Biblio

Residue number systems provide efficient techniques for speeding up calculations and/or protecting against side channel attacks when used in the context of cryptographic engineering. One of the interests of such systems is their scalability, as the existence of large bases for some specialized systems is often an open question. In this paper, we present highly optimized methods for generating large bases for residue number systems and, in some cases, the largest possible bases. We show their efficiency by demonstrating their improvement over the state-of-the-art bases reported in the literature. This work make it possible to address the problem of the scalability issue of finding new bases for a specific system that arises whenever a parameter changes, and possibly open new application avenues.

With the development of artificial intelligence and cryptography, intelligent cryptography will be the trend of encrypted communications in the future. Abadi designed an encrypted communication model based on a generative adversarial network, which can communicate securely when the adversary knows the ciphertext. The communication party and the adversary fight against each other to continuously improve their own capabilities to achieve a state of secure communication. However, this model can only have a better communication effect under the 16 bits communication length, and cannot adapt to the length of modern encrypted communication. Combine the neural network structure in DCGAN to optimize the neural network of the original model, and at the same time increase the batch normalization process, and optimize the loss function in the original model. Experiments show that under the condition of the maximum 2048-bit communication length, the decryption success rate of communication reaches about 0.97, while ensuring that the adversary’s guess error rate is about 0.95, and the training speed is greatly increased to keep it below 5000 steps, ensuring safety and efficiency Communication.

In this paper is devoted a creation cryptographic protocol for the division of mutual authentication and session key. For secure protocols, suitable cryptographic algorithms were monitored.

The cryptology science has gradually gained importance with our digitalized lives. Ensuring the security of data transmitted, processed and stored across digital channels is a major challenge. One of the frequently used components in cryptographic algorithms to ensure security is substitution-box structures. Random selection-based substitution-box structures have become increasingly important lately, especially because of their advantages to prevent side channel attacks. However, the low nonlinearity value of these designs is a problem. In this study, a dataset consisting of twenty different substitution-box structures have been publicly presented to the researchers. The fact that the proposed dataset has high nonlinearity values will allow it to be used in many practical applications in the future studies. The proposed dataset provides a contribution to the literature as it can be used both as an input dataset for the new post-processing algorithm and as a countermeasure to prevent the success of side-channel analyzes.

Real time data analysis in the context of e.g. realtime monitoring or computational steering is an important tool in many fields of science, allowing scientists to make the best use of limited resources such as sensors and HPC platforms. These tools typically rely on large amounts of continuously collected data that needs to be processed in near-real time to avoid wasting compute, storage, and networking resources. Streaming pipelines are a natural fit for this use case but are inconvenient to use on high-performance computing (HPC) systems because of the diverging system software environment with big data, increasing both the cost and the complexity of the solution. In this paper we propose Neon, a clean-slate design of a streaming data processing framework for HPC systems that enables users to create arbitrarily large streaming pipelines. The experimental results on the Bebop supercomputer show significant performance improvements compared with Apache Storm, with up to 2x increased throughput and reduced latency.

Network slicing qualitatively transforms network infrastructures such that they have maximum flexibility in the context of ever-changing service requirements. While the agility of cloud native network functions (CNFs) demonstrates significant promise, virtualization and softwarization severely degrade the performance of such network functions. Considerable efforts were expended to improve the performance of virtualized systems, and at this stage 10 Gbps throughput is a real target even for container/VM-based applications. Nonetheless, the current performance of CNFs with state-of-the-art enhancements does not meet the performance requirements of next-generation 6G networks that aim for terabit-class throughput. The present pace of performance enhancements in hardware indicates that straightforward optimization of existing system components has limited possibility of filling the performance gap. As it would be reasonable to expect a single silver-bullet technology to dramatically enhance the ability of CNFs, an organic integration of various data-plane technologies with a comprehensive vision is a potential approach. In this paper, we show a future vision of system architecture for terabit-class CNFs based on effective harmonization of the technologies within the wide-range of network systems consisting of commodity hardware devices. We focus not only on the performance aspect of CNFs but also other pragmatic aspects such as interoperability with the current environment (not clean slate). We also highlight the remaining missing-link technologies revealed by the goal-oriented approach.

