BestHub
Sign in
Tagged articles
28 articles
Page 1 of 1
DataFunTalk
DataFunTalk
Feb 2, 2026 · Artificial Intelligence

How Alluxio Boosts GPU Utilization to 99.57% for Embodied AI – Inside the MLPerf Success

This article explains how Alluxio’s distributed caching architecture tackles the massive, multimodal data challenges of embodied AI, delivers near‑zero‑millisecond access, achieves 99.57% GPU utilization in MLPerf Storage v2.0, and validates its value through real‑world enterprise deployments.

AI Data PlatformAlluxioEmbodied Intelligence
0 likes · 21 min read
How Alluxio Boosts GPU Utilization to 99.57% for Embodied AI – Inside the MLPerf Success
DataFunTalk
DataFunTalk
Sep 3, 2025 · Artificial Intelligence

How Alluxio’s Distributed Cache Boosts AI Training to 99.57% GPU Utilization

Alluxio’s distributed caching dramatically accelerates AI training and checkpointing workloads, achieving up to 99.57% GPU utilization and linear scaling across clusters in the MLPerf Storage v2.0 benchmark, while using cost‑effective commodity hardware to eliminate I/O bottlenecks.

AI trainingAlluxioGPU utilization
0 likes · 11 min read
How Alluxio’s Distributed Cache Boosts AI Training to 99.57% GPU Utilization
Volcano Engine Developer Services
Volcano Engine Developer Services
Jul 17, 2025 · Artificial Intelligence

How Distributed KVCache (EIC) Revolutionizes Large‑Model Inference Performance

This article examines how Volcano Engine's Elastic Instant Cache (EIC) tackles the memory bottleneck, high‑concurrency latency, and cross‑node coordination challenges of large language model inference by decoupling storage and computation, pooling resources, and applying layered optimizations, ultimately boosting AI inference efficiency, scalability, and cost‑effectiveness across various deployment scenarios.

AI InfrastructureKVCacheLLM inference
0 likes · 30 min read
How Distributed KVCache (EIC) Revolutionizes Large‑Model Inference Performance
Su San Talks Tech
Su San Talks Tech
Sep 30, 2024 · Backend Development

How JD’s Hotkey Framework Detects and Pushes Hot Data in Milliseconds

JD’s Hotkey framework provides millisecond‑level detection and cluster‑wide push of hot data, users, and interfaces, dramatically reducing backend query load, improving performance, and supporting scenarios such as local caching and rate limiting, with proven scalability demonstrated in large‑scale e‑commerce promotions.

Backend PerformanceJavadistributed caching
0 likes · 7 min read
How JD’s Hotkey Framework Detects and Pushes Hot Data in Milliseconds
Lobster Programming
Lobster Programming
Jun 11, 2024 · Fundamentals

Why Consistent Hashing Matters: Solving Cache Distribution and Scaling Issues

Consistent hashing replaces simple modulo‑based distribution to efficiently locate cached data across changing numbers of servers, using a hash ring and virtual nodes to ensure balanced load, minimize data movement, and improve reliability in distributed caching, load balancing, and database sharding scenarios.

consistent hashingdistributed cachingsharding
0 likes · 6 min read
Why Consistent Hashing Matters: Solving Cache Distribution and Scaling Issues
Bilibili Tech
Bilibili Tech
Jan 6, 2023 · Backend Development

Hotspot Detection and Local Cache Framework for High‑Traffic Applications

The presented hotspot detection and local‑cache framework leverages the HeavyKeeper streaming top‑k algorithm with decay‑based burst detection, integrates zero‑code SDK support and a whitelist‑enabled LRU cache, enabling a few megabytes of memory to achieve up to 85% hit rates and dramatically reduce Redis load in high‑traffic applications.

distributed cachingheavykeeperhotspot detection
0 likes · 21 min read
Hotspot Detection and Local Cache Framework for High‑Traffic Applications
JD Tech
JD Tech
Dec 1, 2022 · Databases

Understanding Redis Data Skew and Hotkey Detection with JD Open‑Source hotkey Solution

This article explains the concept of Redis data skew, its causes and impacts, explores data volume and access skew classifications, presents mitigation strategies, and provides a comprehensive source‑code walkthrough of JD's open‑source hotkey framework—including client, worker, and dashboard components—for detecting and handling hot keys in distributed cache clusters.

HotKeyPerformance Optimizationdistributed caching
0 likes · 54 min read
Understanding Redis Data Skew and Hotkey Detection with JD Open‑Source hotkey Solution
Tencent Cloud Developer
Tencent Cloud Developer
Feb 28, 2022 · Big Data

GooseFS: Distributed Caching System for Storage-Compute Separation Architecture

GooseFS, Tencent Cloud’s distributed caching system for storage‑compute separation, links compute frameworks to underlying storage (COS, CHDFS, COSN) and boosts big‑data and AI workloads by 2‑10× through transparent acceleration, robust master‑worker architecture, Raft‑based HA, tiered caching, and metadata optimizations, delivering up to 50% cost savings and 29% faster compute jobs.

