Skip to main content

Overview

Fluence Provider Application is the primary tool for Providers on the Fluence Network to manage all core operations within the Fluence Protocol: from server-level hardware management to interactions with blockchain smart contracts. It is a centralized application, developed and securely hosted by the Fluence team.

Note

A self-hosted version of Provider Application will be available in the near future.

Provider Application contains functionality for managing the following Fluence Network components:

  • Infrastructure Management Software:
    • Talos Linux operation system: Launch and manage Talos Linux as the supported main OS
    • Kubernetes clusters: Provision and manage Kubernetes (K8s) clusters on servers running Talos Linux.
    • Custom Fluence Stack: Provision, configure and automatically update Fluence services running within a Kubernetes cluster
  • Hardware Networking:
    • Private networking: Configure network interfaces, VLANs and Private Subnets for Talos OS and K8s
    • Public IP subnets: Assign IP subnets for Kubernetes clusters for launching workloads, such as Virtual Machines, for Fluence customers
  • Blockchain Smart-contracts (will be supported in early Q4 2025):
    • Offer contracts: Register and update hardware resources and Data Center on-chain info. Manage prices for server resources in stablecoins.
    • Capacity commitments contracts: Enable servers to participate in Fluence network by getting FLT collateral via Staking. Get revenue in FLT token for idle hardware resource.
    • Deal contracts: Get revenue in stablecoins from Fluence Cloud customers renting your infrastructure.

Infrastructure requirements​

Data center requirements​

  • Tier: 3 or 4 certification
  • Additional certificates: not required but highly desirable to have one or more of the following certificates: ISO 27001, SOC2 Type 1, SOC2 Type 2

Hardware requirements​

  • Type: Bare metal servers
  • Server CPU:
    • AMD EPYC or Ryzen Zen2 or better
    • Intel Xeon Cascade Lake or better
    • Unrestricted Turbo Boost enabled
    • Minimum 32 vCPU per CPU unit
  • Server RAM:
    • DDR4+ type
    • Minimum 4 GiB per 2 vCPU
  • DAS Storage:
    • SSD or NVMe types
    • Minimum 25 GiB per 2 vCPU with disabled RAID for hosting customer facing workloads
    • Recommended to use dedicated RAID1 NVMe for Host Operation System with 200+ GiB free space. It is recommended to use different NVMe models in the RAID to reduce the risk of simultaneous failures due to similar disk lifetimes
  • Network interfaces:
    • 1+ GbE interface for Talos OS and Kubernetes
    • 100+ GbE recommended, with 10 GbE minimum, interface for customer facing workloads

Network requirements​

  • Management subnet - management network for Host OS (Talos) and Kubernetes APIs:
    • Allow egress to the Internet
    • Support Bond
    • Can use VLAN
  • Workload subnet - workload network for Fluence customers’ traffic:
    • Should be isolated from internal networks
    • Should have access only to/from the Internet
    • Can’t be used as Bond
    • Can use VLAN

subnets

Possible network topology within data center. Orange area represents a physical server.