We introduce multipartite entanglement distribution protocols that use a quantum switch to deliver stabilizer states to a number of remote end users. As in existing schemes, the first step in our protocols involves Bell pair generation between the switch and each end user. However, unlike existing schemes that wait for all Bell pairs to be established before distributing the desired state -- for example, via a projective measurement -- our approach stores only a minimal subset of Bell pairs while processing every subsequent Bell pair immediately. In doing so, our protocols reduce the average Bell pair storage time compared to existing schemes, resulting in less cumulative noise as a direct consequence. On the theoretical side, our protocol design is grounded in the structure of vertex covers in graph states up to local complementation. Through a comprehensive numerical evaluation, we compare the fidelities of delivered states with those of a baseline scheme, for state sizes up to n = 50 qubits. Simulations also show that our protocols can achieve the critical fidelity threshold of 1/2 for multipartite entanglement in a wider range of depolarization rates and success probabilities of Bell-pair generation. Overall, our protocols always achieve an equal or higher fidelity of the distributed state, and can reduce infidelity by up to 45%.

Quantum networks (QNs) relying on free-space optical (FSO) quantum channels can support quantum applications in environments wherein establishing an optical fiber infrastructure is challenging and costly. However, FSO-based QNs require a clear line-of-sight (LoS) between users, which is challenging due to blockages and natural obstacles. In this paper, a reconfigurable intelligent surface (RIS)-assisted FSO-based QN is proposed as a cost-efficient framework providing a virtual LoS between users for entanglement distribution. A novel modeling of the quantum noise and losses experienced by quantum states over FSO channels defined by atmospheric losses, turbulence, and pointing errors is derived. Then, the joint optimization of entanglement distribution and RIS placement problem is formulated, under heterogeneous entanglement rate and fidelity constraints. This problem is solved using a simulated annealing metaheuristic algorithm. Simulation results show that the proposed framework effectively meets the minimum fidelity requirements of all users’ quantum applications. This is in stark contrast to baseline algorithms that lead to a drop of at least 84% in users’ end-to-end fidelities. The proposed framework also achieves a 63% enhancement in the fairness level between users compared to baseline rate maximizing frameworks. Finally, the weather conditions, e.g., rain, are observed to have a more significant effect than pointing errors and turbulence.

Distributed quantum applications impose requirements on the quality of the quantum states that they consume. When analyzing architecture implementations of quantum hardware, characterizing this quality forms an important factor in understanding their performance. Fundamental characteristics of quantum hardware lead to inherent tradeoffs between the quality of states and traditional performance metrics such as throughput. Furthermore, any real-world implementation of quantum hardware exhibits time-dependent noise that degrades the quality of quantum states over time. Here, we study the performance of two possible architectures for interfacing a quantum processor with a quantum network. The first corresponds to the current experimental state of the art in which the same device functions both as a processor and a network device. The second corresponds to a future architecture that separates these two functions over two distinct devices. We model these architectures as continuous-time Markov chains and compare their quality of executing quantum operations and producing entangled quantum states as functions of their memory lifetimes, as well as the time that it takes to perform various operations within each architecture. As an illustrative example, we apply our analysis to architectures based on Nitrogen-Vacancy centers in diamond, where we find that for present-day device parameters one architecture is more suited to computation-heavy applications, and the other for network-heavy ones. We validate our analysis with the quantum network simulator NetSquid. Besides the detailed study of these architectures, a novel contribution of our work are several formulas that connect an understanding of waiting time distributions to the decay of quantum quality over time for the most common noise models employed in quantum technologies. This provides a valuable new tool for performance evaluation experts, and its applications extend beyond the two architectures studied in this work.