Big Data ArchitectureGooseFSRaft consensus
0 likes · 18 min read
GooseFS: Distributed Caching System for Storage-Compute Separation Architecture
Su San Talks Tech
Su San Talks Tech
Feb 20, 2022 · Backend Development

How Consistent Hashing Solves Cache Scaling and Reduces Data Skew

This article explains the principles of consistent hashing, compares it with simple modulo hashing, and shows how the hash ring, virtual nodes, and key mapping improve load balancing, minimize data loss during node changes, and prevent cache avalanche in distributed caching systems.

consistent hashingdistributed cachingvirtual nodes
0 likes · 12 min read
How Consistent Hashing Solves Cache Scaling and Reduces Data Skew
vivo Internet Technology
vivo Internet Technology
Jan 12, 2022 · Backend Development

Vivo Push Notification Platform: Architecture Evolution and Engineering Practices

The article details Vivo’s push notification platform, describing its evolution from cloud‑based beginnings to a self‑built, three‑region architecture that supports over 1 million concurrent connections, billions of daily messages, and incorporates optimizations such as adaptive heartbeats, advanced load‑balancing, distributed caching, multi‑layer rate limiting, circuit‑breaker mechanisms, and comprehensive content security.

Push NotificationSystem ArchitectureTCP
0 likes · 16 min read
Vivo Push Notification Platform: Architecture Evolution and Engineering Practices
Programmer DD
Programmer DD
May 16, 2021 · Backend Development

What Really Drives Scalable Backend Architecture? Lessons from Weibo’s Massive System

This article explores the essence of system architecture, illustrating how large‑scale services like Uber and Weibo handle massive traffic, data storage, service decomposition, multi‑level caching, and monitoring, and offers practical design principles for building robust, high‑performance backend systems.

Backend ArchitecturePerformance OptimizationScalable Systems
0 likes · 22 min read
What Really Drives Scalable Backend Architecture? Lessons from Weibo’s Massive System
Laravel Tech Community
Laravel Tech Community
May 9, 2021 · Backend Development

Understanding Consistent Hashing: From Simple Modulo Hash to Optimizations

This article explains the drawbacks of a basic modulo hash algorithm for key distribution, demonstrates how consistent hashing resolves scaling and node‑failure issues, and discusses virtual‑node techniques to mitigate data skew and improve load balancing in distributed cache systems.

Data Skewconsistent hashingdistributed caching
0 likes · 5 min read
Understanding Consistent Hashing: From Simple Modulo Hash to Optimizations
dbaplus Community
dbaplus Community
Mar 12, 2018 · Cloud Native

How EVCache Uses Cloud‑Native Architecture for Scalable Distributed Caching

This article explains why distributed caching is essential for large‑scale internet applications, outlines the business and technical benefits of cloud services, and details EVCache’s cloud‑native design, cross‑region replication, high‑availability mechanisms, and real‑world use cases such as Netflix’s recommendation system.

AWSEVCacheNetflix
0 likes · 15 min read
How EVCache Uses Cloud‑Native Architecture for Scalable Distributed Caching
Vipshop Quality Engineering
Vipshop Quality Engineering
Jan 17, 2018 · Backend Development

Uncovering Hidden Cache Failures: A Robustness Test of Memcached with spymemcached

This article details a comprehensive robustness test of a core public service system's caching layer, exposing how decreasing Memcached (MC) instances dramatically impacts TPS and latency, analyzes the underlying Ketama consistent‑hash algorithm, and proposes concrete improvements to mitigate such failures.

Backend testingMemcachedcache robustness
0 likes · 11 min read
Uncovering Hidden Cache Failures: A Robustness Test of Memcached with spymemcached
MaGe Linux Operations
MaGe Linux Operations
Sep 8, 2016 · Backend Development

How Consistent Hashing Powers Distributed Memcached Caching

This article explains how memcached’s client‑side distribution works, illustrates the role of consistent hashing in assigning keys to servers, compares traditional modulo hashing with consistent hashing, discusses hash function choices, and provides complete Java and Python implementations for a scalable distributed cache.

BackendMemcachedconsistent hashing
0 likes · 12 min read
How Consistent Hashing Powers Distributed Memcached Caching
Art of Distributed System Architecture Design
Art of Distributed System Architecture Design
Jun 24, 2016 · Backend Development

Scalable Web Architecture for Startup Companies

This article explains how startup engineers can design and implement a scalable web architecture—covering server clustering, load balancing, distributed caching, database replication, and team organization—to handle rapid user growth without compromising performance or reliability.

