An official website of the United States government
It is challenging to meet the bandwidth and latency requirements of interactive real-time applications (e.g., virtual reality, cloud gam- ing, etc.) on time-varying 5G cellular links. Today’s feedback-based congestion controllers try to match the sending rate at the endhost with the estimated network capacity. However, such controllers can- not precisely estimate the cellular link capacity that changes at timescales smaller than the feedback delay. We instead propose a different approach for controlling congestion on 5G links. We send real-time data streams using an imprecise controller (that errs on the side of overestimating network capacity) to ensure high through- put, and then adapt the transmitted content by dropping appropriate packets in the cellular base stations to match the actual capacity and minimize delay. We build a system called Octopus to realize this ap- proach. Octopus provides parameterized primitives that applications at the endhost can configure differently to express different content adaptation policies. Octopus transport encodes the corresponding app-specified parameters in packet header fields, which the base- station logic can parse to execute the desired dropping behavior. Our evaluation shows how real-time applications involving standard and volumetric videos can be designed to exploit Octopus, and achieve 1.5–18× better performance than state-of-the-art schemes.
more »
« less
A Reality-Conforming Approach for QoS Performance Analysis of AFDX in Cyber-Physical Avionics Systems
AFDX (Avionics Full Duplex Switched Ethernet) is developed to support mission-critical communications while providing deterministic Quality of Service (QoS) across cyber-physical avionics systems. Currently, AFDX utilizes FP/FIFO QoS mechanisms to guarantee its real-time performance. To analyze the real-time performance of avionic systems in their design processes, existing work analyzes the deterministic delay bound of AFDX using NC (Network Calculus). However, existing analytical work is based on an unrealistic assumption leading to assumed worst cases that may not be achievable in reality. In this paper, we present a family of algorithms that can search for realistic worst-case delay scenarios in both preemptive and non-preemptive situations. Then we integrate the proposed algorithms with NC and apply our approach to analyzing tandem AFDX networks. Our reality-conforming approach yields tighter delay bound estimations than the state of the art. When there are 100 virtual links in AFDX networks, our method can provide delay bounds more than 25% tighter than those calculated by the state of the art in our evaluation. Moreover, when using our reality-conforming method in the design process, it leads to 27.2% increase in the number of virtual links accommodated by the network in the tandem scenario.
more »
« less
- PAR ID:
- 10297125
- Publisher / Repository:
- IEEE
- Date Published:
- Journal Name:
- 2021 IEEE/ACM 29th International Symposium on Quality of Service (IWQOS)
- Page Range / eLocation ID:
- 1 to 6
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
It is challenging to meet the bandwidth and latency requirements of interactive real-time applications (e.g., virtual reality, cloud gaming, etc.) on time-varying 5G cellular links. Today’s feedback-based congestion controllers try to match the sending rate at the endhost with the estimated network capacity. However, such controllers cannot precisely estimate the cellular link capacity that changes at timescales smaller than the feedback delay. We instead propose a different approach for controlling congestion on 5G links. We send real-time data streams using an imprecise controller (that errs on the side of overestimating network capacity) to ensure high throughput, and then adapt the transmitted content by dropping appropriate packets in the cellular base stations to match the actual capacity and minimize delay. We build a system called Octopus to realize this approach. Octopus provides parameterized primitives that applications at the endhost can configure differently to express different content adaptation policies. Octopus transport encodes the corresponding app-specified parameters in packet header fields, which the basestation logic can parse to execute the desired dropping behavior. Our evaluation shows how real-time applications involving standard and volumetric videos can be designed to exploit Octopus, and achieve 1.5–18× better performance than state-of-the-art schemes.more » « less
-
Deep neural network (DNN) models are increasingly deployed in real-time, safety-critical systems such as autonomous vehicles, driving the need for specialized AI accelerators. However, most existing accelerators support only non-preemptive execution or limited preemptive scheduling at the coarse granularity of DNN layers. This restriction leads to frequent priority inversion due to the scarcity of preemption points, resulting in unpredictable execution behavior and, ultimately, system failure. To address these limitations and improve the real-time performance of AI accelerators, we propose DERCA, a novel accelerator architecture that supports fine-grained, intra-layer flexible preemptive scheduling with cycle-level determinism. DERCA incorporates an on-chip Earliest Deadline First (EDF) scheduler to reduce both scheduling latency and variance, along with a customized dataflow design that enables intralayer preemption points (PPs) while minimizing the overhead associated with preemption. Leveraging the limited preemptive task model, we perform a comprehensive predictability analysis of DERCA, enabling formal schedulability analysis and optimized placement of preemption points within the constraints of limited preemptive scheduling. We implement DERCA on the AMD ACAP VCK190 reconfigurable platform. Experimental results show that DERCA outperforms state-of-the-art designs using non-preemptive and layer-wise preemptive dataflows, with less than 5 % overhead in worst-case execution time (WCET) and only 6% additional resource utilization. DERCA is open-sourced on GitHub: https://github.com/arc-research-lab/DERCAmore » « less
-
Wireless links using massive MIMO transceivers are vital for next generation wireless communications networks. Precoding in Massive MIMO transmission requires accurate downlink channel state information (CSI). Many recent works have effectively applied deep learning (DL) to jointly train UE-side compression networks for delay domain CSI and a BS-side decoding scheme. Vitally, these works assume that the full delay domain CSI is available at the UE, but in reality, the UE must estimate the delay domain based on a limited number of frequency domain pilots. In this work, we propose a linear pilot-to-delay estimator (P2DE) that acquires the truncated delay CSI via sparse frequency pilots. We show the accuracy of the P2DE under frequency downsampling, and we demonstrate the P2DE’s efficacy when utilized with existing CSI estimation networks. Additionally, we propose to use trainable compressed sensing (CS) networks in a differential encoding network for time-varying CSI estimation, and we propose a new network, MarkovNet-ISTA-ENet (MN-IE), which combines a CS network for initial CSI estimation and multiple autoencoders to estimate the error terms. We demonstrate that MN-IE has better asymptotic performance than networks comprised of only one type of network.more » « less
-
Networks in many safety-critical systems like avionics, automotive, and industrial plants have strict end-to-end delay requirements to be met for correct system operation. Existing software-defined real-time networks do not support data plane programmability provided by recent protocol-independent switch architectures such as P4. Our research enables time-aware flow forwarding in P4-enabled software-defined time-critical networks. In this paper, we introduce time-aware flow scheduling for P4-enabled SDN architectures. We study two scheduling policies: the first one prioritizes flows based on slack (i.e., how much time is left to reach the destination), and the second one uses finish time as a priority metric, which is determined from its data rate requirements. Both approaches were implemented and tested in the P4 software stack. We find that the slack-based forwarding scheme performs better in retaining real-time requirements. Our publicly released scheduler implementations will assist network engineers in adapting programmable switches to safety-critical applications that demand precise timing guarantees.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/IWQOS52092.2021.9521335
