Author

Bilibili Tech

Provides introductions and tutorials on Bilibili-related technologies.

403

Articles

Likes

3.0k

Views

Comments

Latest from Bilibili Tech

100 recent articles max

Bilibili Tech

Jan 24, 2025 · Operations

Design and Implementation of a CDN Edge‑Node Scheduling System for Bilibili Live Streaming

The paper presents Bilibili’s multi‑layer CDN edge‑node scheduling system, which groups heterogeneous nodes by quality and price, uses cost‑aware and resource‑aware heuristics—including maximum‑flow regional borrowing and contextual‑bandit utilization prediction—to allocate bandwidth per business, achieving a 43 % bandwidth reuse increase, 33 % coverage boost, and markedly lower stall rates.

BilibiliCDNcost optimization

0 likes · 10 min read

Design and Implementation of a CDN Edge‑Node Scheduling System for Bilibili Live Streaming

Bilibili Tech

Jan 21, 2025 · Artificial Intelligence

Accelerating Large Model Inference: Challenges and Multi‑Level Optimization Strategies

The article outlines how exploding LLM sizes create compute, memory, and latency bottlenecks and proposes a full‑stack solution—operator fusion, high‑performance libraries, quantization, speculative decoding, sharding, contiguous batching, PageAttention, and specialized frameworks like MindIE‑LLM—to dramatically boost inference throughput and reduce latency, while highlighting future ultra‑low‑bit and heterogeneous hardware directions.

Inference AccelerationMulti-modalcontinuous batching

0 likes · 21 min read

Accelerating Large Model Inference: Challenges and Multi‑Level Optimization Strategies

Bilibili Tech

Jan 17, 2025 · Backend Development

NeighborHash: An Enhanced Batch Query Architecture for Real‑time Recommendation Systems

NeighborHash is a distributed batch‑query architecture for real‑time recommendation systems that combines a cache‑line‑optimized hash table—featuring Lodger Relocation, bidirectional cache‑aware probing, and inline‑chaining—with an NVMe‑backed key‑value service, versioned updates, and asynchronous memory‑access chaining to achieve sub‑microsecond, high‑throughput top‑N retrieval.

AMACNVMebatch query

0 likes · 20 min read

NeighborHash: An Enhanced Batch Query Architecture for Real‑time Recommendation Systems

Bilibili Tech

Jan 14, 2025 · Artificial Intelligence

Technical Practices and Productization of Intelligent Advertising Title Generation for Bilibili

We built an LLM‑powered system for Bilibili that automatically creates ad titles from user keywords, employing fluency, style, and quality classifiers, mixed domain data cleaning, and alignment methods such as SFT, DPO and KTO, resulting in a product that now generates about ten percent of daily titles and drives significant ad spend.

AI alignmentAd Title GenerationBilibili

0 likes · 24 min read

Technical Practices and Productization of Intelligent Advertising Title Generation for Bilibili

Bilibili Tech

Jan 10, 2025 · Frontend Development

Design and Evolution of Bilibili's Activity Platform Low‑Code System

Bilibili’s Activity Platform low‑code system was completely redesigned over two years, introducing a modular, decoupled architecture with an iframe‑based canvas, unified renderer, and Eva‑CLI tools that cut activity build time from three days to 0.48 days, boosted coverage to 77% of activities, and reduced first‑contentful‑paint by 35%.

BilibiliFrontend Architecturelow-code

0 likes · 31 min read

Design and Evolution of Bilibili's Activity Platform Low‑Code System

Bilibili Tech

Jan 7, 2025 · Cloud Native

Design and Implementation of Bilibili's Large-Scale Recall System

Bilibili’s large‑scale recall system separates online processing into a two‑tier merge service and an index service, supports multi‑channel text, item‑to‑item and vector indexes with real‑time updates, uses horizontal sharding, robust CI/CD, monitoring and degradation mechanisms, and is being extended toward model‑based recall and greater automation.

BilibiliSearch ArchitectureVector Search

0 likes · 16 min read

Design and Implementation of Bilibili's Large-Scale Recall System

Bilibili Tech

Jan 3, 2025 · Big Data

Evolution and Production Practices of Apache Celeborn Remote Shuffle Service at Bilibili

Bilibili replaced Spark’s unstable External Shuffle Service with a push‑based approach, then deployed Apache Celeborn’s remote shuffle on Kubernetes using HA masters, tiered workers, extensive monitoring, history‑based routing, chaos testing, and seamless Spark, Flink, and MapReduce integration, while planning self‑healing, elastic scaling, and priority‑aware I/O enhancements.

Apache CelebornBig DataFlink

0 likes · 28 min read

Evolution and Production Practices of Apache Celeborn Remote Shuffle Service at Bilibili

Bilibili Tech

Dec 31, 2024 · Cloud Computing

Design and Implementation of Bilibili AI Compute Network: Topology, Hardware Selection, Load Balancing, and Monitoring

Bilibili designed and deployed an AI compute network for large language model training, choosing a Fat-Tree topology, selecting high‑speed switches, optical modules, and fibers, implementing fixed‑path load balancing, and building a sub‑second telemetry monitoring platform, with plans to scale to ten‑thousand GPUs.

AI compute networkFat-Tree topologyhardware selection

0 likes · 17 min read

Design and Implementation of Bilibili AI Compute Network: Topology, Hardware Selection, Load Balancing, and Monitoring

Bilibili Tech

Dec 27, 2024 · Big Data

Consistency Architecture for Bilibili Recommendation Model Data Flow

The article outlines Bilibili’s revamped recommendation data‑flow architecture that eliminates timing and calculation inconsistencies by snapshotting online features, unifying feature computation in a single C++ library accessed via JNI, and orchestrating label‑join and sample extraction through near‑line Kafka/Flink pipelines, with further performance gains and Iceberg‑based future extensions.

Data ConsistencyFlinkIceberg

0 likes · 12 min read

Consistency Architecture for Bilibili Recommendation Model Data Flow

Bilibili Tech

Dec 24, 2024 · Artificial Intelligence

AniSora: An Integrated System for Anime Video Generation with Data Flywheel, Controllable Diffusion Models, and Evaluation Benchmark

AniSora combines a 10‑million‑pair anime text‑video dataset, a controllable diffusion‑transformer with temporal‑mask conditioning for text‑to‑video, interpolation and region‑guided animation, and a 948‑video benchmark, delivering industry‑leading character and motion consistency and already powering low‑cost dynamic‑comic production for multiple IPs.

AI animationDataset BenchmarkTemporal Masking

0 likes · 21 min read

AniSora: An Integrated System for Anime Video Generation with Data Flywheel, Controllable Diffusion Models, and Evaluation Benchmark