Database Replicationdistributed cachingload balancing
0 likes · 15 min read
Scalable Web Architecture for Startup Companies
Art of Distributed System Architecture Design
Art of Distributed System Architecture Design
Jun 13, 2016 · Backend Development

How to Build a Scalable Web Architecture for Fast‑Growing Startups

This article explains how startup engineers can design and evolve a scalable web architecture—covering server partitioning, load balancing, distributed caching, database replication, and team organization—to handle rapid user growth without compromising performance or reliability.

Database Replicationdistributed cachingload balancing
0 likes · 16 min read
How to Build a Scalable Web Architecture for Fast‑Growing Startups
Qunar Tech Salon
Qunar Tech Salon
May 29, 2016 · Backend Development

Understanding Distributed Caching: Principles and Implementation of Memcached

This article explains the fundamentals of caching, the role of distributed caches like memcached in high‑concurrency environments, and details the algorithms—including remainder hashing and consistent hashing with virtual nodes—that enable memcached to distribute data across multiple servers efficiently.

Memcachedcache algorithmsconsistent hashing
0 likes · 10 min read
Understanding Distributed Caching: Principles and Implementation of Memcached
dbaplus Community
dbaplus Community
May 25, 2016 · Backend Development

How Adding a Distributed Memcache Layer Cut Database Load by 30% in a Telecom CRM

A telecom CRM project introduced a horizontally scalable Memcache cache layer to store frequently accessed dictionary and customer data, reducing database queries by 30%, improving response times by 15%, and lowering expansion costs while detailing the architecture, access logic changes, high‑availability measures, and real‑world results.

Backend ArchitectureScalabilitydistributed caching
0 likes · 13 min read
How Adding a Distributed Memcache Layer Cut Database Load by 30% in a Telecom CRM
Architecture Digest
Architecture Digest
May 12, 2016 · Backend Development

Understanding Distributed Caching with Memcached: Principles and Algorithms

This article explains the fundamentals of caching, the role of memcached in high‑concurrency environments, and details the distributed implementation methods such as remainder hashing and consistent hashing, including their advantages, drawbacks, and optimization techniques.

Memcachedcaching algorithmsconsistent hashing
0 likes · 11 min read
Understanding Distributed Caching with Memcached: Principles and Algorithms
21CTO
21CTO
May 9, 2016 · Backend Development

Unveiling Memcached’s Distributed Caching: Algorithms and Implementation

Under high‑concurrency loads, disk I/O becomes a bottleneck, prompting the use of caches; this article explains the fundamentals of caching, the role of memcached, and details its distributed implementation, covering simple modulo hashing, consistent hashing, and optimized virtual‑node techniques.

Backend DevelopmentMemcachedcache algorithms
0 likes · 9 min read
Unveiling Memcached’s Distributed Caching: Algorithms and Implementation
21CTO
21CTO
Apr 25, 2016 · Backend Development

How to Proactively Refresh Memcached Before Expiration and Avoid DB Thundering Herd

This article explores why Memcached expiration can cause sudden DB overload and presents five practical strategies—including periodic DB refresh, lock‑based queries, dual‑key schemes, and time‑embedded values—to proactively update caches and keep backend performance stable.

BackendJavaMemcached
0 likes · 8 min read
How to Proactively Refresh Memcached Before Expiration and Avoid DB Thundering Herd
Art of Distributed System Architecture Design
Art of Distributed System Architecture Design
Apr 12, 2016 · Cloud Native

How Netflix’s EVCache Enables Global Low‑Latency Caching at Massive Scale

The article explains Netflix’s EVCache—a cloud‑native, memcached‑based distributed cache that provides low‑latency, high‑reliability data access across multiple regions using asynchronous Kafka‑driven replication, detailing its architecture, performance optimizations, and remaining challenges.

Cloud NativeEVCacheKafka replication
0 likes · 10 min read
How Netflix’s EVCache Enables Global Low‑Latency Caching at Massive Scale
Art of Distributed System Architecture Design
Art of Distributed System Architecture Design
Mar 11, 2015 · Cloud Computing

Building a Scalable Cloud Application Platform for the 12306 Railway Ticketing System with Pivotal Gemfire

The article analyzes the rapid growth of China’s 12306 online ticketing platform, the challenges of extreme traffic and concurrency during peak travel periods, and how a cloud‑native, memory‑centric architecture based on Pivotal Gemfire enabled scalable, high‑performance, and highly available ticketing services.

Pivotal Gemfiredistributed cachinghigh concurrency
0 likes · 17 min read
Building a Scalable Cloud Application Platform for the 12306 Railway Ticketing System with Pivotal Gemfire