQ (e.g., tuned to optimize individual regret bounds), and they extend to a broader class of utilities that meet a certain sufficient condition. Finally, we complement our theoretical results with extensive simulations of the repeated Kelly game under several behavioral models, comparing them in terms of convergence speed to the NE, and per-agent time-average utility. The results suggest that BR achieves the fastest convergence and the highest time-average utility, and that convergence to the stage-game NE may fail under heterogeneous update rules.

cs.GT↗

arXiv2025

Multi-Agent Reinforcement Learning for Task Offloading in Wireless Edge Networks

In edge computing systems, autonomous agents must make fast local decisions while competing for shared resources. Existing MARL methods often resume to centralized critics or frequent communication, which fail under limited observability and communication constraints. We propose a decentralized framework in which each agent solves a constrained Markov decision process (CMDP), coordinating implicitly through a shared constraint vector. For the specific case of offloading, e.g., constraints prevent overloading shared server resources. Coordination constraints are updated infrequently and act as a lightweight coordination mechanism. They enable agents to align with global resource usage objectives but require little direct communication. Using safe reinforcement learning, agents learn policies that meet both local and global goals. We establish theoretical guarantees under mild assumptions and validate our approach experimentally, showing improved performance over centralized and independent baselines, especially in large-scale settings.

cs.LG↗

arXiv2025

Performing Load Balancing under Constraints

Join-the-shortest queue (JSQ) and its variants have often been used in solving load balancing problems. The aim of such policies is to minimize the average system occupation, e.g., the customer's system time. In this work we extend the traditional load balancing setting to include constraints that may be imposed, e.g., due to the communication network. We cast the problem into the framework of constrained MDPs, enabling the consideration of both action-dependent constraints, such as, e.g, bandwidth limitation, and state-dependent constraints, such as, e.g., minimum queue utilization. Unlike the state-of-the-art approaches, our load-balancing policies, in particular JSED-$k$ and JSSQ, are both provably safe and yet strive to minimize the system occupancy. Their performance is tested with extensive numerical results under various system settings.

cs.NI↗

arXiv2022

Fair Coflow Scheduling via Controlled Slowdown

The average coflow completion time (CCT) is the standard performance metric in coflow scheduling. However, standard CCT minimization may introduce unfairness between the data transfer phase of different computing jobs. Thus, while progress guarantees have been introduced in the literature to mitigate this fairness issue, the trade-off between fairness and efficiency of data transfer is hard to control. This paper introduces a fairness framework for coflow scheduling based on the concept of slowdown, i.e., the performance loss of a coflow compared to isolation. By controlling the slowdown it is possible to enforce a target coflow progress while minimizing the average CCT. In the proposed framework, the minimum slowdown for a batch of coflows can be determined in polynomial time. By showing the equivalence with Gaussian elimination, slowdown constraints are introduced into primal-dual iterations of the CoFair algorithm. The algorithm extends the class of the sigma-order schedulers to solve the fair coflow scheduling problem in polynomial time. It provides a 4-approximation of the average CCT w.r.t. an optimal scheduler. Extensive numerical results demonstrate that this approach can trade off average CCT for slowdown more efficiently than existing state of the art schedulers.

cs.DC↗

arXiv2022

DCoflow: Deadline-Aware Scheduling Algorithm for Coflows in Datacenter Networks

Datacenter networks routinely support the data transfers of distributed computing frameworks in the form of coflows, i.e., sets of concurrent flows related to a common task. The vast majority of the literature has focused on the problem of scheduling coflows for completion time minimization, i.e., to maximize the average rate at which coflows are dispatched in the network fabric. Modern applications, though, may generate coflows dedicated to online services and mission-critical computing tasks which have to comply with specific completion deadlines. In this paper, we introduce $\mathtt{DCoflow}$, a lightweight deadline-aware scheduler for time-critical coflows in datacenter networks. The algorithm combines an online joint admission control and scheduling logic and returns a $σ$-order schedule which maximizes the number of coflows that attain their deadlines. Extensive numerical results demonstrate that the proposed solution outperforms existing ones.