Deep reinforcement learning has recently been adopted for robot behavior learning, where robot skills are acquired and adapted from data generated by the robot while interacting with its environment through a trial-and-error process. Despite this success, most model-free deep reinforcement learning algorithms learn a task-specific policy from a clean slate and thus suffer from high sample complexity (i.e., they require a significant amount of interaction with the environment to learn reasonable policies and even more to reach convergence). They also suffer from poor initial performance due to executing a randomly initialized policy in the early stages of learning to obtain experience used to train a policy or value function. Model based deep reinforcement learning mitigates these shortcomings. However, it suffers from poor asymptotic performance in contrast to a model-free approach. In this work, we investigate knowledge transfer from a model-based teacher to a task-specific model-free learner to alleviate executing a randomly initialized policy in the early stages of learning. Our experiments show that this approach results in better asymptotic performance, enhanced initial performance, improved safety, better action effectiveness, and reduced sample complexity.

Satellite networks are booming to provide high-speed and low latency Internet access, but the transport layer becomes one of the main obstacles. Legacy end-to-end TCP is designed for terrestrial networks, not suitable for error-prone, propagation delay varying, and intermittent satellite links. It is necessary to make a clean-slate design for the satellite transport layer. This paper introduces a novel Information-centric Hop-by-Hop transport layer design, INTCP. It carries out hop-by-hop packets retransmission and hop-by-hop congestion control with the help of cache and request-response model. Hop-by-hop retransmission recovers lost packets on hop, reduces retransmission delay. INTCP controls traffic and congestion also by hop. Each hop tries its best to maximize its bandwidth utilization and improves end-to-end throughput. The capability of caching enables asynchronous multicast in transport layer. This would save precious spectrum resources in the satellite network. The performance of INTCP is evaluated with the simulated Starlink constellation. Long-distance communication with more than 1000km is carried out. The results demonstrate that, for the unicast scenario INTCP could reduce 42% one-way delay, 53% delay jitters, and improve 60% throughput compared with the legacy TCP. In multicast scenario, INTCP could achieve more than 6X throughput.

Deploying a new network architecture in the Internet requires changing some, but not necessarily all elements between communicating applications. One way to achieve gradual deployment is a proxy or gateway which "translates" between the new architecture and TCP/IP. We present such a proxy, called "Performance Enhancing Proxy for Deploying Network Architectures (PEP-DNA)", which allows TCP/IP applications to benefit from advanced features of a new network architecture without having to be redeveloped. Our proxy is a kernel-based Linux implementation which can be installed wherever a translation needs to occur between a new architecture and TCP/IP domains. We discuss the proxy operation in detail and evaluate its efficiency and performance in a local testbed, demonstrating that it achieves high throughput with low additional latency overhead. In our experiments, we use the Recursive InterNetwork Architecture (RINA) and Information-Centric Networking (ICN) as examples, but our proxy is modular and flexible, and hence enables realistic gradual deployment of any new "clean-slate" approaches.

Recently, network usage has evolved from resource sharing between hosts to content distribution and retrieval. Some emerging network architectures, like Named Data Networking (NDN), focus on the design of content-oriented network paradigm. However, these clean-slate network architectures are difficult to be deployed progressively and deal with the new communication requirements. Multi-Identifier Network (MIN) is a promising network architecture that contains push and pull communication semantics and supports the resolution, routing and extension of multiple network identifiers. MIN's original design was proposed in 2019, which has been improved over the past two years. In this paper, we present the current design and implementation of MIN. We also propose a fallback-based identifier extension scheme to improve the extensibility of the network. We demonstrate that MIN outperforms NDN in the scenario of progressive deployment via IP tunnel.

Name Data Networking (NDN) is a clean-slate network redesign that uses content names for routing and addressing. Facing the fact that TCP/IP is deeply entrenched in the current Internet architecture, NDN has made slow progress in industrial promotion. Meanwhile, new architectures represented by SDN, P4, etc., provide a flexible and programmable approach to network research. As a result, a centralized NDN routing mechanism is needed in the scenario for network integration between NDN and TCP/IP. Combining the NLSR protocol and the P4 environment, we introduce an efficient NDN routing mechanism that offers extensible NDN routing services (e.g., resources-location management and routing calculation) which can be programmed in the control plane. More precisely, the proposed mechanism allows the programmable switches to transmit NLSR packets to the control plane with the extended data plane. The NDN routing services are provided by control plane application which framework bases on resource-location mapping to achieve part of the NLSR mechanism. Experimental results show that the proposed mechanism can reduce the number of routing packets significantly, and introduce a slight overhead in the controller compared with NLSR simulation.

