Identifying the "heavy hitter" ows or ows with large tra c volumes in the data plane is important for several applications e.g., ow-size aware routing, DoS detection, and tra c engineering. However, measurement in the data plane is constrained by the need for linerate processing (at 10-100Gb/s) and limited memory in switching hardware. We propose HashPipe, a heavy hitter detection algorithm using emerging programmable data planes. HashPipe implements a pipeline of hash tables which retain counters for heavy ows in the tables while evicting lighter ows over time. We prototype HashPipe in P4 and evaluate it with packet traces from an ISP backbone link and a data center. On the ISP trace, we nd that HashPipe identi es 95% of the 300 heaviest ows with less than 80KB of memory on a trace that contains 400,000 ows.

With the growing usage of Bitcoin and other cryptocurrencies, many scalability challenges have emerged. A promising scaling solution, exemplified by the Lightning Network, uses a network of bidirectional payment channels that allows fast transactions between two parties. However, routing payments on these networks efficiently is non-trivial, since payments require finding paths with sufficient funds, and channels can become unidirectional over time blocking further transactions through them. Today's payment channel networks exacerbate these problems by attempting to deliver all payments atomically. In this paper, we present the Spider network, a new packet-switched architecture for payment channel networks. Spider splits payments into transaction units and transmits them over time across different paths. Spider uses congestion control, payment scheduling, and imbalance-aware routing to optimize delivery of payments. Our results show that Spider improves the volume and number of successful payments on the network by 10-45% and 5-40% respectively compared to state-of-the-art approaches.

No abstract

No abstract

Video compression is a critical component of Internet video delivery. Recent work has shown that deep learning techniques can rival or outperform human-designed algorithms, but these methods are significantly less compute and power-efficient than existing codecs. This paper presents a new approach that augments existing codecs with a small, content-adaptive super-resolution model that significantly boosts video quality. Our method, SRVC, encodes video into two bitstreams: (i) a content stream, produced by compressing downsampled low-resolution video with the existing codec, (ii) a model stream, which encodes periodic updates to a lightweight super-resolution neural network customized for short segments of the video. SRVC decodes the video by passing the decompressed low-resolution video frames through the (time-varying) super-resolution model to reconstruct high-resolution video frames. Our results show that to achieve the same PSNR, SRVC requires 16% of the bits-per-pixel of H.265 in slow mode, and 2% of the bits-per-pixel of DVC, a recent deep learning-based video compression scheme. SRVC runs at 90 frames per second on a NVIDIA V100 GPU.