Papers by Abderrahmen Mtibaa
Whispering: Joint Service Offloading and Computation Reuse in Cloud-Edge Networks
Due to the proliferation of Internet of Things (IoT) and application/user demands that challenge ... more Due to the proliferation of Internet of Things (IoT) and application/user demands that challenge communication and computation, edge computing has emerged as the paradigm to bring computing resources closer to users. In this paper, we present Whispering, an analytical model for the migration of services (service offloading) from the cloud to the edge, in order to minimize the completion time of computational tasks offloaded by user devices and improve the utilization of resources. We also empirically investigate the impact of reusing the results of previously executed tasks for the execution of newly received tasks (computation reuse) and propose an adaptive task offloading scheme between edge and cloud. Our evaluation results show that Whispering achieves up to 35% and 97% (when coupled with computation reuse) lower task completion times than cases where tasks are executed exclusively at the edge or the cloud.
Coupled with the rapid increase in mobile device users and the bandwidth and latency demands are ... more Coupled with the rapid increase in mobile device users and the bandwidth and latency demands are the continuous increase of devices' processing capabilities, storage, and wireless connectivity options. The multiple radio access technology (multi-RAT) is proposed to satisfy mobile users' increasing needs. The Information-Centric Networking (ICN) paradigm is better tuned (than the current Internet Protocol approach) to support multi-RAT communications. ICN eschews the connection-based content retrieval model used today and has desirable features such as data naming, in-network caching, and device mobility-a paradigm ripe for exploration. We propose DICE, an ICN forwarding strategy that helps a device dynamically select a subset of its multi-RAT interfaces for communication. DICE assesses the state of edge links and network congestion to determine the minimum number of interfaces required to to perform data delivery. We perform simulations to compare DICE's performance with bestroute2 and multicast strategies (part of the named data networking simulator, ndnSIM). We show that DICE is the best of both worlds: providing a higher delivery ratio (0.2-2 times) and much lower overhead (by 2-8 times) for different packet rates.
arXiv (Cornell University), Feb 4, 2022
Multi-Agent Reinforcement Learning (MARL) is nowadays widely used to solve real-world and complex... more Multi-Agent Reinforcement Learning (MARL) is nowadays widely used to solve real-world and complex decisions in various domains. While MARL can be categorized into independent and cooperative approaches, we consider the independent approach as a simple, more scalable, and less costly method for large-scale distributed systems, such as network packet forwarding. In this paper, we quantitatively and qualitatively assess the benefits of leveraging such independent agents learning approach, in particular IQL-based algorithm, for packet forwarding in computer networking, using the Named Data Networking (NDN) architecture as a driving example. We put multiple IQL-based forwarding strategies (IDQF) to the test and compare their performances against very basic forwarding schemes and simple topologies/traffic models to highlight major challenges and issues. We discuss the main issues related to the poor performance of IDQF, and quantify the impact of these issues on isolation when training and testing the IDQF models under different model tuning parameters and network topologies/characteristics.
Towards Service Discovery and Invocation in Data-Centric Edge Networks
The efforts exploring Named Data Networking (NDN) have mainly focused on addressing the lack of s... more The efforts exploring Named Data Networking (NDN) have mainly focused on addressing the lack of scalable data distribution by today’s Internet. In this paper, we argue that NDN offers a richer environment for edge computing applications. We consider a scenario, where applications need to discover the services running in the edge network. We demonstrate the design and implementation of a distributed service discovery mechanism over NDN through an example use-case of a mobile application for vision impairment patient. The paper discusses three main edge computing challenges, namely service discovery, service invocation, and user mobility management, to highlight NDN’s architectural advantages for edge computing systems. Experimental results show that our framework design can effectively utilize the available resources at the network edge, being able to satisfy 95-98% of mobile users’ service requests.
The smart grid, with its large array of networked devices and bidirectional data flow between the... more The smart grid, with its large array of networked devices and bidirectional data flow between the end-users and the grid, presents new requirements in service reliability, communication latency, and data delivery. The traditional TCP/IP communication paradigm was not designed to handle these requirements at the envisioned scale. This calls for a novel networking paradigm. This paper makes the case for the use of the Information Centric Networking (ICN) paradigm to create the smart grid network architecture. We quantitatively assess the gains resulting from ICN's inherent functionalities, such as concurrent use of multiple interfaces, request aggregation, and stateful forwarding, which enable timely critical message delivery and fast packet re-transmissions. We perform simulations to compare IP and ICN-based smart grid deployments. Our results show that the ICN-based solution outperforms the IP-based solution, especially in a network with packet losses.