The six papers in this special section focuses on services and software computing. Services computing provides a foundation to build software systems and applications over the Internet as well as emerging hybrid networked platforms motivated by it. Due to the open, dynamic, and evolving nature of the Internet, new features were born with these Internet-scale and service-based software systems. Such systems should be situation- aware, adaptable, and able to evolve to effectively deal with rapid changes of user requirements and runtime contexts. These emerging software systems enable and require novel methods in conducting software requirement, design, deployment, operation, and maintenance beyond existing services computing technologies. New programming and lifecycle paradigms accommodating such Internet- scale and service-based software systems, referred to as Internetware, are inevitable. The goal of this special section is to present the innovative solutions and challenging technical issues, so as to explore various potential pathways towards Internet-scale and service-based software systems.

Widespread deployment of centralized clouds has changed the way internet services are developed, deployed and operated. Centralized clouds have substantially extended the market opportunities for online services, enabled new entities to create and operate internet-scale services, and changed the way traditional companies run their operations. However, there are types of services that are unsuitable for today's centralized clouds such as highly interactive virtual and augmented reality (VR/AR) applications, high-resolution gaming, virtualized RAN, mass IoT data processing and industrial robot control. They can be broadly categorized as either latency-sensitive network functions, latency-sensitive applications, and/or high-bandwidth services. What these basic functions have in common is the need for a more distributed cloud infrastructure—an infrastructure we call edge clouds. In this paper, we examine the evolution of clouds, and edge clouds especially, and look at the developing market for edge clouds and what developments are required in networking, hardware and software to support them.

Deterministic latency is the key challenge that must be addressed in numerous 5G applications such as AR/VR. However, it is difficult to make customized end-to-end resource reservation across multiple ISPs using IP-based QoS mechanisms. Information-Centric Networking (ICN) provides scalable and efficient content distribution at the Internet scale due to its in-network caching and native multicast capabilities, and the deterministic latency can promisingly be guaranteed by caching the relevant content objects in appropriate locations. Existing proposals formulate the ICN cache placement problem into numerous theoretical models. However, the underlying mechanisms to support such cache coordination are not discussed in detail. Especially, how to efficiently make cache reservation, how to avoid route oscillation when content cache is updated and how to conduct the real-time latency measurement? In this work, we propose Latency Deterministic Information-Centric Networking (LD-ICN). LD-ICN relies on source routing-based latency telemetry and leverages an on-path caching technique to avoid frequent route oscillation while still achieve the optimal cache placement under the SDN architecture. Extensive evaluation shows that under LD-ICN, 90.04% of the content requests are satisfied within the hard latency requirements.

In this paper, we address the web service selection problem for linear workflows. Given a linear workflow specifying a set of ordered tasks and a set of candidate services providing different features for each task, the selection problem deals with the objective of selecting the most eligible service for each task, given the ordering specified. A number of approaches to solving the selection problem have been proposed in literature. With web services growing at an incredible pace, service selection at the Internet scale has resurfaced as a problem of recent research interest. In this work, we present our approach to the selection problem using an abstraction refinement technique to address the scalability limitations of contemporary approaches. Experiments on web service benchmarks show that our approach can add substantial performance benefits in terms of space when compared to an approach without our optimization.

Malicious scripts are an important computer infection threat vector for computer users. For internet-scale processing, static analysis offers substantial computing efficiencies. We propose the ScriptNet system for neural malicious JavaScript detection which is based on static analysis. We also propose a novel deep learning model, Pre-Informant Learning (PIL), which processes Javascript files as byte sequences. Lower layers capture the sequential nature of these byte sequences while higher layers classify the resulting embedding as malicious or benign. Unlike previously proposed solutions, our model variants are trained in an end-to-end fashion allowing discriminative training even for the sequential processing layers. Evaluating this model on a large corpus of 212,408 JavaScript files indicates that the best performing PIL model offers a 98.10% true positive rate (TPR) for the first 60K byte subsequences and 81.66% for the full-length files, at a false positive rate (FPR) of 0.50%. Both models significantly outperform several baseline models. The best performing PIL model can successfully detect 92.02% of unknown malware samples in a hindsight experiment where the true labels of the malicious JavaScript files were not known when the model was trained.