cs.DC↗

arXiv2022

Joint Traffic Offloading and Aging Control in 5G IoT Networks

The widespread adoption of 5G cellular technology will evolve as one of the major drivers for the growth of IoT-based applications. In this paper, we consider a Service Provider (SP) that launches a smart city service based on IoT data readings: in order to serve IoT data collected across different locations, the SP dynamically negotiates and rescales bandwidth and service functions. 5G network slicing functions are key to lease an appropriate amount of resources over heterogeneous access technologies and different site types. Also, different infrastructure providers will charge slicing service depending on specific access technology supported across sites and IoT data collection patterns. We introduce a pricing mechanism based on Age of Information (AoI) to reduce the cost of SPs. It provides incentives for devices to smooth traffic by shifting part of the traffic load from highly congested and more expensive locations to lesser charged ones, while meeting QoS requirements of the IoT service. The proposed optimal pricing scheme comprises a two-stage decision process, where the SP determines the pricing of each location and devices schedule uploads of collected data based on the optimal uploading policy. Simulations show that the SP attains consistent cost reductions tuning the trade-off between slicing costs and the AoI of uploaded IoT data.

cs.NI↗

arXiv2020

Delay and Price Differentiation in Cloud Computing: A Service Model, Supporting Architectures, and Performance

Many cloud service providers (CSPs) provide on-demand service at a price with a small delay. We propose a QoS-differentiated model where multiple SLAs deliver both on-demand service for latency-critical users and delayed services for delay-tolerant users at lower prices. Two architectures are considered to fulfill SLAs. The first is based on priority queues. The second simply separates servers into multiple modules, each for one SLA. As an ecosystem, we show that the proposed framework is dominant-strategy incentive compatible. Although the first architecture appears more prevalent in the literature, we prove the superiority of the second architecture, under which we further leverage queueing theory to determine the optimal SLA delays and prices. Finally, the viability of the proposed framework is validated through numerical comparison with the on-demand service and it exhibits a revenue improvement in excess of 200%. Our results can help CSPs design optimal delay-differentiated services and choose appropriate serving architectures.

cs.PF↗

arXiv2019

A Framework for Allocating Server Time to Spot and On-demand Services in Cloud Computing

Cloud computing delivers value to users by facilitating their access to computing capacity in periods when their need arises. An approach is to provide both on-demand and spot services on shared servers. The former allows users to access servers on demand at a fixed price and users occupy different periods of servers. The latter allows users to bid for the remaining unoccupied periods via dynamic pricing; however, without appropriate design, such periods may be arbitrarily small since on-demand users arrive randomly. This is also the current service model adopted by Amazon Elastic Cloud Compute. In this paper, we provide the first integral framework for sharing the time of servers between on-demand and spot services while optimally pricing spot instances. It guarantees that on-demand users can get served quickly while spot users can stably utilize servers for a properly long period once accepted, which is a key feature to make both on-demand and spot services accessible. Simulation results show that, by complementing the on-demand market with a spot market, a cloud provider can improve revenue by up to 464.7%. The framework is designed under assumptions which are met in real environments. It is a new tool that cloud operators can use to quantify the advantage of a hybrid spot and on-demand service, eventually making the case for operating such service model in their own infrastructures.

cs.PF↗

arXiv2018

Cutting Throughput on the Edge:App-Aware Placement in Fog Computing

Fog computing extends cloud computing technology to the edge of the infrastructure to let IoT applications access objects' data with reduced latency, location awareness and dynamic computation. By displacing workloads from the central cloud to the edge devices, fog computing overcomes communication bottlenecks avoiding raw data transfer to the central cloud, thus paving the way for the next generation IoT-based applications. In this paper we study scheduling and placement of applications in fog computing, which is key to ensure profitability for the involved stakeholders. We consider a scenario where the emerging microservice architecture allows for the design of applications as cascades of coupled microservice modules. It results into a mixed integer non linear problem involving constraints on both application data flows and computation placement. Due to the complexity of the original problem, we resort to a simplified version, which is further solved using a greedy algorithm. This algorithm is the core placement logic of the FogAtlas platform, a fog computing platform based on existing virtualization technologies. Extensive numerical results validate the model and the scalability of the proposed solution, showing it attains performance close to the optimal solution and, in our real implementation, it scales well with respect to the number of served applications.

cs.NI↗

arXiv2018

Cloud4IoT: a heterogeneous, distributed and autonomic cloud platform for the IoT

We introduce Cloud4IoT, a platform offering automatic deployment, orchestration and dynamic configuration of IoT support software components and data-intensive applications for data processing and analytics, thus enabling plug-and-play integration of new sensor objects and dynamic workload scalability. Cloud4IoT enables the concept of Infrastructure as Code in the IoT context: it empowers IoT operations with the flexibility and elasticity of Cloud services. Furthermore it shifts traditionally centralized Cloud architectures towards a more distributed and decentralized computation paradigm, as required by IoT technologies, bridging the gap between Cloud Computing and IoT ecosystems. Thus, Cloud4IoT is playing a role similar to the one covered by solutions like Fog Computing, Cloudlets or Mobile Edge Cloud. The hierarchical architecture of Cloud4IoThosts a central Cloud platform and multiple remote edge Cloud modules supporting dedicated devices, namely the IoT Gateways, through which new sensor objects are made accessible to the platform. Overall, the platform is designed in order to support systems where IoT-based and data intensive applications may pose specific requirements for low latency, restricted available bandwidth, or data locality. Cloud4IoT is built on several Open Source technologies for containerisation and implementations of standards, protocols and services for the IoT. We present the implementation of the platform and demonstrate it in two different use cases.

