FABRIC is a national and international scale research infrastructure built to enable cutting-edge research in a wide range of networking and computing areas and domain sciences that depend on advanced networking and computing capabilities. Since its full operation in October 2023, FABRIC has seen rapidly emerging experiments in different research topics. This paper provides a description of FABRIC’s hardware resources, software services, and an analysis of the characteristics of the emerging experiments that are leveraging FABRIC’s unique capabilities. The analysis showed a wide spectrum of topic areas, frequent emphasis on at-scale and high performance experiments that interact with real facilities and real people, and growing complexity and rigor in tools, data and measurement. Researchers are seen sharing well-developed artifacts with others to enable larger experiments and reproducible research.

Denoising and demosaicking are two essential steps to reconstruct a clean full-color image from the raw data. Recently, joint denoising and demosaicking (JDD) for burst images, namely JDD-B, has attracted much attention by using multiple raw images captured in a short time to reconstruct a single high-quality image. One key challenge of JDD-B lies in the robust alignment of image frames. State-of-the-art alignment methods in feature domain cannot effectively utilize the temporal information of burst images, where large shifts commonly exist due to camera and object motion. In addition, the higher resolution (e.g., 4K) of modern imaging devices results in larger displacement between frames. To address these challenges, we design a differentiable two-stage alignment scheme sequentially in patch and pixel level for effective JDD-B. The input burst images are firstly aligned in the patch level by using a differentiable progressive block matching method, which can estimate the offset between distant frames with small computational cost. Then we perform implicit pixel-wise alignment in full-resolution feature domain to refine the alignment results. The two stages are jointly trained in an end-to-end manner. Extensive experiments demonstrate the significant improvement of our method over existing JDD-B methods. Codes are available at https://github.com/GuoShi28/2StageAlign.

Video super-resolution (VSR) aims to improve the spatial resolution of low-resolution (LR) videos. Existing VSR methods are mostly trained and evaluated on synthetic datasets, where the LR videos are uniformly downsampled from their high-resolution (HR) counterparts by some simple operators (e.g., bicubic downsampling). Such simple synthetic degradation models, however, cannot well describe the complex degradation processes in real-world videos, and thus the trained VSR models become ineffective in real-world applications. As an attempt to bridge the gap, we build a real-world video super-resolution (RealVSR) dataset by capturing paired LR-HR video sequences using the multi-camera system of iPhone 11 Pro Max. Since the LR-HR video pairs are captured by two separate cameras, there are inevitably certain misalignment and luminance/color differences between them. To more robustly train the VSR model and recover more details from the LR inputs, we convert the LR-HR videos into YCbCr space and decompose the luminance channel into a Laplacian pyramid, and then apply different loss functions to different components. Experiments validate that VSR models trained on our RealVSR dataset demonstrate better visual quality than those trained on synthetic datasets under real-world settings. They also exhibit good generalization capability in cross-camera tests. The dataset and code can be found at https://github.com/IanYeung/RealVSR.

A powerful path computation element is a must for the next-generation IP-over-Optical networks to support on-demand service provisioning crossing layers. The main purpose of path computation in multi-layer networks (MLN) is to enable traffic engineering (TE) under a multitude of constraints. A common objective of such traffic engineering is to minimize resource utilization while satisfying all the applicable constraints. In this study, we investigate a variety of MLN path computation methods based on real-world control plane implementations and network settings. We present some techniques for handling the specific TE constraints that concern the IP-over-Optical networks. Experimental results are presented for performance evaluation and comparison of the investigated multi-constraint path computation methods and techniques.

Lightpath scheduling is an important capability in next-generation wavelength-division multiplexing (WDM) optical networks to reserve resources in advance for a specified time period while provisioning end-to-end lightpaths. In a dynamic environment, the end user requests for dynamic scheduled lightpath demands (D-SLDs) need to be serviced without the knowledge of future requests. Even though the starting time of the request may be hours or days from the current time, the end-user however expects a quick response as to whether the request could be satisfied. We propose a two- phase approach to dynamically schedule and provision D-SLDs. In the first phase, termed the deterministic lightpath scheduling phase, upon arrival of a lightpath request, the network control plane schedules a path with guaranteed resources so that the user can get a quick response with a deterministic lightpath schedule. In the second phase, termed the lightpath re-optimization phase, we re-provision some already scheduled lightpaths to re-optimize for improving network performance. We study two re- optimization scenarios to reallocate network resources while maintaining the existing lightpath schedules. Experimental results show that our proposed two-phase dynamic lightpath scheduling approach can greatly reduce network blocking.