Photo service providers are facing critical challenges of dealing with the huge amount of photo storage, typically in a magnitude of billions of photos, while ensuring national-wide or world-wide satisfactory user experiences. Distributed photo caching architecture is widely deployed to meet high performance expectations, where efficient still mysterious caching policies play essential roles. In this work, we present a comprehensive study on internet-scale photo caching algorithms in the case of QQPhoto from Tencent Inc., the largest social network service company in China. We unveil that even advanced cache algorithms can only perform at a similar level as simple baseline algorithms and there still exists a large performance gap between these cache algorithms and the theoretically optimal algorithm due to the complicated access behaviors in such a large multi-tenant environment. We then expound the reasons behind this phenomenon via extensively investigating the characteristics of QQPhoto workloads. Finally, in order to realistically further improve QQPhoto cache efficiency, we propose to incorporate a prefetcher in the cache stack based on the observed immediacy feature that is unique to the QQPhoto workload. The prefetcher proactively prefetches selected photos into cache before they are requested for the first time to eliminate compulsory misses and promote hit ratios. Our extensive evaluation results show that with appropriate prefetching we improve the cache hit ratio by up to 7.4 percent, while reducing the average access latency by 6.9 percent at a marginal cost of 4.14 percent backend network traffic compared to the original system that performs no prefetching.

The Internet, the de facto platform for large-scale content distribution, suffers from two issues that limit its manageability, efficiency, and evolution. First, the IP-based Internet is host-centric and agnostic to the content being delivered and, second, the tight coupling of the control and data planes restrict its manageability, and subsequently the possibility to create dynamic alternative paths for efficient content delivery. Here, we present the CURLING system that leverages the emerging Information-Centric Networking paradigm for enabling cost-efficient Internet-scale content delivery by exploiting multicasting and in-network caching. Following the software-defined networking concept that decouples the control and data planes, CURLING adopts an interdomain hop-by-hop content resolution mechanism that allows network operators to dynamically enforce/change their network policies in locating content sources and optimizing content delivery paths. Content publishers and consumers may also control content access according to their preferences. Based on both analytical modeling and simulations using real domain-level Internet subtopologies, we demonstrate how CURLING supports efficient Internet-scale content delivery without the necessity for radical changes to the current Internet.

The tremendous growth of cloud computing and large-scale data analytics highlight the importance of reducing datacenter power consumption and environmental impact of brown energy. While many Internet service operators have at least partially powered their datacenters by green energy, it is challenging to effectively utilize green energy due to the intermittency of renewable sources, such as solar or wind. We find that the geographical diversity of internet-scale services can be carefully scheduled to improve the efficiency of applying green energy in datacenters. In this paper, we propose a holistic heterogeneity-aware cloud workload management approach, sCloud, that aims to maximize the system goodput in distributed self-sustainable datacenters. sCloud adaptively places the transactional workload to distributed datacenters, allocates the available resource to heterogeneous workloads in each datacenter, and migrates batch jobs across datacenters, while taking into account the green power availability and QoS requirements. We formulate the transactional workload placement as a constrained optimization problem that can be solved by nonlinear programming. Then, we propose a batch job migration algorithm to further improve the system goodput when the green power supply varies widely at different locations. Finally, we extend sCloud by integrating a flexible batch job manager to dynamically control the job execution progress without violating the deadlines. We have implemented sCloud in a university cloud testbed with real-world weather conditions and workload traces. Experimental results demonstrate sCloud can achieve near-to-optimal system performance while being resilient to dynamic power availability. sCloud with the flexible batch job management approach outperforms a heterogeneity-oblivious approach by 37 percent in improving system goodput and 33 percent in reducing QoS violations.

Network dynamics, such as bandwidth fluctuation and unexpected latency, hurt users' quality of experience (QoE) greatly for media services over the Internet. In this work, we propose a neural adaptive transport (NAT) framework to tackle the network dynamics for Internet-scale interactive media services. The entire NAT system has three major components: a learning based cloud overlay routing (COR) scheme for the best delivery path to bypass the network bottlenecks while offering the minimal end-to-end latency simultaneously; a residual neural network based collaborative video processing (CVP) system to trade the computational capability at client-end for QoE improvement via learned resolution scaling; and a deep reinforcement learning (DRL) based adaptive real-time streaming (ARS) strategy to select the appropriate video bitrate for maximal QoE. We have demonstrated that COR could improve the user satisfaction from 5% to 43%, CVP could reduce the bandwidth consumption more than 30% at the same quality, and DRL-based ARS can maintain the smooth streaming with \textbackslashtextless; 50% QoE improvement, respectively.

