Wanxin Shi

The increasing popularity of large models and datasets has highlighted the significance of distributed training networks. As gradient synchronization generates substantial traffic, in-network aggregation (INA) has emerged as a solution to offload aggregation onto the switch, alleviating network congestion and accelerating distributed training. However, the limited memory capacity of the INA switch becomes a potential bottleneck as computation shifts into the network, especially in multi-tenant scenarios. To address this bottleneck and enhance network throughput, we propose the Aggregation with Innetwork Resource Pooling (AIRP) framework. Unlike existing approaches that optimize individual switches in a localized manner, AIRP takes a holistic view and efficiently pools switch memory resources across the entire network, allocating them to multiple tenants. Evaluation using the ns-3 simulator and P4 testbed demonstrates that AIRP can accelerate the training of various models, including computer vision and language models. The experimental results show that AIRP outperforms existing INA approaches by up to 7 times in terms of network throughput in multi-tenant scenarios, while also achieving great flexibility and efficiency in deployment.

The increasing volume of video traffic presents significant challenges to network transmission, while edge computing accelerates video delivery by leveraging caching and computation to optimize content forwarding. However, as edge computing is generally deployed by service providers in a transparent manner, clients cannot perceive edge states, e.g., cache availability, potentially resulting in suboptimal bitrate decisions. This issue persists even with intelligent bitrate selection approaches on the client side, as the inaccurate estimation of network delivery capacity due to edge cache transparency remains unresolved. Meanwhile, single-edge servers or nodes, with limited cache space and computational capacity for a small number of users, can be more effective by aggregating into clusters to better serve users and optimize resource utilization. Therefore, we propose an edge-assisted bitrate re-adaptation scheme (e-BitRead) for adaptive streaming, utilizing neighbor edges to accelerate video deliveries. e-BitRead introduces three key innovations: (i) it employs a bitrate re-adaptation mechanism that intelligently selects alternative bitrates from edge servers instead of strictly responding with the requested bitrate, (ii) it utilizes collaborative caching across multiple edge servers to expand available bitrate options through coordinated resource sharing, and (iii) it enhances the learning efficiency through joint optimization of network architecture and reward design, which leverages actor-critic structure to fit into multi-edge bitrate adaptation. In experiments with an intelligent client ABR, e-BitRead demonstrates its superiority by achieving a 1.43x higher hit ratio compared to the baseline, while improving QoE by 1.93x over the scheme without smart bitrate matching and delivering a 33% gain over the single-edge re-adaptation approach.

Programmable packet scheduling enables the integration of scheduling algorithms into switches without the need for hardware redesign. The Push-In First-Out (PIFO) queue facilitates a programmable packet scheduler, supporting a single scheduling algorithm flexibly. However, hierarchical scheduling required in Multi-Tenant Data Centers (MTDCs) remains non-programmable. Dynamic and diverse hierarchical scheduling algorithms necessitate alterations in both the number of PIFO queues and their connection topology, posing a significant challenge to support them on fixed hardware. In this paper, we introduce the virtualized PIFO (vPIFO) system, a hardware virtualization solution for programmable hierarchical packet scheduling in MTDCs. The vPIFO system, leveraging a single physical PIFO, can flexibly establish PIFO Trees with diverse shapes, thereby enabling network operators to customize fine-grained traffic scheduling. We present a high-performance and large-scale hardware design for the vPIFO and implement a prototype on FPGA and ASIC. The synthesized results on GlobalFoundries 28nm technology showcase that vPIFO can flexibly support hierarchical scheduling within the scale of 128 PIFO instances while achieving an impressive speed of 400 Gbps with 6 levels of hierarchical scheduling. To the best of our knowledge, vPIFO is the first hardware virtualization work for packet schedulers to support accurate programmable hierarchical scheduling.

The digital age has brought a significant increase in video traffic. This traffic growth, driven by rapid internet advancements and a surge in multimedia applications, presents both challenges and opportunities to video transmissions. Users seek high-quality video content, prompting service providers to offer high-definition options to improve user experience and increase profits. However, traditional end-to-end best-effort networks struggle to meet the demands of extensive video streaming and ensure good user Quality of Experience (QoE), especially in high user mobility scenarios or fluctuating network conditions. Addressing some of these challenges, content delivery networks (CDN) are instrumental in delivering video content, but they are under increased pressure to support high quality and reduce their deployment and maintenance costs. Currently, cloud-edge-end fusion technologies have become one of the optimization directions for network services due to their flexibility and scalability. At the same time, in the context of the recent advancements in computing-focused network paradigms, intelligent enhancement techniques (e.g., super-resolution), commonly utilized in image optimization, have been adopted as a pivotal solution for increasing video delivery quality. To illustrate the essence and employment of the intelligent enhancement solutions for video streaming, this paper first outlines the video streaming process, discusses relevant evaluation metrics, and examines aspects related to the intelligent solutions. Then the paper presents the intelligent enhancement process of video streaming, analyzes various typical intelligent models for content enhancement and highlights their distinct characteristics. This exploration delves deeper into various intelligent quality-improved solutions, scrutinizing their applicability across different transmission scenarios like Video on Demand (VoD) and live streaming, and shedding light on their strengths and weaknesses from a cloud-edge-end fusion perspective. Additionally, the intelligent quality-enhanced video delivery systems are analysed comprehensively, exploring their impact on network traffic, computational demand, and storage needs, and aligning them with potential deployment scenarios and use cases. Finally, the article identifies open issues and key challenges that warrant attention in future research endeavors.