IEEE Internet of Things Journal, May 1, 2020
In today's era of explosion of Internet of Things (IoT) and end-user devices and their data volum... more In today's era of explosion of Internet of Things (IoT) and end-user devices and their data volume, emanating at the network's edge, the network should be more in-tune with meeting the needs of these demanding edge computing applications. To this end, we design and prototype Information-Centric edge (ICedge), a general-purpose networking framework that streamlines service invocation and improves reuse of redundant computation at the edge. ICedge runs on top of Named-Data Networking, a realization of the Information-Centric Networking vision, and handles the "low-level" network communication on behalf of applications. ICedge features a fully distributed design that: (i) enables users to get seamlessly on-boarded onto an edge network, (ii) delivers application invoked tasks to edge nodes for execution in a timely manner, and (iii) offers naming abstractions and network-based mechanisms to enable (partial or full) reuse of the results of already executed tasks among users, which we call "compute reuse", resulting in lower task completion times and efficient use of edge computing resources. Our simulation and testbed deployment results demonstrate that ICedge can achieve up to 50× lower task completion times leveraging its networkbased compute reuse mechanism compared to cases, where reuse is not available.
arXiv (Cornell University), Jan 16, 2022
Named-Data Networking (NDN), a prominent realization of the Information-Centric Networking (ICN) ... more Named-Data Networking (NDN), a prominent realization of the Information-Centric Networking (ICN) vision, offers a request-response communication model where data is identified based on application-defined names at the network layer. This amplifies the ability of censoring authorities to restrict user access to certain data/websites/applications and monitor user requests. The majority of existing NDN-based frameworks have focused on enabling users in a censoring network to access data available outside of this network, without considering how data producers in a censoring network can make their data available to users outside of this network. This problem becomes especially challenging, since the NDN communication paths are symmetric, while producers are mandated to sign the data they generate and identify their certificates. In this paper, we propose Harpocrates, an NDN-based framework for anonymous data publication under censorship conditions. Harpocrates enables producers in censoring networks to produce and make their data available to users outside of these networks while remaining anonymous to censoring authorities. Our evaluation demonstrates that Harpocrates achieves anonymous data publication under different settings, being able to identify and adapt to censoring actions.
This paper discusses leveraging the Named Data Networking (NDN) architecture and Named Function N... more This paper discusses leveraging the Named Data Networking (NDN) architecture and Named Function Networking (NFN) to facilitate in-network edge computing. In the NDN context, we consider a the Augmented Reality (AR) use-case-a challenging application-to discuss how NDN functionalities can be leveraged for addressing inherent edge computing challenges, such as efficient resource discovery, compute re-use, mobility management, and security. We present several options to tackle the highlighted challenges and where possible provide solutions.
A Case for Compute Reuse in Future Edge Systems: An Empirical Study
Edge computing has been proposed as a solution to the ultra-low latency requirements of novel com... more Edge computing has been proposed as a solution to the ultra-low latency requirements of novel compute-intensive applications and systems such as autonomous driving, smart cities, etc. However, the scale of such compute-intensive systems coupled with stricter application requirements causes the computation load to exceed the current rate at which we provision the resources at the edge. In this paper, we advocate that reusing (partly or fully) the results of already executed computational tasks among multiple users to calculate new results, which we call "compute reuse", has the potential to significantly reduce resource utilization and lessen the time needed for the completion (i.e., execution) of new tasks. We conduct an experimental study to quantify the performance gain and improved utilization of resources that stem from systems being able to reuse previous computations. Our results demonstrate that such systems can result in up to 50× lower completion times and substantially lower utilization of computing resources. We also discuss alternative design options and tradeoffs of building systems aware of compute reuse semantics.
IEEE Network, Nov 1, 2020
Named Data Networking (NDN) architectural features, including multicast data delivery, stateful f... more Named Data Networking (NDN) architectural features, including multicast data delivery, stateful forwarding, and in-network data caching, have shown promise for applications such as video streaming and file sharing. However, collaborative applications, requiring a multi-producer participation introduce new NDN design challenges. In this paper, we highlight these challenges in the context of the Network Time Protocol (NTP) and one of its most widely-used deployments for NTP server discovery, the NTP pool project. We discuss the design requirements for the support of NTP and NTP pool and present general directions for the design of a time synchronization protocol over NDN, coined Named Data Networking Time Protocol (NDNTP).
