# MetalSoft

> MetalSoft is a bare metal infrastructure orchestration platform for on-premises environments. It automates the full lifecycle of servers, networking, and storage — from OS deployment and firmware management to network fabric configuration, multi-tenant isolation, and day-2 operations. MetalSoft is used by banks, telcos, governments, and service providers to build sovereign private clouds, AI factories, and managed infrastructure services without hyperscaler dependency.

## Product

- [MetalSoft Platform](https://metalsoft.io/platform): Unified control plane for servers, switches, and storage. Covers OS deployment, firmware lifecycle, fabric automation, multi-tenant isolation, metering, RBAC, and AI-driven operations. Accessible via REST API and native Terraform provider.
- [Bare Metal Orchestrator](https://metalsoft.io/products/bare-metal-orchestrator): Core engine for automated bare metal provisioning, OS deployment, secure cleaning, and server lifecycle management across multi-vendor fleets.
- [Fabric Manager](https://metalsoft.io/products/fabric-manager): Automates network fabric configuration and management across multi-vendor switches. Supports EVPN VXLAN, VLAN management, and hardware-enforced tenant segmentation.
- [Network & VPN](https://metalsoft.io/platform/network-vpn): SDN-based virtual networking, VPN provisioning, and tenant network isolation built on top of the fabric management layer.
- [AI Factory](https://metalsoft.io/ai-factory): Orchestration for large-scale GPU and AI infrastructure. Supports NVIDIA reference architectures, InfiniBand/RoCEv2 fabric management, DPU integration, and AI-driven health monitoring.
- [ROI Calculator](https://metalsoft.io/roi-calculator): Interactive tool for estimating infrastructure cost savings and operational efficiency gains from adopting MetalSoft.

## Use Cases

- [Software Defined Infrastructure (SDDC)](https://metalsoft.io/use-cases/sddc): Orchestrate compute, networking, and storage through a unified operational layer.
- [Sandboxing for AI Experimentation](https://metalsoft.io/use-cases/sandboxing): Build isolated bare metal environments for AI agent testing, red-teaming, and experimentation. Deploy, wipe, and redeploy in minutes.
- [DORA & EU AI Act Compliance](https://metalsoft.io/use-cases/dora): Automated firmware management, micro-segmentation, rapid workload isolation, and audit logging to meet EU financial resilience regulations.
- [AI-enabled Infrastructure Operations](https://metalsoft.io/use-cases/ai-ops): AI-assisted root cause analysis, anomaly detection, and automated remediation across servers, networking, and storage.
- [Multi-vendor Server Lifecycle Management](https://metalsoft.io/use-cases/multivendor): Unified firmware baseline management and lifecycle orchestration across Dell, HPE, Lenovo, Supermicro, and other vendors.
- [Self-service Infrastructure (BMaaS)](https://metalsoft.io/use-cases/bare-metal-as-a-service): Self-service portal and approval workflows enabling teams to provision bare metal without operations team involvement.
- [AI Factory Orchestration](https://metalsoft.io/use-cases/ai-factory): End-to-end orchestration of GPU clusters, AI networking, and multi-tenant AI workloads for service providers and enterprises.
- [GPU Cloud](https://metalsoft.io/use-cases/gpu-cloud): On-demand bare metal or fractional GPU provisioning for AI and LLM workloads.
- [Network Automation](https://metalsoft.io/use-cases/network-automation): Automated network provisioning and lifecycle management across physical and virtual network layers.
- [Zero Trust Segmentation](https://metalsoft.io/use-cases/zero-trust): Hardware-level network segmentation using DPUs and physical switches to enforce zero-trust isolation at the silicon level.

