Energy Efficient Congestion Control for Multipath TCP in Heterogeneous Networks
Not Available
SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Not Available
Congestion in urban areas is a chronic problem that faces many US and worldwide cities. It results in inefficient use of the infrastructure as well as increased fuel consumption and vehicle emissions. Congestion is intertwined with delay. Road users waste precious hours on the road, which reduces productivity. Throughout this decade researchers were and are developing tools and systems that alleviate this problem. Network perimeter control is one of these tools that has been studied extensively. It attempts to control the flow of vehicles entering a protected area so as not to reach the congested regime that is pre-determined using the Network Fundamental Diagram (NFD). In this paper, we present a novel and simple approach derived from sliding mode control theory that achieves the desired objective. Its main advantages over other control strategies include but are not limited to (1) ease of implementation, (2) minimal and straightforward model tuning, and (3) high performance. This controller was implemented and tested on a congested grid network. The results show that, the new logic performs as expected, producing network-wide delay savings that are consistent with other control approaches described in the literature. Furthermore, unlike other controllers, we demonstrate that the controller is robust to its hyper-parameter values, making the algorithm easy to implement and calibrate to local conditions. http://pubsindex.trb.org/
We analyze some systems of partial differential equations arising in the theory of mean field type control with congestion effects. We look for weak solutions. Our main result is the existence and uniqueness of suitably defined weak solutions, which are characterized as the optima of two optimal control problems in duality.
This work deals with a numerical method for solving a mean-field type control problem with congestion. It is the continuation of an article by the same authors, in which suitably defined weak solutions of the system of partial differential equations arising from the model were discussed and existence and uniqueness were proved. Here, the focus is put on numerical methods: a monotone finite difference scheme is proposed and shown to have a variational interpretation. Then an Alternating Direction Method of Multipliers for solving the variational problem is addressed. It is based on an augmented Lagrangian. Two kinds of boundary conditions are considered: periodic conditions and more realistic boundary conditions associated to state constrained problems. Various test cases and numerical results are presented.
Summary of Report In work prior to this grant, NCDM developed a high performance data transport protocol called SABUL. During this grant, we refined SABUL’s functionality, and then extended both the capabilities and functionality and incorporated them into a new protocol called UDP-based Data transport Protocol, or UDT. We also began preliminary work on Composable UDT, a version of UDT that allows the user to choose among different congestion control algorithms and implement the algorithm of his choice at the time he compiles the code. Specifically, we: · Investigated the theoretical foundations of protocols similar to SABUL and UDT. · Performed design and development work of UDT, a protocol that uses UDP in both the data and control channels. · Began design and development work of Composable UDT, a protocol that supports the use of different congestion control algorithms by simply including the appropriate library when compiling the code. · Performed experimental studies using UDT and Composable UDT using real world applications such as the Sloan Digital Sky Survey (SDSS) astronomical data sets. · Released several versions of UDT and Composable, the most recent being v3.1.
Cloud networks utilize virtual connections to connect virtual machines distributed across cloud sites. They are increasingly deployed due to flexible provisioning using software and cost-effectiveness in not requiring to build physical network infrastructure. However, their extensive virtualization makes it unclear how well the established practices of conventional networks translate to them. Here, we study throughput measurements over a Google Cloud network using a matching hardware emulated conventional network, which provide production and exploratory conditions, respectively. The measurements span connections representing local, cross-continental and around the Earth distances. We study the effects of parallel flows, congestion control algorithms and retransmissions on the network throughput profile expressed as a function of RTT. We compare the throughput profile of Google Cloud network with those of emulated network under various loss conditions, including those too disruptive or expensive in the former. Our analysis based on the concave-convex shape and utilization-concavity coefficients of throughput profiles indicates an overall agreement of performance between the two networks, thereby justifying the use of conventional network emulations to analyze cloud networks. In terms of practical use, our study establishes that BBR and BBRv2 alpha TCP achieve higher throughput compared to loss-based congestion control algorithms under most network configurations, especially, under losses at large RTT.
Active distribution networks (ADNs) can provide grid services such as peak load shaving, loss reduction, VoltNar control, congestion management, and frequency regulation. The fast response of inverter-based distributed energy resources (DERs) and battery storage systems enables them to provide ramping support and frequency control. This paper proposes a real-time optimization approach for behind-the-meter DERs to participate in the secondary frequency regulation. The proposed approach determines optimal set-points of output power of DERs to respond to requests from system operators (SOs) in realtime to maintain the frequency of the system at the nominal value. To satisfy the real-time requirement for the secondary frequency control, the nonlinear AC power flow model is linearized considering power losses in the system to achieve both high computational speed and acceptable accuracy. The optimization problem is rerun in a closed loop manner until the mismatch between the requested power by the system operators and actual delivered power at the substation reaches an acceptable tolerance. Furthermore, the proposed approach is validated using modified versions of the IEEE 33-bus and IEEE 69-bus distribution systems. Although the contribution of a single distribution system in the frequency regulation may not be significant, stacked and coordinated contributions from several distribution systems can provide frequency regulation and other grid services at scale.
