Compute / Edge Computing

Edge Computing

WLStack Edge brings container, bare-metal and network capabilities closer to end users, devices and the field — delivering lower-latency, more cost-effective local compute for media distribution, real-time interaction, IoT access and edge inference.

Apply for a trial

<20ms

Nearby access latency

Multi-region

Edge node coverage

99.9%

Service availability

Edge+Cloud

Unified scheduling

Push compute and service capabilities closer to users

WLStack Edge addresses low-latency, distributed and regional workloads by moving compute, cache and access capabilities closer to users and devices. It offers edge containers, edge bare metal and K8s@Edge, suited to media distribution, real-time interaction, edge inference, IoT access and regional processing — coordinated with the central cloud for unified management, tiered processing and data collaboration.

Key advantages

Smart scheduling, lower cost

A self-developed scheduling platform plus nearby resource reuse fuses backbone, aggregation and edge scheduling to cut service, content-delivery and central compute cost while preserving experience.

  • Nearby resource reuse
  • Lower origin bandwidth
  • Intelligent scheduling
  • Predictable cost

Nearby access, lower latency

Widely distributed edge nodes cover multiple regions and major carriers; data is collected and processed locally to shorten transmission distance and reduce jitter.

  • Near-user deployment
  • Local collection & processing
  • Faster first response
  • Less access jitter

Heterogeneous, peak performance

Edge nodes use X86, ARM and NV/AMD GPU heterogeneous resources and can provide bare metal directly, meeting high-IO and compute-heavy performance demands.

  • X86 / ARM / GPU mix
  • Direct bare metal
  • High-IO, compute-heavy
  • Peak device performance

Reliable, edge-cloud coordinated

Node-level monitoring, alerts and system-level support, with edge nodes and the central cloud working as one to keep edge workloads continuous and available.

  • Node monitoring & alerts
  • >99.9% availability
  • Unified control plane
  • Fast fault handling

Platform capabilities & product forms

Lightweight HCI, standardized edge nodes

A software-defined lightweight hyper-converged platform consolidates compute, storage and network into standardized edge nodes, pooling hardware via virtualization to run distributed services on low-power devices.

  • Dynamic allocation of CPU/memory/storage
  • Unified CPU/GPU/ARM heterogeneous management
  • Edge persistence: power-loss protection & local cache
  • Container-native: K8s and Docker

Layer 4–7 global smart scheduling engine

A global scheduling engine using federated and reinforcement learning analyzes network topology, node load and QoS metrics in real time for intelligent task orchestration across edge clusters.

  • Multi-dim sensing: latency probe (≤10ms), health scoring
  • Elastic scaling on traffic peaks
  • Self-healing: fast service migration
  • Path planning: BGP L4 routing + L7 scheduling

Connecting core, aggregation and edge access

Backbone transport connects the BGP Anycast core, IDC aggregation and edge access networks via a programmable data plane, able to integrate 5G smart networks and cover campus and metro networks.

  • Efficient transport: QUIC/HTTP3 and TCP acceleration
  • Protocol decoupling: IP/Non-IP passthrough (MQTT, CoAP)
  • Zero-config mesh with zero-trust auth
  • Slice isolation: hard-pipe bandwidth (min 1Mbps)

Edge container, edge bare metal and edge-cloud

K8s@Edge edge containers, dedicated-resource edge bare metal, and edge-cloud coordination with central-cloud governance — combined to fit your needs for performance, isolation and elasticity.

  • K8s@Edge edge containers
  • High-performance edge bare metal
  • Edge-cloud coordination & backhaul
  • Second-level deploy & elastic scaling

Edge-oriented monitoring, alerts and support

Full observability of node state, traffic and jobs, backed by 24/7 technical support and a visual alerting platform for stable operation.

  • Node monitoring
  • Real-time alerts
  • 24/7 support
  • Fault handling

Edge scenarios that fit your business

VOD and download distribution

For video-on-demand, game downloads and updates, phone OTA and app downloads, large-capacity caches on widely distributed aggregation edge nodes raise hit rate and throughput while cutting cost.

  • Large edge caches
  • Higher download hit rate
  • High-throughput delivery
  • Lower origin cost

Audio & video processing

Deploying transcoding, distribution and processing at edge nodes lowers latency and saves central bandwidth via nearby transcoding, with capabilities like beautification, filters and content moderation.

  • Nearby transcoding
  • Lower network latency
  • Save central bandwidth
  • Moderation / enhancement

E-commerce & interactive media

Software-defined real-time transport builds a high-quality, cross-carrier, fully covered network for video conferencing, online education and interactive live streaming — smooth low-latency interaction and viewing.

  • Real-time transport
  • Cross-carrier coverage
  • Low-latency interaction
  • Smooth HD experience

Digital-intelligence apps & AI

Near data and application sources, edge nodes provide traditional compute and large-model appliances plus speech, text and image algorithm APIs — cutting transfer cost, improving responsiveness and keeping data within-city and within-domain.

  • Compute near data source
  • Large-model appliances
  • Algorithm APIs
  • Local data compliance

Images & pre-installed apps

A range of OS images and one-click pre-installed applications, ready on boot.

FreeBSD
FreeBSD
Windows
Windows
Linux
Linux
PostgreSQL
PostgreSQL
Nginx
Nginx
Docker
Docker
WordPress
WordPress
LAMP
LAMP
LEMP
LEMP

FAQ

VOD and download distribution, audio/video processing, real-time interaction, cloud gaming, low-latency APIs, IoT access, edge inference, field data processing and regional application deployment.

Edge containers (K8s@Edge), edge bare metal and edge-cloud coordination — combined to fit your needs for performance, isolation and elasticity.

Yes — edge nodes coordinate with the central cloud for unified control, tiered processing and centralized operations.

Edge nodes use X86, ARM and NV/AMD GPU heterogeneous resources and can provide bare metal directly, meeting high-IO and compute-heavy demands.

Nearby caching and resource reuse cut origin bandwidth and central compute consumption; combined with intelligent scheduling, this optimizes overall cost while preserving experience.