uDiscover: User-Driven Service Discovery in Pervasive Edge Computing using NDN
2022 IEEE International Conference on Edge Computing and Communications (EDGE)
Proceedings of the 27th ACM on Symposium on Access Control Models and Technologies
Named-Data Networking (NDN), a realization of the Information-Centric Networking (ICN) vision, of... more Named-Data Networking (NDN), a realization of the Information-Centric Networking (ICN) vision, offers a request-response communication model where data is identified based on application-defined names at the network layer. This amplifies the ability of censoring authorities to restrict access to certain data/websites/applications and monitor user requests. The majority of existing NDN-based frameworks have focused on enabling users in a censoring network to access data available outside of this network, without considering how data producers in a censoring network can make their data available to users outside of this network. This problem becomes especially challenging, since the NDN communication paths are symmetric, while producers are mandated to sign the data they generate and identify their certificates. In this paper, we propose Harpocrates, an NDN-based framework for anonymous data publication under censorship conditions. Harpocrates enables producers in censoring networks to produce and make their data available to users outside of these networks while remaining anonymous to censoring authorities. Our evaluation demonstrates that Harpocrates achieves anonymous data publication under different settings, being able to identify and adapt to censoring actions. • Networks → Network protocol design; • Security and privacy → Pseudonymity, anonymity and untraceability; Privacypreserving protocols.
Multi-Agent Reinforcement Learning (MARL) is nowadays widely used to solve real-world and complex... more Multi-Agent Reinforcement Learning (MARL) is nowadays widely used to solve real-world and complex decisions in various domains. While MARL can be categorized into independent and cooperative approaches, we consider the independent approach as a simple, more scalable, and less costly method for large-scale distributed systems, such as network packet forwarding. In this paper, we quantitatively and qualitatively assess the benefits of leveraging such independent agents learning approach, in particular IQL-based algorithm, for packet forwarding in computer networking, using the Named Data Networking (NDN) architecture as a driving example. We put multiple IQL-based forwarding strategies (IDQF) to the test and compare their performances against very basic forwarding schemes and simple topologies/traffic models to highlight major challenges and issues. We discuss the main issues related to the poor performance of IDQF and quantify the impact of these issues on isolation when training and...
Comparison of Text Mining Feature Extraction Methods Using Moderated vs Non-Moderated Blogs
Proceedings of the 9th International Conference on Digital Public Health, 2019
Online social media is being widely used by social scientists to study human behavior. Researcher... more Online social media is being widely used by social scientists to study human behavior. Researchers have explored different feature extraction (FE) and classification techniques to perform sentiment analysis, topic identification, etc. Most studies tend to evaluate FE and classification methods using only one particular class of datasets---well-defined with little/no noise or with well-defined noise. For instance, when the datasets under study have different noise characteristics, various FE and/or classification methods may fail to identify a given topic. In this paper, we fill this gap by quantitatively comparing multiple FE methods and classifiers using three different datasets (two moderator-controlled blogs and one single-authored personal blogs) related to Autism Spectrum Disorder (ASD). Our result shows that no particular combination of FE and classifier is the best overall, but choosing the right ones can improve accuracy by over 30%.
Impact of Locational Factors on Business Ratings/Reviews: A Yelp and TripAdvisor Study
Lecture Notes in Social Networks, 2021
The proliferation and the success of crowdsourcing reviews of businesses became a major indicator... more The proliferation and the success of crowdsourcing reviews of businesses became a major indicator of the success or failure of establishments. Particularly, Yelp, Trip Advisor, and Zomato are using a social media platform where users can share feedback, scores, photos, and make reservations. Business owners keep a close eye on these crowd-sourced indicators in order to maintain or improve their ratings. While improvements can be related to many factors such as services, food, cleanliness, etc., in this chapter, we focus on the impact of the location of the business on its success based on these crowdsourced ratings. We perform an empirical study to quantify the impact of location characteristic indicators (parameters), such as cost of living, housing affordability, or tourism, on the success of restaurants as a business example using two datasets: 2019 Yelp, and TripAdvisor. We first, performed a state wise preliminary experiments to verify the correlation between location parameters and business success. We have verified that some location parameters alone, such as education index, can determine the success of a business with a 0.72 correlation ratio. Next, we propose a clustering method that group similar zip code locations to better estimate the influence of the said location parameters on the success scores.
