Cloud Computing 13 min read

How Huawei Cloud Achieves Ultra‑Low Latency for E‑Commerce Live Streaming

This article examines the technical challenges of massive e‑commerce live streams, explains why traditional streaming suffers from high latency, and details Huawei Cloud's low‑latency architecture, protocols, intelligent scheduling and ultra‑fast transcoding that enable sub‑second product link delivery to millions of viewers.

Huawei Cloud Developer Alliance
Huawei Cloud Developer Alliance
Huawei Cloud Developer Alliance
How Huawei Cloud Achieves Ultra‑Low Latency for E‑Commerce Live Streaming

Live streaming has become a major e‑commerce sales channel, with hundreds of millions of users online during events like 618. Delivering product links instantly to millions of viewers requires second‑level response and system stability.

#01 Traditional Live Streaming: Lag and Delay

Live streaming offers real‑time interaction, but latency remains a problem. The end‑to‑end live‑streaming chain includes capture, encoding, sending, distribution, receiving, decoding, and rendering. Network jitter during sending and distribution causes unpredictable delays.

Online classes: students ask questions, but the delay means the teacher has already moved on.

E‑commerce live streams: fans hear the host start a sale but cannot click the link in time.

Sports broadcasts: viewers notice a goal only after the crowd reaction.

Cross‑platform and cross‑region live streams require push (streaming from source to server) and pull (viewers retrieving the stream). The industry commonly uses RTMP for push and RTMP/HTTP‑FLV for pull, with latency around 3‑5 seconds; H5 often relies on HLS with >10 seconds delay.

Traditional live‑streaming architecture consists of three layers: a single‑line CDN edge node, a multi‑line CDN core node, and the origin server. The scheduler directs the host to the optimal edge node, which receives the stream, forwards it to the core, and finally to the origin; viewers then pull from the edge node.

#02 The Hard‑Core Tech Behind “3,2,1 Link”: Massive Ultra‑Low Latency

Traditional streaming cannot meet the high‑concurrency, low‑latency demands of interactive scenarios. Low‑latency live streaming is emerging as a new focus.

Huawei Cloud combines CDN protocol optimization, dynamic link selection, and ultra‑low‑latency transcoding to reduce end‑to‑end latency from 3‑5 seconds to under 800 ms, with transcoding delay kept below 150 ms, while remaining fully compatible with existing architectures.

#03 Architecture Analysis

Beyond protocol selection, architectural refinements further cut latency and improve user experience.

The design maintains forward compatibility with traditional RTMP push streams. Media transcoding and callback notifications on the origin server are seamlessly supported. Additionally, the H5 side adds RTC protocol playback, and iOS/Android support private RTC playback.

Huawei Cloud’s ultra‑low‑latency streaming optimization focuses on three core areas:

Replacing the TCP‑based last‑mile with a UDP‑based solution, using Huawei Cloud algorithms to achieve millisecond‑level latency while preserving robustness.

Transforming the static tree‑shaped scheduling architecture into an intelligent dynamic mesh, enabling CDN internal routing to be dynamically planned.

Introducing ultra‑low‑latency transcoding, reducing typical 500 ms transcoding delay to a stable <150 ms.

#04 Intelligent Dynamic Mesh Architecture

The “intelligent dynamic mesh” routes viewers more efficiently. For example, a viewer in Shenzhen can be directly scheduled to the same node used by the host, shortening the path to a single hop. A viewer in Guangxi is routed to the nearest node before reaching the host’s node, while a distant Beijing viewer may still follow the traditional path to ensure quality.

#05 Smart Scheduling Media Brain

The scheduling system, built on the origin server, consists of four core modules:

Content Management Center – monitors real‑time stream information such as host online/offline status.

Quality Map – gathers network quality data from edge agents, aggregates it in a big‑data platform, and produces a real‑time coverage quality map.

Scheduling Controller – holds baseline data like node traffic and user metrics.

Scheduling Decision Center – combines inputs from the previous modules to generate global scheduling strategies for user access and intra‑node back‑haul, deploying decisions to edge executors to minimize latency.

#05 Ultra‑Low‑Latency Transcoding

Video transcoding is essential for low‑latency streaming. Huawei’s transcoding keeps delay under 150 ms, reduces bitrate by over 30 % at the same visual quality, and supports high‑definition low‑bit-rate streams, quality enhancement, and ROI optimization.

With 5G rollout, the “live +” model expands into education, e‑commerce, sports, and show‑type streams. 2021 is expected to see 20 % of internet live streams upgraded to ultra‑low latency, and in the next 2‑3 years this technology will replace traditional streaming, driving new commercial models.

#06 Conclusion

Technology drives e‑commerce competition. Cloud computing, AI, and big data reshape shopping, while AR/VR, 5G, IoT, edge computing, machine vision, and autonomous logistics automate warehouses and deliveries, creating a more efficient, immersive consumer experience.

Original Source

Signed-in readers can open the original source through BestHub's protected redirect.

Sign in to view source
Republication Notice

This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactadmin@besthub.devand we will review it promptly.

e‑commercecloud computinglive streamingCDNLow latencyVideo Transcoding
Huawei Cloud Developer Alliance
Written by

Huawei Cloud Developer Alliance

The Huawei Cloud Developer Alliance creates a tech sharing platform for developers and partners, gathering Huawei Cloud product knowledge, event updates, expert talks, and more. Together we continuously innovate to build the cloud foundation of an intelligent world.

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.