Single Pane of Glass
Every GPU, every job, every user — on one screen. Stop tab-switching between consoles — Control Center pulls the entire fleet into a single, fast UI.
/ CONTROL CENTER
One pane of glass for your entire AI fleet
Tscale Control Center is a single web UI for every GPU, every job, every dollar. Built for platform teams who manage hundreds of users — and loved by finance for the cost breakdowns.
Real-time Visibility
Sub-second metric updates via streaming. Watch a job land, scale, and complete in real time — no refresh, no polling.
Secure By Default
SSO, RBAC, audit logs, and signed audit trails for every action. SOC 2 and ISO 27001-aligned out of the box.
/ WORKSPACES & TEAMS
Multi-tenant, zero friction
Run dozens of research groups on one platform without spreadsheets. Workspaces, RBAC, and project-level budgets keep every team in their own lane — and finance happy.
- Workspaces — isolated environments per team, project, or business unit. Each with its own GPUs, budgets, and policies.
- Role-based access — Owner, Admin, Member, Viewer, Billing. Define custom roles for fine-grained control.
- SSO & SCIM — Okta, Azure AD, Google Workspace, JumpCloud. Auto-provision users from your IdP.
- Project budgets — hard caps with email alerts at 80% and 100%. Stop runaway costs before they happen.
/ COST & USAGE
Show me the money
Know exactly where every dollar goes. Break spend down by team, project, GPU type, or region. Export to CSV, push to your data warehouse, or wire alerts directly into Slack.
-
Real-time cost analytics
Spend updates every 60 seconds. No more waiting for the monthly bill to know you went over budget.
-
Per-team, per-project
Chargeback reports that map spend to cost centres, projects, or customers — automatically.
-
Budget alerts
Email, Slack, or webhook alerts at 50%, 80%, 100% of project budgets. Hard caps available on request.
/ PLATFORM
Everything in one console
Control Center isn’t just a dashboard — it’s the operational hub for every Tscale service. Manage compute, storage, jobs, costs, and teams without leaving the browser.
Fleet Management
- Real-time instance dashboard
- One-click start / stop / restart
- Live resource utilization
- Bulk operations
- Console & SSH access
Job Management
- Slurm queue management
- Kubernetes pod inspector
- Live log streaming
- Job history & audit
- Cancellation & requeue
Storage
- Object browser & uploader
- Volume manager
- Snapshot scheduling
- Lifecycle policies
- Cross-region replication
Cost Management
- Real-time spend analytics
- Per-team, per-project breakdowns
- Budget alerts (Slack · email · webhook)
- Reserved capacity planning
- CSV / Parquet export
Access Control
- SSO (SAML / OIDC)
- SCIM auto-provisioning
- Role-based access control
- Audit logs (immutable)
- API keys & service accounts
Integrations
- Slack & Microsoft Teams
- PagerDuty & Opsgenie
- Datadog & Grafana
- Snowflake & BigQuery export
- Webhooks for everything else
Performance
Streaming metric updates
WebSocket-driven live data — no polling, no refresh, no stale numbers.
Managed from one screen
Tscale’s UI scales to fleets of 100,000+ GPUs without breaking a sweat.
Immutable audit logs
Every action, every change, every login — cryptographically signed and exportable.
Backed by enterprise SLA
Multi-region active-active, 24/7 SRE on call. Your team can always log in.
For teams that run at scale
Platform Teams
Manage hundreds of users across dozens of teams. Workspaces, RBAC, project-level budgets, chargeback reports — everything a platform team needs to scale.
Learn MoreFinance & Procurement
Cost breakdowns, chargeback exports, reserved capacity planning, and budget alerts — built for the people who write the cheques.
Learn MoreControl Center manages everything
Every Tscale service is reachable from the same console. Click into an instance, see its job queue, check its cost, and audit who did what — without leaving the browser.