Whispering: Joint Service Offloading and Computation Reuse in Cloud-Edge Networks
ICC 2021 - IEEE International Conference on Communications, 2021
Due to the proliferation of Internet of Things (IoT) and application/user demands that challenge ... more Due to the proliferation of Internet of Things (IoT) and application/user demands that challenge communication and computation, edge computing has emerged as the paradigm to bring computing resources closer to users. In this paper, we present Whispering, an analytical model for the migration of services (service offloading) from the cloud to the edge, in order to minimize the completion time of computational tasks offloaded by user devices and improve the utilization of resources. We also empirically investigate the impact of reusing the results of previously executed tasks for the execution of newly received tasks (computation reuse) and propose an adaptive task offloading scheme between edge and cloud. Our evaluation results show that Whispering achieves up to 35% and 97% (when coupled with computation reuse) lower task completion times than cases where tasks are executed exclusively at the edge or the cloud.
RC-UDP: On Raptor Coding over UDP for Reliable High-Bandwidth Data Transport
2018 IEEE International Conference on Communications (ICC), 2018
Data-driven and collaborative research has become the trend for today's scientific communitie... more Data-driven and collaborative research has become the trend for today's scientific communities, resulting in large- scale datasets being shared and transported through networks every day. Most of these large data transfers use TCP sockets which are known to be limited in long-distance and high-bandwidth scenarios. UDP, on the other hand, while fast and efficient does not implement any reliability mechanisms. In this paper, we investigate the use of erasure coding techniques, namely fountain codes, on top of UDP to help high speed and reliable data transfer applications to attain high bandwidth in the face of packet losses. We propose RC-UDP, a Raptor Code over UDP framework that enables reliable data transfers for high bandwidth networks. We implement RC-UDP and evaluate its performance using computer simulation (ns-3) and real world testbed experimentations. We compare RC-UDP to HighSpeed and CUBIC TCP. Our results show that RC-UDP, which achieves up to 75X time reduction while incurring minimum overhead, is beneficial when the network is subject to high congestion or packet drop rates.
Proceedings of the 2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, 2019
Today, with the emergence of various business review sites such as Yelp, Trip Advisor, and Zomato... more Today, with the emergence of various business review sites such as Yelp, Trip Advisor, and Zomato, people can write reviews and provide an assessment (often as 1-5 score rating). The success of a business on the crowd-sourced review platform has taken the form of positive reviews and high star ratings (failure are associated with negative reviews and low star ratings). We often claim that location plays a major role in determining the success or the failure of a given business. This paper attempts to verify this claim and quantifies the impact of location, solely, on business success, using two data sets; a Yelp dataset for business information and reviews, and another Location dataset that gathers location-based information in a city or an area. We perform an empirical study to quantify the impact of (i) relative location to well known landmarks and (ii) parameterized location (such as cost of living in a given zip code), on the success of restaurants. In our study, we found that parameterized location using location characteristic parameters such as housing affordability correlate highly with restaurant success with more than 0.81 correlation ratio. We also observe that the closer the restaurant to a landmark (relative location) the more likelihood it succeeds.
STAR: STAble routing for hidden interfering primary user problems in mobile cognitive radio networks
MILCOM 2017 - 2017 IEEE Military Communications Conference (MILCOM), 2017
In mobile Cognitive Radio Networks (CRNs), the frequent link breakage and wireless interference u... more In mobile Cognitive Radio Networks (CRNs), the frequent link breakage and wireless interference urged the need for novel stable routing solutions. In this paper, we propose STAR, a stable routing protocol that aims at selecting interference-free and durable paths. We investigate the particular Hidden interfering Primary User (HIPU) problem caused by the un-detectable presence of the primary user in the transmission range of the secondary users resulting on severe packet loss. While existing CRNs routing solution, generally, overlook the interference caused by the Hidden Interfering Primary User (HIPU), STAR defines a HIPU-free transmission zone, that enables successful transmission and selects a stable route that is free from interference along the path to the destination. STAR uses prediction based model to estimate the link stability and the link propagation delay. To quantitatively select the best stable links along the route, STAR implements an objective function, cognitive stab...
Open J. Internet Things, 2019
The technological advancements along with the proliferation of smart and connected devices (thing... more The technological advancements along with the proliferation of smart and connected devices (things) motivated the exploration of the creation of smart cities aimed at improving the quality of life, economic growth, and efficient resource utilization. Some recent initiatives defined a smart city network as the interconnection of the existing independent and heterogeneous networks and the infrastructure. However, considering the heterogeneity of the devices, communication technologies, network protocols, and platforms the interoperability of these networks is a challenge requiring more attention. In this paper, we propose the design of a novel Information-Centric Smart City architecture (iSmart), focusing on the demand of the future applications, such as efficient machineto-machine communication, low latency computation offloading, large data communication requirements, and advanced security. In designing iSmart, we use the Named-Data Networking (NDN) architecture as the underlying co...
Uploads
Papers by Abderrahmen Mtibaa