This paper presents the vision of establishing a global service for Global IoT Device Discovery and Integration (GIDDI). The establishment of a GIDDI will: (1) make IoT application development more efficient and cost-effective via enabling sharing and reuse of existing IoT devices owned and maintained by different providers, and (2) promote deployment of new IoT devices supported by a revenue generation scheme for their providers. More specifically, this paper proposes a distributed IoT blockchain ledger that is specifically designed for managing the metadata needed to describe IoT devices and the data they produce. This GIDDI Blockchain is Internet-owned (i.e., it is not controlled by any individual or organization) and is Internet-scaled (i.e., it can support the discovery and reuse billions of IoT devices). The paper also proposes a GIDDI Marketplace that provides the functionality needed for IoT device registration, query, integration, payment and security via the proposed GIDDI Blockchain. We outline the GIDDI Blockchain and Marketplace implementation. We also discuss ongoing research for automatically mining the IoT Device metadata needed for IoT Device query and integration from the data produce. This significantly reduces the need for IoT device providers to supply the metadata descriptions the devices and the data they produce during the registration of IoT Devices in the GIDDI Blockchain.

Software defined systems use virtualization technologies to provide an abstraction of the hardware infrastructure at different layers. Ultimately, the adoption of software defined systems in all cloud infrastructure components will lead to Software Defined Cloud Computing. Nevertheless, virtualization has already been used for years and is a key element of cloud computing. Traditionally, virtual machines are deployed in cloud infrastructure and used to execute applications on common operating systems. New lightweight virtualization technologies, such as containers and unikernels, appeared later to improve resource efficiency and facilitate the decomposition of big monolithic applications into multiple, smaller services. In this work, we present and empirically evaluate four popular unikernel technologies, Docker containers and Docker LinuxKit. We deployed containers both on bare metal and on virtual machines. To fairly evaluate their performance, we created similar applications for unikernels and containers. Additionally, we deployed full-fledged database applications ported on both virtualization technologies. Although in bibliography there are a few studies which compare unikernels and containers, in our study for the first time, we provide a comprehensive performance evaluation of clean-slate and legacy unikernels, Docker containers and Docker LinuxKit.

The explosive growth of mobile traffic in the Internet makes content delivery a challenging issue to cope with. To promote efficiency of content distribution and reduce network cost, Internet Service Providers (ISPs) and content providers (CPs) are motivated to cooperatively work. As a clean-slate solution, nowadays Information-Centric Networking architectures have been proposed and widely researched, where the thought of in-network caching, especially edge caching, can be applied to mobile wireless networks to fundamentally address this problem. Considered the profit split issue between ISPs and CPs and the influence of content popularity is largely ignored, in this paper, we propose a Stackelberg-based optimal network profit split scheme for content delivery in information-centric wireless networks. Simulation results show that the performance of our proposed model is comparable to its centralized solution and obviously superior to current ISP-CP cooperative schemes without considering cache deployment in the network.

Research on Future Internet has gained traction in recent years, with a variety of clean-slate network architectures being proposed. The realization of such proposals may lead to a period of coexistence with the current Internet, creating a heterogeneous Future Internet. In such a vision, mobile nodes (MNs) can move across access networks supporting different network architectures, while being able to maintain the access to content during this movement. In order to support such scenarios, this paper proposes an inter-network architecture mobility framework that allows MNs to move across different network architectures without losing access to the contents being accessed. The usage of the proposed framework is exemplified and evaluated in a mobility scenario targeting IP and NDN network architectures in a content retrieval use case. The obtained results validate the proposed framework while highlighting the impact on the overall communication between the MN and content source.

This paper reports the definition, setup and obtained results of the Fed4FIRE + medium experiment ERASER, aimed to evaluate the actual Quality of Service (QoS) guarantees that the clean-slate Recursive InterNetwork Architecture (RINA) can deliver to heterogeneous applications at large-scale. To this goal, a 37-Node 5G metro/regional RINA network scenario, spanning from the end-user to the server where applications run in a datacenter has been configured in the Virtual Wall experimentation facility. This scenario has initially been loaded with synthetic application traffic flows, with diverse QoS requirements, thus reproducing different network load conditions. Next,their experienced QoS metrics end-to-end have been measured with two different QTA-Mux (i.e., the most accepted candidate scheduling policy for providing RINA with its QoS support) deployment scenarios. Moreover, on this RINA network scenario loaded with synthetic application traffic flows, a real HD (1080p) video streaming demonstration has also been conducted, setting up video streaming sessions to end-users at different network locations, illustrating the perceived Quality of Experience (QoE). Obtained results in ERASER disclose that, by appropriately deploying and configuring QTA-Mux, RINA can yield effective QoS support, which has provided perfect QoE in almost all locations in our demo when assigning video traffic flows the highest (i.e., Gold) QoS Cube.