Platform deployment

Deployed in your environment, managed with your team.

TRAIN is a managed Causal AI application installed inside infrastructure you control. The architecture is infra-agnostic (container, virtual machine, or internal HPC) and runs on cloud, on-premises or hybrid. Software installation typically takes one to two weeks after infrastructure readiness; a scoped pilot or application rollout typically takes 6–12 weeks, depending on data and validation requirements. Parabole supports configuration, updates and maintenance from there.

Scroll down · delivery, the three architectures, and requirements
Delivery

What is handed over, and how.

The platform is delivered through a secure, credential-protected portal or supported cloud marketplace, with signed installers, infrastructure templates, a manifest, release notes, and checksums.

Artifacts

Installers & templates

Local and network installer packages, ARM templates for Azure, and Terraform configurations for multi-cloud. A manifest carries version details, release notes and SHA-256 checksums; optional configuration files and scripts are digitally signed.

Channels

Portal or marketplace

Secure, credential-protected HTTPS links by portal or email, or Azure Marketplace for one-click provisioning. Patches and upgrades are delivered the same way as the initial deployment, with email notification of each release.

Versioning

Semantic & licensed

Semantic versioning (for example, v1.2.0) supports traceability. The license is organization-specific and validated during installation; it defines access and feature entitlements for the deployment.

Reference architectures

Three deployment architectures.
Choose what fits your operations.

Every model runs the same platform components: a network layer, teaching and runtime services, language and optimization services, and the stateful stores. What changes is how the compute is packaged. Each diagram shows the components, their interconnections and deployment tiers, inside your private network.

Container / Kubernetes

Platform components run as isolated, scalable containers under an orchestrator such as Kubernetes: high availability, horizontal scaling, rolling updates and fault isolation, suited to teams already running CI/CD and cloud-native practices. Services split across a data-services node, a CPU node set and a GPU node set; databases and messaging run as managed services or self-hosted in the cluster.

Figure 1: container / Kubernetes deployment architecture.
ResourceDescriptionConfigurationRuns
Data services nodeNode, no GPU16 core · 128 GB RAMUser management, application frontend
CPU node setNode set, no GPU16 core · 128 GB RAMAPI services, data & knowledge services
GPU node setNodes with GPU80 GB VRAMLanguage services, teaching & runtime services
DatabasesManaged or self-hosted16 core · 128 GB RAM (self-hosted)PostgreSQL and GraphDB
StorageCloud object storageN/AArtifact storage

Best for · cloud-native teams · CI/CD · elastic scaling

VM-based

Each core service runs on a dedicated virtual machine (separation of concerns, resource isolation and straightforward debugging), for environments where container orchestration isn't available or preferred: on-premises or regulated cloud setups needing long-running VMs, tight control over the OS and networking layers, manual provisioning and custom security hardening.

Figure 2: VM-based deployment architecture.
ServerTypeConfigurationRuns
Server 1VM with GPU40 GB VRAM · 60 core · 480 GB RAMParabole platform: teaching & runtime
Server 2VM, no GPU16 core · 128 GB RAMUtility services, graph database
Server 3VM with GPU80 GB VRAM · 60 core · 480 GB RAMLanguage services
Server 4VM, no GPU8 core · 64 GB RAMOptimization service
StorageCloud / local disk / NASN/AArtifact storage

Best for · on-premises · regulated cloud · traditional IT operations

Internal HPC

For organizations with internal high-performance computing (HPC), the TRAIN server hosts application services and submits GPU workloads (including language modeling, training, and inference) to the cluster through SlurmCTLD.

Figure 3: internal HPC (SLURM) deployment architecture.
ResourceDescriptionConfigurationRuns
TRAIN serverOn-prem Linux VM60 core · 512 GB RAMTeaching, runtime, job & utility services, GraphDB, PostgreSQL
SLURM head nodeHPC controllerManaged by customer ITAccepts job submissions
SLURM compute nodesGPU-enabled HPC nodes80 GB VRAMExecutes ML / AI workloads
Shared storageNFS or parallel file systemHigh IOPS, reachable by bothModel data and job output

Best for · sites with an existing Slurm-based HPC cluster

Scalability

One instance per environment. Unlimited applications on top.

The architecture supports multiple applications on a shared platform instance. Commercial terms scale by deployment, model, and application scope.

From IT environment to sites, processes and products: a model is an address in the physical plant. Applications can reuse the same governed model; commercial terms are defined in the applicable agreement.
01

One TRAIN instance per IT environment

Each environment (its own hardware and software, its own MES, ERP, SCADA and historian) runs one instance. Environments never share an instance, so a deployment in one region and one in another scale, upgrade and are governed independently.

02

Add a site, not a platform

A single instance serves many sites. Bringing a new plant onto an existing instance is a configuration and connectivity step against infrastructure you already run, not a new deployment.

03

A model is an address: Site × Process × Product

Every causal model is pinned to one point in the plant: a site (the data origin), a manufacturing process and a product or family. Change any one coordinate and it is a different model; keep all three and one model covers that address. Retraining and updates keep the same address; they do not create a new model.