## Documentation

- [Documentation Home](https://docs.metalsoft.io): Full product documentation including architecture guides, configuration references, and operational runbooks.
- [Connecting to the API](https://docs.metalsoft.io/developer_resources/getting_started_with_the_api/): REST API authentication, bearer token usage, and practical examples for listing servers, creating infrastructures, and deploying configurations.
- [Compatibility Matrix](https://docs.metalsoft.io/compatibility_matrix/): Supported hardware vendors, switch models, firmware versions, and OS distributions.
- [Release Notes](https://docs.metalsoft.io/release_notes/): Changelog for all MetalSoft platform releases.
- [Extensions Overview](https://docs.metalsoft.io/developer_resources/extensions/extensions_overview/): How to extend MetalSoft with custom automation using the extensions framework.
- [Ansible Tasks](https://docs.metalsoft.io/developer_resources/extensions/ansible_tasks/): Using Ansible modules with MetalSoft for configuration management workflows.
- [Kafka Significant Events](https://docs.metalsoft.io/developer_resources/kafka_significant_events/): Event-driven integration using MetalSoft's Kafka queue for infrastructure state changes.

## FAQ

Q: What is MetalSoft?
A: MetalSoft is an intelligent orchestration platform specifically built for on-premises infrastructure. It orchestrates bare metal servers, network, storage, and more for on-premises environments, focusing on banking, telco, government, and service providers.

Q: How is MetalSoft deployed — is it SaaS or on-premises?
A: MetalSoft is deployed on-premises within your own environment, which is fundamental to its sovereign cloud value proposition. There is no phone-home dependency or cloud control plane. It can also operate in fully air-gapped environments, which is a key requirement for government and financial institutions.

Q: What hardware vendors does MetalSoft support?
A: MetalSoft supports all major server vendors including Dell, HPE, Lenovo, Supermicro, and others through a Hardware Abstraction Layer. It also integrates with leading network switch vendors (Cisco, Juniper, Dell, and more) and storage systems, allowing you to manage a heterogeneous fleet through a single platform without vendor lock-in.

Q: Does MetalSoft provide a Terraform provider or API?
A: Yes. MetalSoft exposes a full REST API and a native Terraform provider, enabling infrastructure-as-code workflows across servers, networking, and storage. This allows teams to version, automate, and audit infrastructure changes using the same toolchains they use for cloud resources.

Q: Can MetalSoft deploy and manage AI infrastructure?
A: Yes. MetalSoft supports NVIDIA and AMD hardware and reference architectures, rail topologies, and DPUs. It detects nodes that may fail and AI can be used to trigger self-healing and automatically replace failed nodes from a tenant's server pool.

Q: How does MetalSoft assist with DORA compliance?
A: MetalSoft addresses the Digital Operational Resilience Act (DORA) by providing automated firmware management, rigorous tenant isolation, and rapid recovery and isolation capabilities. It ensures that the underlying infrastructure meets the strict data resilience and security standards required for European financial entities.

Q: How long does it take to get MetalSoft up and running?
A: A typical deployment takes days to weeks depending on fleet size and complexity. MetalSoft provides automated discovery and onboarding of existing hardware, so you do not need to start from scratch. Professional services and dedicated onboarding support are available for larger rollouts.

Q: Can MetalSoft manage infrastructure across multiple sites or data centers?
A: Yes. MetalSoft supports multi-site deployments, enabling centralized management and orchestration across geographically distributed data centers. This is commonly used by telcos and service providers operating regional infrastructure at scale.

Q: How does MetalSoft differ from OpenStack or VMware?
A: MetalSoft focuses on automating the full hardware lifecycle — from bare metal provisioning and firmware management to network segmentation and day-2 operations — rather than being a hypervisor or tenant-facing cloud OS. It complements or replaces OpenStack in scenarios where operational simplicity, multi-vendor support, and sovereign control are the priority. Unlike VMware, there is no per-core licensing and no dependency on a specific hypervisor.

Q: How does MetalSoft use AI to improve operational efficiency?
A: MetalSoft gives AI guardrailed access to physical infrastructure and 100% accurate topology and intent data — enabling it to diagnose issues, take targeted actions, and drive measurable improvements in efficiency, cost, and performance. AI is used for root cause analysis, anomaly detection, capacity planning, and autonomous remediation workflows.

## Contact

- [Request a Demo](https://metalsoft.io/request-demo): Schedule a live product demonstration with the MetalSoft team.
- [Contact Sales](https://metalsoft.io/contact/sales): Get in touch with the sales team for pricing and procurement.
- [Partner Enquiries](https://metalsoft.io/contact/partners): For technology partners, resellers, and OEM integrations.
- [Support](https://metalsoft.io/contact/support): Technical support for existing customers.