The Transmission Control Protocol (TCP) utilizes a congestion avoidance and control mechanism as a preventive measure against congestive collapse and as an adaptive measure in the presence of changing network conditions. The set of available congestion control algorithms is diverse, and while many have been studied from empirical and simulation perspectives, there is a notable lack of analytical work for some variants. To gain more insight into the dynamics of these algorithms, we: (1) propose a general modeling scheme consisting of a set of functional differential equations of retarded type (RFDEs) and of the congestion window as a function of time; (2) apply this scheme to TCP Reno and demonstrate its equivalence to a previous, well known model for TCP Reno; (3) show applications of the new framework to the widely-deployed congestion control algorithm TCP CUBIC, for which analytical models are few and limited; as well as to H-TCP, another high-speed congestion control algorithm; and (4) validate the model using simulations. Our modeling framework yields a fluid model for window- or rate-based congestion control variants. From a theoretical analysis of this model with TCP CUBIC, we discover that CUBIC is locally uniformly asymptotically stable - a property of the algorithm previously unknown. Through further analysis, we derive a sufficient condition for H-TCP's stability, but observe via a numerical analysis and simulations that H-TCP rarely converges to an equilibrium and is usually not asymptotically stable in practical high-speed settings.

Many learning problems are formulated as minimization of some loss function on a training set of examples. Distributed gradient methods on a cluster are often used for this purpose. In this paper, we study how the variability of task execution times at cluster nodes affects the system throughput. In particular, a simple but accurate model allows us to quantity how the time to solve the minimization problem depends on the network of information exchanges among the nodes. Interestingly, we show that, even when communication overhead may be neglected, the clique is not necessarily the most effective topology, as commonly assumed in previous works.

The Transmission Control Protocol (TCP) utilizes congestion avoidance and control mechanisms as a preventive measure against congestive collapse and as an adaptive measure in the presence of changing network conditions. The set of available congestion control algorithms is diverse, and while many have been studied from empirical and simulation perspectives, there is a notable lack of analytical work for some variants. To gain more insight into the dynamics of these algorithms, we: (1) propose a general modeling scheme consisting of a set of functional differential equations of retarded type (RFDEs) and of the congestion window as a function of time; (2) apply this scheme to TCP Reno and demonstrate its equivalence to a previous, well known model for TCP Reno; (3) show an application of the new framework to the widely-deployed congestion control algorithm TCP CUBIC, for which analytical models are few and limited; and (4) validate the model using simulations. Our modeling framework yields a fluid model for TCP CUBIC. From a theoretical analysis of this model, we discover that TCP CUBIC is locally uniformly asymptotically stable-a property of the algorithm previously unknown.

ide-area data transfers in high-performance computing infrastructures are increasingly being carried over dynamically provisioned dedicated network connections that provide high capacities with no competing traffic. We present extensive TCP throughput measurements and time traces over a suite of physical and emulated 10 Gbps connections with 0-366 ms round-trip times (RTTs). Contrary to the general expectation, they show significant statistical and temporal variations, in addition to the overall dependencies on the congestion control mechanism, buffer size, and the number of parallel streams. We analyze several throughput profiles that have highly desirable concave regions wherein the throughput decreases slowly with RTTs, in stark contrast to the convex profiles predicted by various TCP analytical models. We present a generic throughput model that abstracts the ramp-up and sustainment phases of TCP flows, which provides insights into qualitative trends observed in measurements across TCP variants: (i) slow-start followed by well-sustained throughput leads to concave regions; (ii) large buffers and multiple parallel streams expand the concave regions in addition to improving the throughput; and (iii) stable throughput dynamics, indicated by a smoother Poincare map and smaller Lyapunov exponents, lead to wider concave regions. These measurements and analytical results together enable us to select a TCP variant and its parameters for a given connection to achieve high throughput with statistical guarantees.

Dedicated wide-area network connections are increasingly employed in high-performance computing and big data scenarios. One might expect the performance and dynamics of data transfers over such connections to be easy to analyze due to the lack of competing traffic. However, non-linear transport dynamics and end-system complexities (e.g., multi-core hosts and distributed filesystems) can in fact make analysis surprisingly challenging. We present extensive measurements of memory-tomemory and disk-to-disk file transfers over 10 Gbps physical and emulated connections with 0-366 ms round trip times (RTTs). For memory-to-memory transfers, profiles of both TCP and UDT throughput as a function of RTT show concave and convex regions; large buffer sizes and more parallel flows lead to wider concave regions, which are highly desirable. TCP and UDT both also display complex throughput dynamics, as indicated by their Poincarέmaps and Lyapunov exponents. For diskto-disk transfers, we determine that high throughput can be achieved via a combination of parallel I/O threads, parallel network threads, and direct I/O mode. Our measurements also show that Lustre filesystems can be mounted over long-haul connections using LNet routers, although challenges remain in jointly optimizing file I/O and transport method parameters to achieve peak throughput.

