Official service account of 360 Smart Cloud, dedicated to building a high-quality, secure, highly available, convenient, and stable one‑stop cloud service platform.
This article explains the background, overall architecture, and detailed design and implementation of PrivateLink and ClassicLink gateways within a cloud VPC environment, covering their control and forwarding planes, workflow steps, and future optimization plans.
This article explains how to construct high‑performance RoCE v2 and InfiniBand networks within a cloud‑native Kubernetes environment, detailing the underlying technologies, required components, configuration steps, and performance test results that demonstrate significant communication speed improvements for large‑scale AI model training.
This article provides a comprehensive tutorial on fine‑tuning the Qwen‑14B large language model, covering the motivation, fine‑tuning concepts, step‑by‑step workflow, required code, DeepSpeed training parameters, testing scripts, and deployment using FastChat and the 360AI platform.
FFmpeg provides a comprehensive, cross‑platform hardware acceleration framework that abstracts diverse GPU and dedicated video codec interfaces, defines HWContext types, device and frame contexts, and various codec configuration methods, enabling efficient video encoding, decoding, and filtering while addressing performance, compatibility, and pipeline complexity challenges.
This article describes how 360 unified Kubernetes pod networking with OpenStack VPC by developing a custom CNI plugin that leverages Neutron elastic ports, IPVLAN L2 mode, and OVS to achieve layer‑2 connectivity between VMs and pods, including detailed implementation steps and command examples.
This article explains virtual network security mechanisms, including security groups, network ACLs, QoS policies, and quality‑assurance techniques, detailing implementations with Linux Bridge, iptables, OVS, conntrack, and DPDK, and discusses active probing and passive observation for network reliability.
This article explains how virtual machine traffic is forwarded across compute nodes in an IAAS cloud using overlay protocols such as VXLAN, detailing L2/L3 inter‑node forwarding, control‑plane designs, and forwarding‑plane implementations with Linux‑Kernel OVS, DPDK, NIC offload, and VDPA.
This article explains how virtual networks operate in cloud computing by covering virtual NIC implementation, Virtio frontend/backend mechanisms, KVM signal interaction, and both layer‑2 and layer‑3 forwarding within a node using OVS and related technologies.
The rapid growth of AI compute demand is forcing data centers to redesign cabinet power capacity, adopt advanced cooling solutions such as liquid cooling, and re‑evaluate space density and construction timelines to meet the high‑density, high‑power needs of modern AI workloads.
This article reviews distributed training techniques for large deep‑learning models, covering data parallelism, model parallelism (including pipeline and tensor parallelism), gradient bucketing and accumulation, 3D parallelism, and practical implementations such as Megatron‑LM and 360AI platform optimizations.
