A hidden authentication step causes Elasticsearch clients to send each request twice—once without credentials and again after a 401 response—doubling bandwidth usage, but configuring pre‑emptive authentication in Java or Python eliminates the waste and cuts traffic costs.
This article provides a comprehensive guide to Linux network performance tuning, covering key metrics, TCP window adjustments, Fast Open, congestion control algorithms, kernel parameter optimizations, zero‑copy transmission, NIC bonding, connection limits, and essential monitoring tools to achieve higher bandwidth and lower latency.
A high‑concurrency data‑collection service quickly filled a 100 Mbps uplink because each GET request incurred a 1.68 KB TLS handshake, but switching to plain HTTP or enabling Keep‑Alive can dramatically reduce per‑request size and server load, saving up to 70% bandwidth.
The team built a reusable low‑latency live‑streaming platform for the European Cup that replaces HLS with RTMP over QUIC, reuses VOD bandwidth, automates stream control, and cuts playback latency by 20 seconds while dramatically lowering bandwidth costs and delivering stable 1080p 50 fps HDR video.
Bilibili’s VOD player cuts bandwidth waste and costs by up to 15% through an intelligent caching and pre‑loading system that evaluates network risk and user habits to size cache dynamically, reduces stalls by 20% in vertical mode, and is continuously refined via a multi‑layer monitoring framework.
This article analyzes the fundamental goals of video transcoding, presents an information‑theoretic framework for bitrate reduction, compares traditional and deep‑learning codecs, and shares Bilibili's practical system design, parameter‑decision strategies, and visual‑quality‑aware optimizations that dramatically lower bandwidth consumption.
The 全民K歌 app reduced video bandwidth by 59% and improved playback experience by adopting Tencent's TXAV1 encoder, achieving 60% AV1 playback share, 20% better compression than H.265, and extensive mobile device support through careful transcoding strategies and performance testing.
The article examines how Apollo’s long‑polling and push‑pull design cause massive database bandwidth spikes during large namespace updates in promotion periods, analyzes the RemoteConfigLongPollService and repository code, and proposes three optimizations—key‑level synchronization, MD5‑based change detection, and circuit‑breaker handling—while weighing real‑time versus periodic update trade‑offs.
At LiveVideoStackCon 2023, XiaoHongShu unveiled an edge‑cloud super‑resolution framework guided by the human‑perception‑aligned RedVQA model, which jointly optimizes video quality and bandwidth by generating a dedicated SR bitrate tier in the cloud and applying a lightweight SR algorithm on the client, achieving notable QoE gains and narrow‑band HD delivery.
This article examines the technical background, core AI‑assisted perceptual encoding methods, practical implementations, and performance results of Baidu's intelligent video cloud, showing how content‑aware preprocessing, ROI‑based bitrate allocation, and AI‑enhanced super‑resolution can dramatically reduce bandwidth consumption while improving user experience.
This article explains how AI‑assisted perceptual video encoding techniques, such as ROI‑based bitrate allocation and content‑aware preprocessing, dramatically reduce bitrate while improving user experience, and describes Baidu Cloud’s "Smart Perceptual Ultra‑Clear" solutions, practical deployments, and future trends in intelligent video codecs.
Tencent Video slashed static image bandwidth costs by over 60% using AVIF compression and on‑the‑fly image sampling, which together cut average image size by more than half, reduced memory, latency and decode time, and saved tens of millions of yuan across billions of daily requests.
This article details the challenges and solutions for building a scalable, low‑latency danmaku service for live streaming, covering background requirements, bandwidth constraints, protocol choices, short‑polling implementation, reliability measures, and performance results that supported 700,000 concurrent users during a major event.
By leveraging the Yugong Platform’s predictive control—combining Prophet‑based threshold forecasts, custom real‑time bandwidth models, and a token‑bucket mechanism—to smooth peaks and fill valleys, enterprises can dramatically improve CDN bandwidth utilization, automate adjustments, and substantially lower peak‑based billing costs.
This article details Bilibili's multi‑datacenter offline architecture, explaining the capacity challenges, the chosen scale‑out design, and the implementation of job placement, data replication, routing, versioning, throttling, and traffic analysis to efficiently handle massive batch workloads across geographically distributed clusters.
This article describes the challenges of delivering real‑time danmaku for a Southeast Asian live‑streaming service, analyzes bandwidth pressure, network instability, and reliability issues, and presents a series of backend optimizations—including HTTP compression, response simplification, short‑polling, service splitting, caching, and a custom ring‑buffer—to reliably support up to one million concurrent users.
This article presents a detailed case study of designing a high‑throughput danmu (bullet‑screen) system for Southeast Asian live streaming, analyzing bandwidth constraints, latency issues, and evaluating long‑polling, WebSocket, and short‑polling approaches, ultimately describing the chosen short‑polling solution, service partitioning, caching, and ring‑buffer techniques to achieve stable performance for millions of concurrent users.
