VarolioVarolio Docs
Deployment scenarios

Dedicated Cloud

Single-tenant Varolio environment in a dedicated, Varolio-operated AWS account

Dedicated Cloud

Dedicated Cloud is Varolio's single-tenant deployment model for customers who need a fully isolated environment and a choice of region, without owning or operating the underlying cloud account. The environment is dedicated to one customer, hosted in a separate AWS account that Varolio owns and operates — there are no other tenants and no shared services.

It sits between the other models: more isolated than Cloud SaaS, which is multi-tenant; lighter to adopt than Cloud-Prem, which runs inside the customer's own account. The software is functionally identical across all models — same services, same data layer, same AI surface.

Varolio Operations
GitOps · same release pipeline
Customer
Dedicated AWS Account · customer-chosen region (e.g. ca-central-1)
Proxy
AWS WAF
AWS WAF
Application Load Balancer
Application Load Balancer
Istio Ingress Gateway
Istio Ingress Gateway
Object storage — dedicated, in-region
Amazon S3 (in-region)
Amazon S3 (in-region)
Dedicated Amazon EKS cluster
Engine
API service · control-plane access
Agent
AI orchestration · retrieval
Worker
Temporal workflow worker
MCP Server
MCP endpoint
External Managed Services — dedicated instances
Supabase
Supabase
PostgreSQL
Elastic Cloud
Elastic Cloud
Redis
Redis
Temporal Cloud
Temporal Cloud
workflow orchestration
Vertex AI (Gemini)
Vertex AI (Gemini)
regional endpoint
Amazon Bedrock
Amazon Bedrock
regional endpoint

Region and environment

Each Dedicated Cloud customer gets a separate AWS account with its own VPC, EKS cluster, S3 buckets, and secret store, plus dedicated Supabase (PostgreSQL), Elastic Cloud, Redis, and Temporal Cloud (workflow orchestration) instances. Nothing is shared with any other customer.

Unlike Cloud SaaS, which is hosted in Frankfurt, the customer chooses the region at deployment time, so data-residency requirements are met by where the environment is provisioned. Canada (ca-central-1) is validated and running in production today; other AWS regions are available on request.

Tenancy and isolation

The environment is single-tenant by construction: there is exactly one tenant in the account, and isolation comes from the account and VPC boundary rather than only from application-layer filtering.

  • The operational database, search index, object storage, and message fabric are dedicated to the one customer. No data path crosses into another customer's environment because no other customer exists in the account.
  • All the application-layer isolation controls from Cloud SaaS (workspace-scoped queries, namespaced object keys, per-tenant credentials, separate admin surface) still apply on top of the account boundary.

Public surface

HostnamePurpose
app.{region}.varolio.ioMain user web application (Studio)
{region}.api.varolio.ioPlatform API (engine)
mcp-{region}.varolio.ioMCP server for AI-tool access

For example, the Canadian production environment is served at app.ca.varolio.io, ca.api.varolio.io, and mcp-ca.varolio.io. The edge uses the same AWS WAF + ALB (TLS 1.2+, ACM) configuration as Cloud SaaS, provisioned per environment.

Encryption

Identical to Cloud SaaS: AES-256 at rest across Supabase (PostgreSQL), Elastic Cloud, and S3 (SSE-S3); secrets in AWS Secrets Manager with CSI sync; TLS 1.2+ in transit at every external hop.

Operational ownership

AreaOwner
AWS account, VPC, EKS, data storesVarolio (dedicated account)
OS and node patchingVarolio (EKS-managed node groups)
Application code and releasesVarolio (same GitOps pipeline as Cloud SaaS)
Database backupsVarolio
Secret rotationVarolio
Monitoring, alerting, on-callVarolio
Cloud billVarolio (included in subscription)
Customer user accounts and rolesCustomer
Customer SSO configurationShared — customer provides IdP (OIDC), Varolio configures
Integration credential authorizationCustomer

Upgrade cadence

Dedicated Cloud environments are configured in the same GitOps repository as Cloud SaaS, as a parallel environment tree. Every release flows through the same CI and review pipeline, then rolls out per environment — which means the upgrade window for a dedicated environment can be scheduled with the customer where contractually required.

AI and embedding providers

Same defaults as Cloud SaaS (Application Architecture — AI providers), with inference routed to regional provider endpoints where available. Customers who require inference against their own model deployment can combine Dedicated Cloud with BYO Data Store.

When to choose a different model

  • Choose Cloud SaaS if EU (Frankfurt) residency is sufficient and you do not need single-tenant isolation.
  • Choose BYO Data Store if the search index, LLM, or object storage must run on infrastructure you host.
  • Choose Cloud-Prem if the environment must run inside a cloud account you own.

AI Tools

Ask ChatGPTAsk Claude

On this page