cs.DC↗

arXiv2018

Foggy: A Platform for Workload Orchestration in a Fog Computing Environment

In this paper we present Foggy, an architectural framework and software platform based on Open Source technologies. Foggy orchestrates application workload, negotiates resources and supports IoT operations for multi-tier, distributed, heterogeneous and decentralized Cloud Computing systems. Foggy is tailored for emerging domains such as 5G Networks and IoT, which demand resources and services to be distributed and located close to data sources and users following the Fog Computing paradigm. Foggy provides a platform for infrastructure owners and tenants (i.e., application providers) offering functionality of negotiation, scheduling and workload placement taking into account traditional requirements (e.g. based on RAM, CPU, disk) and non-traditional ones (e.g. based on networking) as well as diversified constraints on location and access rights. Economics and pricing of resources can also be considered by the Foggy model in a near future. The ability of Foggy to find a trade-off between infrastructure owners' and tenants' needs, in terms of efficient and optimized use of the infrastructure while satisfying the application requirements, is demonstrated through three use cases in the video surveillance and vehicle tracking contexts.

cs.DC↗

arXiv2018

Performance Analysis of QoS-Differentiated Pricing in Cloud Computing: An Analytical Approach

A fundamental goal in the design of IaaS service is to enable both user-friendly and cost-effective service access, while attaining high resource efficiency for revenue maximization. QoS differentiation is an important lens to achieve this design goal. In this paper, we propose the first analytical QoS-differentiated resource management and pricing architecture in the cloud computing context; here, a cloud service provider (CSP) offers a portfolio of SLAs. In order to maximize the CSP's revenue, we address two technical questions: (1) how to set the SLA prices so as to direct users to the SLAs best fitting their needs, and, (2) determining how many servers should be assigned to each SLA, and which users and how many of their jobs are admitted to be served. We propose optimal schemes to jointly determine SLA-based prices and perform capacity planning in polynomial time. Our pricing model retains high usability at the customer's end. Compared with standard usage-based pricing schemes, numerical results show that the proposed scheme can improve the revenue by up to a five-fold increase.

cs.DC↗

arXiv2017

Optimal curing policy for epidemic spreading over a community network with heterogeneous population

The design of an efficient curing policy, able to stem an epidemic process at an affordable cost, has to account for the structure of the population contact network supporting the contagious process. Thus, we tackle the problem of allocating recovery resources among the population, at the lowest cost possible to prevent the epidemic from persisting indefinitely in the network. Specifically, we analyze a susceptible-infected-susceptible epidemic process spreading over a weighted graph, by means of a first-order mean-field approximation. First, we describe the influence of the contact network on the dynamics of the epidemics among a heterogeneous population, that is possibly divided into communities. For the case of a community network, our investigation relies on the graph-theoretical notion of equitable partition; we show that the epidemic threshold, a key measure of the network robustness against epidemic spreading, can be determined using a lower-dimensional dynamical system. Exploiting the computation of the epidemic threshold, we determine a cost-optimal curing policy by solving a convex minimization problem, which possesses a reduced dimension in the case of a community network. Lastly, we consider a two-level optimal curing problem, for which an algorithm is designed with a polynomial time complexity in the network size.

math.OC↗

arXiv2017

Optimal Control of Storage Regeneration with Repair Codes

High availability of containerized applications requires to perform robust storage of applications' state. Since basic replication techniques are extremely costly at scale, storage space requirements can be reduced by means of erasure or repairing codes. In this paper we address storage regeneration using repair codes, a robust distributed storage technique with no need to fully restore the whole state in case of failure. In fact, only the lost servers' content is replaced. To do so, new cleanslate storage units are made operational at a cost for activating new storage servers and a cost for the transfer of repair data. Our goal is to guarantee maximal availability of containers' state files by a given deadline. activation of servers and communication cost. Upon a fault occurring at a subset of the storage servers, we aim at ensuring that they are repaired by a given deadline. We introduce a controlled fluid model and derive the optimal activation policy to replace servers under such correlated faults. The solution concept is the optimal control of regeneration via the Pontryagin minimum principle. We characterise feasibility conditions and we prove that the optimal policy is of threshold type. Numerical results describe how to apply the model for system dimensioning and show the tradeoff between