Wide-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.
Network congestion is one of the biggest problems facing HPC systems today, affecting system throughput, performance, user experience, and reproducibility. Congestion manifests as run-to-run variability due to contention for shared resources (e.g., filesystems) or routes between compute endpoints. Despite its significance, current network benchmarks fail to proxy the real-world network utilization seen on congested systems. We propose a new open-source benchmark suite called the Global Performance and Congestion Network Tests (GPCNeT) to advance the state of the practice in this area. The guiding principles used in designing GPCNeT are described and the methodology employed to maximize its utility is presented. The capabilities of GPCNeT are evaluated by analyzing results from several world's largest HPC systems, including an evaluation of congestion management on a next-generation network. The results show that systems of all technologies and scales are susceptible to congestion and this work motivates the need for congestion control in next-generation networks.
Escalating bandwidth demand strains high-performance data networks, posing potential performance risks. TCP congestion control algorithms enhance reliability and optimize bandwidth usage. Network performance is influenced by factors such as AQM algorithms and router buffer size. In the context of constrained network resources, understanding how TCP flows share networks and the resulting performance impact is essential. This paper introduces insights into TCP fairness and performance involving a comparison of TCP CUBIC, Reno, Hamilton, and BBR versions 1 and 2 across real-world networks supporting high bandwidths of up to 25 Gbps. The research explores TCP behaviors with AQM algorithms like FIFO, FQ_CODEL, and RED, alongside diverse buffer sizes. Notably, findings reveal that manipulating buffers and queuing methods yields contrasting outcomes based on bandwidth. BBRv2 emerges as a superior fair algorithm, pivotal for swift transfers, particularly in scientific data scenarios. These results provide crucial guidance for future network design, ensuring equitable performance optimization.
DISTRI is an advanced network simulator designed for multi-facility computational infrastructures with agentic behavior. It simulates HPC facilities where computational resources act as autonomous agents, making intelligent decisions about job scheduling, load balancing, and resource allocation. The simulator focuses on developing and testing decentralized algorithms that promote resilience and efficiency in multi-facility environments. Key Features: - Agentic Resource Behavior: Processors and DTNs act as autonomous agents with decision-making capabilities - Pheromone-Based Load Balancing: Decentralized load balancing inspired by ant colony optimization - Dual Topology Support: Mesh (normal operations) and Dumbell (network testing) topologies - Comprehensive TCP Simulation: Realistic TCP implementations with multiple congestion control algorithms - Failure Resilience Testing: Processor failure simulation with automatic job reassignment - Extensive Visualization: Detailed performance analysis and metrics collection - Research-Ready: Designed for algorithm development and benchmarking
The combination of analytical models and measurements provide practical configurations and parameters to achieve high data transport rates: (a) buffer sizes and number of parallel streams for improved memory and file transfer rates, (b) Hamilton and Scalable TCP congestion control modules for memory transfers in place of default CUBIC, and (c) direct IO mode for Lustre file systems for wide-area transfers. Conventional parameter selection using full sweeps is impractical in many cases since it takes months. By exploiting the unimodality of throughput profiles, we developed the d-w method that significantly reduces the number of measurements needed for parameter identification. This heuristic method was effective in practice in reducing the measurements by about 90% for Lustre and XFS file transfers.
In recent years, the computer networking community has seen a steady growth in bandwidth-delay products (BDPs). Several TCP variants were created to combat the shortcomings of legacy TCP when it comes to operation in high-BDP environments. These variants, among which are CUBIC, STCP, and H-TCP, have been extensively studied in some empirical contexts, and some analytical models exist for CUBIC and STCP. However, since these studies have been conducted, BDPs have risen even more, and new bulk data transfer tools have emerged that utilize multiple 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 help resolve this question, we contribute the following: (i) using first principles, we develop a general throughput-prediction framework that takes into account buffer sizes and maximum window constraints; (ii) we validate the models using measurements and achieve low prediction errors; (iii) we note differences in TCP dynamics between two experimental configurations and find one of them to be significantly more deterministic than the other; we also find that CUBIC and H-TCP outperform STCP, especially when multiple streams are used; and (iv) we present preliminary results for modelling multiple TCP streams for CUBIC and STCP.