In recent years, there has been a steady growth in network bandwidths. This is especially true in scientific and big data environments, where high bandwidth-delay products (BDPs) are common. It is well-understood that legacy TCP (e.g. TCP Reno) is not appropriate for such environments, and several TCP variants were developed to address this shortcoming. These variants, including CUBIC, STCP, and H-TCP, have been studied in some empirical contexts, and some analytical models exist for CUBIC and STCP. However, since these studies were conducted, BDPs further increased, and new bulk data transfer methods have emerged that utilize parallel TCP streams. In view of these new developments, it is imperative to revisit the question: `Which congestion control algorithms are best adapted to current networking environments?' In order to answer this question, (i) we create a general theoretical framework within which to develop mathematical models of TCP variants that account for finite buffer sizes, maximum window constraints, and parallel TCP streams; (ii) we validate the models using measurements collected over a high-bandwidth testbed and achieve low prediction errors; (iii) we find that CUBIC and H-TCP outperform STCP, especially when multiple streams are used.

We investigate the file transfer scheduling problem, where transfers among different endpoints must be scheduled to maximize pertinent metrics. We propose two new algorithms that exploit the fact that the aggregate bandwidth obtained over a network or at a storage system tends to increase with the number of concurrent transfers---but only up to a certain limit. The first algorithm, SEAL, uses runtime information and data-driven models to approximate system load and adapt transfer schedules and concurrency so as to maximize performance while avoiding saturation. We implement this algorithm using GridFTP as the transfer protocol and evaluate it using real transfer logs in a production WAN environment. Results show that SEAL can improve average slowdowns and turnaround times by up to 25% and worst-case slowdown and turnaround times by up to 50%, compared with the best-performing baseline scheme. Our second algorithm, STEAL, further leverages user-supplied categorization of transfers as either "interactive" (requiring immediate processing) or "batch" (less time-critical). Results show that STEAL reduces the average slowdown of interactive transfers by 63% compared to the best-performing baseline and by 21% compared to SEAL. For batch transfers, compared to the best-performing baseline, STEAL improves by 18% the utilization of the bandwidth unused by interactive transfers. By elegantly ensuring a sufficient, but not excessive, allocation of concurrency to the right transfers, we significantly improve overall performance despite constraints.

Data generated by experimental, simulation, and observational science is growing exponentially. The resulting datasets are often transported over wide-area networks for storage, analysis, or visualization. Network bandwidth, which is not increasing at the same rate as dataset sizes, is becoming a key obstacle to data-driven sciences. In this paper, we focus on how bandwidth allocation can be controlled at the level of a protocol such as Grid FTP, in view of goals such as maintaining certain priorities or performing scheduling with specified objectives. In particular, we explore how Grid FTP transfer performance can be controlled by using parallelism and concurrency. We find that concurrency turns out to be a more powerful control knob than is parallelism. For a source where most bandwidth is consumed by transfers to as mall number of other destinations, we build a model for each destination's achieved throughput in terms of its concurrency and total concurrency (over Grid FTP transfers) to other major destinations. We then enhance this model by including an indicator of the time-varying external load, using multiple ways to measure this external load. We study the effectiveness of the proposed models in controlling the bandwidth allocation. After evaluating the numerous combinations of models and methods of measuring external load, we narrow in on the four best-performing ones, based on both their validation results and their applicability. After extensive testing of these four approaches, we find that they can obtain desired bandwidth allocations with a mean(median) error rate of19.8%(13.8%), with 38% of the errors in our benchmark tests being less than 10% and 54% of them being less than 15%.