04

Multiple applications, one governed model

Multiple causal applications can run on a governed model. Licensing and metering are defined in the applicable commercial agreement.

Environments & methods

Cloud, on-premises, or hybrid:
installed to fit your environment.

Cloud

Azure · OCI · AWS

Azure deployment is supported through ARM templates; supported cloud environments can be provisioned through validated Terraform configurations.

On-premises

Bare metal or VMs

Full control of the OS, network and upgrade path, including air-gapped environments.

Hybrid

Cloud + on-prem

Integrates cloud and on-premises resources, including HPC and edge nodes.

Deployment methods
MethodWhen to use
Local installerAir-gapped or on-premises environments with no internet access or strict security controls. Transfer the package, run the installer, validate.
Network installerEnvironments with outbound HTTPS access to the Parabole repository: the quickest path where internet access is allowed.
ARM templateAzure-only deployments that want native platform integration.
TerraformCustom or multi-cloud and hybrid deployments that need flexibility across providers.
HPC integrationOn-premises environments with a Slurm-based HPC cluster; requires shell access to submit jobs to SlurmCTLD.
Infrastructure requirements

The software, network and storage the platform expects.

Hardware is listed per architecture above. These requirements are common to every model.

Software
LayerRequirement
Operating systemUbuntu 22+ or RHEL 8+ (64-bit, systemd)
RuntimesPython 3.10+ · OpenJDK 17+
DatabasesPostgreSQL 15+ (on-prem or cloud-managed) · Ontotext GraphDB 10+ or equivalent RDF store
MessagingKafka and Zookeeper for data ingestion
GPUCUDA 12.4+ with NVIDIA drivers
WebNginx for UI, reverse proxy and SSL
Containers (if used)Docker · Kubernetes 1.25+ · Helm 3.10+
Network

Ports & DNS

Inbound: 443 HTTPS, 22 SSH, 8585 APIs, 8888 UI, 5010 internal. Static DNS entries for service discovery and SSL. The network installer additionally needs outbound 443 to the Parabole repository, proxy-aware for corporate firewalls.

Storage

Objects & backups

Cloud object storage (Azure Blob, OCI Object Storage) or local high-IOPS storage for logs, models and backups. Backups are run by the customer's IT team according to organizational policy.

Security

Access & encryption

Supports least-privilege IAM/RBAC, MFA for privileged access, audit logging, encryption at rest, and TLS in transit. Supported secrets-management integrations are confirmed during architecture review.

High-availability and disaster-recovery options, including supported multi-region patterns, are defined during architecture review; RPO and RTO are agreed in the applicable SLA.

Responsibilities

Your team owns the environment. Parabole owns the platform.

A RACI split makes ownership explicit across infrastructure, security and the installation itself.

TaskCloud adminDevOps / customer ITSecurity adminVendor support
Infrastructure provisioningRACR/C
Network configurationRCAR/C
Deployment script executionCR/AIR/C
Security & access configurationCRAC
Platform installationCR/AIR
Post-install validationCRCR/C
Backup & DR setupRACC
Monitoring setupRACC
Troubleshooting & supportCRCR/A

R responsible  ·  A accountable  ·  C consulted  ·  I informed. Roles needed: cloud administrator, DevOps engineer or customer IT, security administrator, and Parabole vendor support, with working knowledge of Linux CLI, YAML and Bash, Terraform or the Azure CLI, and SSH / TLS / VPN / firewall administration; Kubernetes where the container model is used.

Rollout

Software installation: 1–2 weeks, five milestones.

Duration depends on complexity. A pre-check script validates the environment before anything is installed.

01

Planning & assessment

Define scope, assign roles against the RACI, and confirm prerequisites. Run precheck.sh from the package to validate CPU, RAM, Python, CUDA, open ports, DNS resolution and disk.

02

Infrastructure provisioning

Stand up servers or nodes, networking and storage for the chosen architecture.

03

Installation & configuration

Run the installer for the chosen method, then wire in external data sources (object storage, PostgreSQL, GraphDB) and define user roles and permissions from the admin dashboard.

04

Validation & testing

Confirm that the health endpoint returns 200 OK, review logs, and run smoke-tests.sh. Monitor CPU and memory utilization, GPU availability, database health, and latency.

05

Handover

Document the deployment, train the operating team, and transition to support.

Run & support

After go-live.

Maintenance

Updates & upgrades

Regular releases for bug fixes, features and security patches, delivered the same way as the initial deployment. Customers are notified by email of every new release or patch.

Troubleshooting

Logs & common issues

Logs are stored under <parabole-path>/logs/. For installation, API, or license errors, review the service logs and troubleshooting guide before contacting support.

Support

SLA & escalation

Response targets are defined by severity and the applicable support agreement. Contact support@parabole.ai with a summary, diagnostics, and priority.

Get started

Plan a deployment with your infrastructure team.

Bring your cloud, DevOps and security leads. We will map TRAIN onto your environment and agree the topology at an architecture review.

Request a demo