cs.IT↗

arXiv2016

Competitive Caching of Contents in 5G Edge Cloud Networks

The surge of mobile data traffic forces network operators to cope with capacity shortage. The deployment of small cells in 5G networks is meant to reduce latency, backhaul traffic and increase radio access capacity. In this context, mobile edge computing technology will be used to manage dedicated cache space in the radio access network. Thus, mobile network operators will be able to provision OTT content providers with new caching services to enhance the quality of experience of their customers on the move. In turn, the cache memory in the mobile edge network will become a shared resource. Hence, we study a competitive caching scheme where contents are stored at given price set by the mobile network operator. We first formulate a resource allocation problem for a tagged content provider seeking to minimize the expected missed cache rate. The optimal caching policy is derived accounting for popularity and availability of contents, the spatial distribution of small cells, and the caching strategies of competing content providers. It is showed to induce a specific order on contents to be cached based on their popularity and availability. Next, we study a game among content providers in the form of a generalized Kelly mechanism with bounded strategy sets and heterogeneous players. Existence and uniqueness of the Nash equilibrium are proved. Finally, extensive numerical results validate and characterize the performance of the model.

cs.GT↗

arXiv2016

Backward-Shifted Coding (BSC) based on Scalable Video Coding for HAS

The main task of HTTP Adaptive Streaming is to adapt video quality dynamically under variable network conditions. This is a key feature for multimedia delivery especially when quality of service cannot be granted network-wide and, e.g., throughput may suffer short term fluctuations. Hence, robust bitrate adaptation schemes become crucial in order to improve video quality. The objective, in this context, is to control the filling level of the playback buffer and maximize the quality of the video, while avoiding unnecessary video quality variations. In this paper we study bitrate adaptation algorithms based on Backward-Shifted Coding (BSC), a scalable video coding scheme able to greatly improve video quality. We design bitrate adaptation algorithms that balance video rate smoothness and high network capacity utilization, leveraging both on throughput-based and buffer-based adaptation mechanisms. Extensive simulations using synthetic and real-world video traffic traces show that the proposed scheme performs remarkably well even under challenging network conditions.

cs.MM↗

arXiv2015

Epidemic Outbreaks in Networks with Equitable or Almost-Equitable Partitions

We study the diffusion of epidemics on networks that are partitioned into local communities. The gross structure of hierarchical networks of this kind can be described by a quotient graph. The rationale of this approach is that individuals infect those belonging to the same community with higher probability than individuals in other communities. In community models the nodal infection probability is thus expected to depend mainly on the interaction of a few, large interconnected clusters. In this work, we describe the epidemic process as a continuous-time individual-based susceptible-infected-susceptible (SIS) model using a first-order mean-field approximation. A key feature of our model is that the spectral radius of this smaller quotient graph (which only captures the macroscopic structure of the community network) is all we need to know in order to decide whether the overall healthy-state defines a globally asymptotically stable or an unstable equilibrium. Indeed, the spectral radius is related to the epidemic threshold of the system. Moreover we prove that, above the threshold, another steady-state exists that can be computed using a lower-dimensional dynamical system associated with the evolution of the process on the quotient graph. Our investigations are based on the graph-theoretical notion of equitable partition and of its recent and rather flexible generalization, that of almost equitable partition.

cs.SI↗

arXiv2013

Differential Games of Competition in Online Content Diffusion

Access to online contents represents a large share of the Internet traffic. Most such contents are multimedia items which are user-generated, i.e., posted online by the contents' owners. In this paper we focus on how those who provide contents can leverage online platforms in order to profit from their large base of potential viewers. Actually, platforms like Vimeo or YouTube provide tools to accelerate the dissemination of contents, i.e., recommendation lists and other re-ranking mechanisms. Hence, the popularity of a content can be increased by paying a cost for advertisement: doing so, it will appear with some priority in the recommendation lists and will be accessed more frequently by the platform users. Ultimately, such acceleration mechanism engenders a competition among online contents to gain popularity. In this context, our focus is on the structure of the acceleration strategies which a content provider should use in order to optimally promote a content given a certain daily budget. Such a best response indeed depends on the strategies adopted by competing content providers. Also, it is a function of the potential popularity of a content and the fee paid for the platform advertisement service. We formulate the problem as a differential game and we solve it for the infinite horizon case by deriving the structure of certain Nash equilibria of the game.

cs.SI↗