Wide-area data transfers in high-performance computing infrastructures and big data applications are increasingly carried over dynamically provisioned dedicated network connections, which provide high capacities with no competing traffic. Throughput of TCP and UDT over wide-area connections depends, often in a complex non-linear way, on the congestion control mechanism, buffer sizes, and the number of parallel streams. In addition, our extensive TCP and UDT throughput measurements and time traces over a suite of physical and emulated 10 Gbps connections with 0-366 ms round-trip times (RTTs) show significant statistical and temporal variations. Consequently, parameter selection and optimization for transport methods require data analytics to complement analytical throughput models. We present analytics based on the concavity-convexity of throughput profiles which provide insights into peak throughput and superior/inferior trend compared to linear interpolations based on RTT. In particular, we propose the utilization-concavity coefficient, a scalar metric that incorporates both utilization and concavity profile to characterize the overall performance of a transport protocol. These measurements-based anaytics enable us to select a transport protocol and its parameters for a given host and connection configuration to achieve high throughput with statistical guarantees.
Under this SBIR Phase II, Reservoir Labs has developed G2 Analytics, a new technology that allows network operators to analyze bottleneck and flow performance with high precision. G2 delivers a new analytical approach and framework to resolve a variety of key problems found in modern communication networks, including: traffic engineering, routing, flow scheduling, network design, capacity planning, resiliency analysis, network slicing, or service level agreement (SLA) management, among others. G2 leverages the bottleneck structure of congestion-controlled communication networks, a recent mathematical discovery by the Reservoir team [RL19b, RL20a, RL20b, RL21a]. Bottleneck structures reveal how perturbations on flows and links propagate through the network, providing an analytical framework to measure (qualitatively and quantitatively) the ripple effects induced as they traverse the network. Leveraging the mathematics of bottleneck structures, Reservoir Labs is developing the G2 technology to provide network operators with a framework to design, optimize and troubleshoot network performance. This delivery includes the G2 software stack.
Techniques based on the Theory of Bottleneck Ordering can reveal the bottleneck structure of a network, and the Theory of Flow ordering can take advantage of the revealed bottleneck structure to manage and configure network flows so as to improve the overall network performance. These two techniques provide insights into the inherent topological properties of a network at least in three areas: (1) identification of the regions of influence of each bottleneck; (2) the order in which bottlenecks (and flows traversing them) may converge to their steady state transmission rates in distributed congestion control algorithms; and (3) the design of optimized traffic engineering policies.
Traffic congestion is a serious problem in the US, and traffic signal control is one of the effective solutions to congestion. Previous research on model predictive control (MPC)-based traffic signal control showed substantial benefits over conventional methods. This study focused on implementing MPC over a large-scale network with complex intersections and the impact of cycle length, network size, and imperfect state estimation on performances. This study implemented a virtual phase link (VPL)-based model predictive control method which used the number of vehicles in each VPL as input state variables and was suitable for National Electrical Manufacturing Association (NEMA) ring-barrier control. To test the impact of network size, the performance of distributed MPC (36 intersections in the network are divided into five subnetworks) was compared with that of MPC over the full network for a set of cycle lengths. To test the impact of imperfect state estimation, we synthetically infused estimation error and developed two scenarios, MPC-error and MPC-error narrow, which had higher and lower estimation errors, respectively. The performance of these MPC methods was compared with that of the existing time-of-day (TOD) method and an offline method that used Webster's method for split and MULTIBAND for cycle length and offset optimization. Trajectory and linkwise signal performance measures were collected from the simulation to evaluate performance. The distributed MPC method with perfect state estimation had the lowest delay and highest energy efficiency of all the methods. The performance of MPC decreased as the prediction inaccuracy increased. MPC-error had 7% and 11% more delay than MPC-error narrow in the morning and evening peaks, respectively. Overall, simulation results suggest that even with imperfect state estimation, MPC methods will outperform offline methods significantly.
The uncoordinated charging of electric vehicles (EVs) in time and space brings congestion issues to the distribution network. This paper proposes an EV charging demand forecasting-based model predictive control (MPC) method for distribution system congestion management. To effectively forecast the time-series EV station charging demand, a hybrid forecasting model that integrates the long short-term memory network (LSTM) and Transformer is proposed. The Transformer-LSTM model is trained using a one-year real historical charging dataset of EV stations to forecast future charging demand in 15-minute intervals. This informs the MPC for distribution network congestion management and minimization of PV curtailment. Numerical results carried out on the modified IEEE 123-bus distribution system demonstrate that the proposed method can effectively resolve line congestion issues through EV smart charging and PV curtailment while outperforming other benchmarks.