A new breed of network intensive "e-science" applications are emerging that require a new perspective on how networks provide resources to the user. These e-science applications often have large raw capacity requirements (tens of gigabits/second) and require global reach. The discipline scientists expect this type of performance to be deterministic and repeatable, and available in conjunction with other resources such as sensors, instruments, computational clusters, storage arrays, and the like. The DRAGON project, an NSF funded testbed in the Washington DC metropolitan area, has been developing open source generalized multi-protocol label swapping (GMPLS) based service concepts that provide dedicated and deterministic network resources to the e-science application. Using the GMPLS control plane and DRAGON extensions, an entire "application specific topology" can be established in an automated fashion across administrative boundaries, across heterogeneous network technologies, and can be instantiated in seconds rather than weeks or months. This paper describes the technologies being developed by the DRAGON project and some of the ways application specific topologies are being applied to advanced networked applications.

Lightpath scheduling is an important capability in next-generation wavelength-division multiplexing (WDM) optical networks to reserve resources in advance for a specified time period while provisioning end-to-end lightpaths. In this study, we propose an approach to support dynamic lightpath scheduling in such networks. To minimize blocking probability in a network that accommodates dynamic scheduled lightpath demands (D- SLDs), resource allocation should be optimized in a dynamic manner. However, for the network users who desire deterministic services, resources must be reserved in advance and guaranteed for future use. These two objectives may be mutually incompatible. Therefore, we propose a two-phase dynamic lightpath scheduling approach to tackle this issue. The first phase is the deterministic lightpath scheduling phase. When a lightpath request arrives, the network control plane schedules a path with guaranteed resources so that the user can get a quick response with the deterministic lightpath schedule. The second phase is the lightpath re-optimization phase, in which the network control plane re-provisions some already scheduled lightpaths. Experimental results show that our proposed two-phase dynamic lightpath scheduling approach can greatly reduce WDM network blocking.

Emerging network applications tend to be built over heterogeneous network resources spanning multiple management domains. Many such applications have dynamic demands for dedicated, deterministic, high-bandwidth connections. The generalized multi-protocol label switching (GMPLS) networks under development can address these kinds of demands by using policy-based resource management and service provisioning technologies. In this paper, we present the architecture and implementation for policy-based resource management and service provisioning as part of the work on the NSF funded dynamic resource allocation via GMPLS optical networks (DRAGON) Project. This work captures several critical features of service-oriented GMPLS networks, including a) collaborative interdomain resource management; b) interdomain end-to-end path computation; c) advance scheduled provisioning; and d) authentication, authorization and accounting (AAA). These features rely on the capability of exchanging and coordinating resource and policy information among multiple participating domains. In pursuit of this capability, we propose a three-dimensional (3D) resource computation model (RCM). The three broad dimensions of this model are traffic engineering (TE) constraints, time schedule constraints, and AAA policy constraints. The 3D RCM facilitates policy based resource allocation based on many specific constraints within these three broad categories. Based on this model we describe our approach to GMPLS interdomain end-to-end path computation, advance scheduled provisioning, and AAA policy based provisioning, respectively. Our current DRAGON implementation status is also reported in this paper.

Rapid progress in deployment of national and regional optical network infrastructures holds the promise to provide abundant, inexpensive bandwidth to scientific communities. The expectation is that this will provide a catalyst for innovation across a range of eScience applications. However, there remains considerable architecture and research work in order to allow such applications to access a common shared high-speed infrastructure efficiently and flexibly. This paper presents an experimental solution currently being designed and tested over the Dynamic resource allocation via GMPLS optical networks (DRAGON) and the Internet2 hybrid optical and packet infrastructure (HOPI). This implementation provides deterministic network services to high-end eScience applications in the form of Ethernet "circuits" using GMPLS to dynamically provision user specific VLAN based configurations. This requires addressing several research issues associated with multi-layer networks, interdomain routing, signaling, and multi-constraint path computation. The larger vision is that of a shared infrastructure consisting of multiple network technology layers over which an intelligent control plane can provide for a unified service model. The instantiation described here of Ethernet over WDM provides an ideal platform for both experimental research and, real services, and a migration path to a production infrastructure.

Inter-domain path computation, or the ability to compute end-to-end paths across multiple domains, is the next step toward wide deployment of a distributed control plane with support for traffic engineering. A key enabler to achieve this goal is the introduction of a Path Computation Element (PCE) in each domain. There are various ways these elements can collaborate to compute an end-to-end path; of particular interest to us in this paper is cooperative path computation, a scheme where PCEs exchange path information in the context of a specific end-to-end path computation instance, often prior to signaling the path. We show that depending on the information available to each PCE, cooperation can take one of two forms, which we call model-based and ad hoc. We demonstrate that model-based cooperation is essentially a multistage decision problem, and offer a probabilistic analysis which we believe is the key to understanding the problem and developing efficient inter-domain path computation heuristics. In particular, we argue that having an estimate of the blocking probability in each domain can be helpful in determining the path computation effort needed to find an end-to-end path.