In 2015‑2016, Twitter engineer Steve Krenzel reduced mobile bandwidth consumption by about 40% through gzip‑compressed request logging, only to confront an ethical clash when a telecom client demanded identifiable user location data, leading him to reject the request and eventually leave the company.
This article explains how the veImageX platform leverages a self‑developed HEIF codec and AI‑driven optimization to shrink image sizes by over 30%, reduce bandwidth costs, maintain visual quality, and provide flexible integration options for both native apps and web pages.
A visual‑lossless deep‑learning pre‑processor that employs a DCT‑based low‑rank loss and an ultra‑lightweight CPU‑friendly model achieves up to 20% bitrate reduction for 1080p videos while preserving high‑frequency details, enabling real‑time transcoding and bandwidth savings for popular content on Bilibili.
This article describes Bilibili's offline multi‑datacenter architecture, explaining why a scale‑out approach was chosen over scale‑up, and detailing the unit‑based design, job placement, data replication, routing, versioning, bandwidth throttling, traffic analysis, and the operational results and future directions.
This article details Bilibili's multi‑datacenter solution for offline big‑data workloads, covering the challenges of capacity limits, the design of a unit‑based architecture, job placement, data replication, routing, versioning, bandwidth throttling, traffic analysis, and future directions.
To overcome rapid data growth and on‑premise capacity limits, Bilibili adopted a scale‑out, unit‑based multi‑datacenter architecture that isolates failures, intelligently places jobs, replicates data via an enhanced DistCp service, routes reads with an IP‑aware HDFS router, and throttles cross‑site traffic, enabling stable offline big‑data processing of hundreds of petabytes while preserving throughput.
Enterprise A consolidated its numerous VPC peering links and dedicated‑line gateways into a single Tencent Cloud Connect Network instance, enabling automatic route learning, ECMP load‑balancing across four 10 Gbps lines, boosting bandwidth utilization from about 50 % to 75 % while cutting operational complexity and enhancing cross‑region high availability.
Video encoding standards—from H.264 to AV1 and China’s AVS series—have successively doubled compression efficiency, lowering bandwidth and storage costs, while patent royalties steer industry adoption, prompting Chinese platforms like iQIYI to develop proprietary, high‑speed encoders and prepare for next‑gen codecs such as AVS3 and H.266.
Alibaba’s entertainment engineers explain how their cloud‑edge P2P‑based PCDN architecture tackles the massive bandwidth demands of video streaming by leveraging multi‑tier node hierarchies, intelligent resource allocation, and adaptive download scheduling to deliver smooth HD playback while continuously reducing average bandwidth costs.
Alibaba Entertainment’s 2019 Double‑11 Cat Night live stream combined FPGA‑based H.265 transcoding, smart‑profile QoE adaptation, Dolby Atmos audio and frame‑aligned multi‑view video to deliver ultra‑HD 4K60 streams at 35% lower bandwidth cost, achieving zero failures, zero degradations and over 99.99% service availability.
At the 2019 Cloudwise Entertainment Technology Forum, Youku unveiled Narrowband HD 3.0, a private video transmission protocol that reduces bandwidth by roughly 30% compared with its 2.0 predecessor, integrates cloud‑based AI rendering signals, and makes ultra‑high‑definition streaming feasible for ordinary mobile users and low‑cost devices in the upcoming 5G era.
This article shares Tencent SNG's practical experience in bandwidth cost optimization, detailing how advanced image and video compression techniques, adaptive resolution, AI‑driven super‑resolution, and efficient transcoding pipelines reduced over a billion yuan in cash flow while preserving user experience and product quality.
The article reviews recent developments in image compression formats such as JPEG, WebP, HEVC, and Tencent's proprietary WXAM/SHARP, explains Guetzli's perceptual encoding, details extensive GPU‑based performance optimizations, and demonstrates how these techniques dramatically reduce bandwidth usage in Tencent's massive image storage platform.
Tencent's video team shares how they grew daily QQ Space video plays from 50 million to over a billion by redesigning the mobile playback architecture, implementing bandwidth and cost controls, accelerating start‑up times, adopting H.265, and building robust monitoring and error‑handling pipelines.
Tencent’s social services cut two hundred million RMB in bandwidth costs by flattening traffic peaks through seven practices—disabling auto‑play, pre‑pushing hot content, compressing media, offering on‑demand quality, using segmented downloads, applying technical breakthroughs like P2P and codec optimization, and filtering illegal content—while preserving user experience.
This article shares Tencent's seven practical bandwidth‑saving techniques—ranging from disabling auto‑play to intelligent pre‑push, file compression, on‑demand usage, segmented download, technical breakthroughs, and content compliance—to dramatically reduce operational costs while maintaining user experience.
This article examines the characteristics and billing models of multi‑data‑center networks, analyzes external traffic patterns, identifies challenges in optimizing Internet‑facing bandwidth, and proposes practical scheduling strategies to better utilize idle bandwidth and reduce carrier costs.