With the rapid popularity of short video applications, a large number of short video transmissions occupy the bandwidth, placing a heavy load on the Internet. Due to the extensive number of short videos and the predominant service for mobile users, traditional approaches (e.g., CDN delivery, edge caching) struggle to achieve the expected performance, leading to a significant number of redundant transmissions. In order to reduce the amount of traffic, we design a Novel Coded Transmission Mechanism (NCTM), which transmits XOR-coded data instead of the original video content. NCTM caches the short videos that users have already watched in user devices, and encodes, multicasts, and decodes XOR-coded files separately at the server, edge nodes, and clients, with the assistance of cached content. This approach enables NCTM to deliver more short video data given the limited bandwidth. Our extensive trace-driven simulations show how NCTM reduces network load by 3.02%-14.75%, cuts peak traffic by 23.01%, and decreases rebuffering events by 43%-85% in comparison to a CDN-supported scheme and a naive edge caching scheme. Additionally, NCTM also increases the user's buffered video duration by 1.21x-13.53x, ensuring improved playback smoothness.

Live video streaming has grown dramatically in recent years. A key challenge is achieving high video quality of experience (QoE) in low-rate networks. To tackle this problem, recent streaming approaches strategically drop video frames, thus reducing the bandwidth required. However, these methods are usually designed for video on demand (VoD) services and perform poorly in live video streaming. In this paper, we design a new live video streaming approach, Reparo, which aims to improve users' QoE in low-rate networks. On the upload client side, Reparo discards video frames such that they are never encoded or transmitted. To decide which frames should be dropped, we design a real-time Video Frame Discarding (VFD) model, which strives to minimize the impact on video quality while maximizing bandwidth savings. To complement this, Reparo further proposes a modified adaptive bitrate algorithm and two encoding modes, targeting low-frame-rate encoding. On the server side, Reparo then recovers the dropped frames using a lightweight Video Frame Interpolation Deep Neural Network (VFI-DNN). Experimental results show that, compared with vanilla DASH, Reparo reaches an SSIM gain of 0.018, or reduces bandwidth consumption by 30.86%. With an average bandwidth of 0.974Mbps, it improves QoE by 18.13% on average compared to DASH.

The real-time video streaming market is currently undergoing rapid development. Various existing methods aim to optimize network bandwidth utilization by adjusting the sending rate and implementing forward error correction techniques. Meanwhile, other methods enhance the quality of video streams using frame-level super-resolution (SR), leveraging the computational power on the receiving end, which often consumes a lot of time and increases the end-to-end delay of a real-time video stream. To this end, we proposed a real-time video stream enhancement scheme based on patch-level super-resolution. The approach adaptively selects the locations of super-resolution patches based on the time interval between video frame receiving and rendering. Additionally, the quality benefit of SR patches continuously would be transferred to subsequent video frames during the decoding and upscaling process. We evaluate the performance of our scheme on several commercial mobile devices with real network traces and construct an open-source real-time video streaming system. Overall, our scheme can improve the PSNR of real-time video streaming by 2.3dB.

The user experience of mobile web video streaming is often impacted by insufficient and dynamic network bandwidth. In this paper, we design Bidirectionally Optimized Super-Resolution (BiSR) to improve the quality of experience (QoE) for mobile web users under limited bandwidth. BiSR exploits a deep neural network (DNN)-based model to super-resolve key frames efficiently without changing the inter-frame spatial-temporal information. We then propose a downscaling DNN and a mobile-specific optimized lightweight super-resolution DNN to enhance the performance. Finally, a novel reinforcement learning-based adaptive bitrate (ABR) algorithm is proposed to verify the performance of BiSR on real network traces. Our evaluation, using a full system implementation, shows that BiSR saves 26% of bitrate compared to the traditional H.264 codec and improves the SSIM of video by 3.7% compared to the prior state-of-the-art. Overall, BiSR enhances the user-perceived quality of experience by up to 30.6%.

Datacenters currently deploy shallow-buffered switches to achieve low latency by avoiding long waiting time in the data plane. However, the limited buffer space in the switch causes the frequent overflow and the notorious TCP incast problem. Moreover, the simple scheduling strategy in buffer deprives the switch of the ability to offer deeply differentiated services. Therefore, we present a novel priority-based flow-aware in-network caching scheme, named Poche, which supplies more control capabilities for the network side through introducing some additional cache resource into switches. Poche classifies network traffic into multiple priorities according to the latency requirements of flows. The end server adds priority tags to packets and sets different RTO values for flows with distinct priorities. The switch monitors the buffer utilization of each port and performs the priority-based flow-aware caching and injecting strategies based on the analysis of the scheduling model between the buffer and cache. We conduct comprehensive experiments to compare Poche with the state-of-the-art traffic optimization schemes. The results demonstrate that Poche can reduce the FCTs of latency-sensitive flows by at least 59.1% and improve the network throughput by at least 54.4%, while ensuring the finite cached volume and effectively addressing the incast problem.