The static provisioning problem in wavelength-routed optical networks has been studied for many years. However, service providers are still facing the challenges arising from the special requirements for provisioning services at the optical layer. In this paper, we incorporate some realistic constraints into the static provisioning problem, and formulate it under different network resource availability conditions. We consider three classes of shared risk link group (SRLG)-diverse path protection schemes: dedicated, shared, and unprotected. We associate with each connection request a lightpath length constraint and a revenue value. When the network resources are not sufficient to accommodate all the connection requests, the static provisioning problem is formulated as a revenue maximization problem, whose objective is maximizing the total revenue value. When the network has sufficient resources, the problem becomes a capacity minimization problem with the objective of minimizing the number of used wavelength-links. We provide integer linear programming (ILP) formulations for these problems. Because solving these ILP problems is extremely time consuming, we propose a tabu search heuristic to solve these problems within a reasonable amount of time. We also develop a rerouting optimization heuristic, which is based on previous work. Experimental results are presented to compare the solutions obtained by the tabu search heuristic and the rerouting optimization heuristic. For both problems, the tabu search heuristic outperforms the rerouting optimization heuristic.

The next-generation SONET metro network is evolving into a service-rich infrastructure. At the edge of such a network, multi-service provisioning platforms (MSPPs) provide efficient data mapping enabled by generic framing procedure (GFP) and virtual concatenation (VC). The core of the network tends to be a meshed architecture equipped with multi-service switches (MSSs), In the context of these emerging technologies, we propose a load-balancing spare capacity reallocation approach to improve network utilization in the next-generation SONET metro networks. Using our approach, carriers can postpone network upgrades, resulting in increased revenue with reduced capital expenditures (CAPEX). For the first time, we consider the spare capacity reallocation problem from a capacity upgrade and network planning perspective. Our approach can operate in the context of shared-path protection (with backup multiplexing) because it reallocates spare capacity without disrupting working services. Unlike previous spare capacity reallocation approaches which aim at minimizing total spare capacity, our load-balancing approach minimizes the network load vector (NLV), which is a novel metric that reflects the network load distribution. Because NLV takes into consideration both uniform and non-uniform link capacity distribution, our approach can benefit both uniform and non-uniform networks. We develop a greedy load-balancing spare capacity reallocation (GLB-SCR) heuristic algorithm to implement this approach. Our experimental results show that GLB-SCR outperforms a previously proposed algorithm (SSR) in terms of established connection capacity and total network capacity in both uniform and non-uniform networks.

Next-generation optical transport networks will automatically and dynamically provision end-to-end connections. In this paper, we study the problem of inter-domain dynamic routing under a multi-layer multi-domain network model, which allows the end-to-end connections to be set up not only across multiple routing domains but also through two transport layers: the optical layer and the digital layer. In this model, a connection can traverse the domain boundary either through optical bypass or through optical-electrical-optical (O/E/O) processing. We propose an inter-domain dynamic routing scheme with modest time complexity to address the problem from an algorithmic perspective.

Translucent WDM optical networks use sparse placement of regenerators to overcome the impairments and wavelength contention introduced by fully transparent networks, and achieve a performance close to fully opaque networks with much less cost. Our previous study proved the feasibility of translucent networks using the sparse regeneration technique. We addressed the placement of regenerators based on static schemes allowing only fixed number of regenerators at fixed locations. This paper furthers the study by proposing a suite of dynamical routing schemes. Dynamic allocation, advertisement and discovery of regeneration resources are proposed to support sharing transmitters and receivers between regeneration and access functions. This study follows the current trend in the optical networking industry by utilizing the extension of IP control protocols. Dynamic routing algorithms, aware of current regeneration resources and link states, are designed to smartly route the connection requests under quality constraints. A hierarchical network model, supported by the MPLS-based control plane, is also proposed to provide scalability. Experiments show that network performance is improved without placement of extra regenerators.

An analytical model for virtual topology reconfiguration (VTR) in optical networks is developed. It aims at the optical networks with a circuit-based data plane and an IP-like control plane. By identifying and analyzing the important factors impacting the network performance due to VTR operations on both planes, we can compare the benefits and penalties of different VTR algorithms and policies. The best VTR scenario can be adaptively chosen from a set of such algorithms and policies according to the real-time network situations. For this purpose, a cost model integrating all these factors is created to provide a comparison criterion independent of any specific VTR algorithm and policy. A case study based on simulation experiments is conducted to illustrate the application of our models.

This paper proposes an alternate approach to fully transparent and fully opaque optical networks for operating a wavelength routed optical network. The architecture of the regeneration node that performs sparse regeneration (or translucency) is discussed. The regeneration demands generated from call blocking and signal quality requirements are addressed. Two implementation strategies for incorporating sparse regeneration are introduced and their relative merits are studied.