Costs & Billing

Cloud resources are billed on a pay-as-you-go model — you pay for what you use, typically by the hour or second. This section explains how billing works in the Campus Cloud, the discounts available to UCSB, and how to keep costs under control. For how cost responsibilities are divided between you, the Cloud Team, and the provider, see Shared Responsibility.


Baseline Costs

These baseline services include:

  • Audit log storage
  • Security Hub / Defender for Cloud / Security Command Center
  • GuardDuty (AWS)
  • CloudTrail (AWS)

Tracking Your Costs

Each provider gives you built-in tools to view your spend, set budget alerts, and break costs down by tag. A budget alert notifies you as you approach a threshold — it does not stop your account from spending — so keep an eye on your account, especially in the first few weeks of a new workload.

AWS

Dashboards

View your spend in AWS Cost Explorer — in the console, go to Billing → Cost Explorer.

For richer cross-account analytics, the Cloud Team can grant access to the AWS Cost Intelligence Dashboard and CUDOS dashboard; request access through a ServiceNow ticket.

Budget Alerts

Use the Fixed Monthly Budget with Notification product in the Service Catalog — see AWS First Steps.

Categorizing Costs by Tag

A tag only appears in the AWS cost tools after it’s activated as a cost allocation tag — the Cloud Team does this centrally, so you just apply the ucsb: tags to your resources. You can then group or filter costs by a tag in Cost Explorer, in Cost and Usage Reports (one column per tag), and in the dashboards above.

Tagging isn’t retroactive — only usage after a resource is tagged is attributed, and new tag data can take up to 24 hours to appear.

Azure

Dashboards

View your spend in Azure Cost Management + Billing — in the portal, go to Cost Management.

Budget Alerts

Use Azure Cost Management budgets — see Azure First Steps.

Categorizing Costs by Tag

Azure needs no activation step — tags appear in cost reports automatically. In Cost analysis, group or filter by the required ucsb: tags — for example, ucsb:mission to split spend across academic, research, and administrative work, or ucsb:environment to separate prod from dev. Tag the resources themselves, not just their resource groups — Azure attributes cost by the tags on each resource.

GCP

Dashboards

View your spend in GCP Billing Reports — in the console, go to Billing → Reports.

Budget Alerts and Limits

Use Cloud Billing budgets — see GCP First Steps.

Categorizing Costs by Tag or Label

GCP cost data carries both Resource Manager tags and project labels (see Tags vs. Labels), so you can group or filter billing reports by either — no activation step needed. The required tags you already set for compliance (environment, mission, dept) double as your cost-attribution dimensions; add labels when you need a breakdown those don’t cover.


Purchase Orders and Invoicing

Cloud costs are tracked against the Purchase Order you created in Gateway. Monthly usage is invoiced approximately three weeks after the end of each month. Keep your PO active until all invoices have been processed — see Close an Account for details on the closure timeline.

If your costs are approaching your PO limit, renew or increase your PO before it expires. Contact your departmental financial administrator and the Cloud Team.


Pricing, Discounts & Credits

UC Enterprise Discounts

UCSB participates in system-wide enterprise agreements with all three providers. You receive these discounts automatically — no action required.

Provider Program Key Discount
AWS Enterprise Discount Program (EDP) Percentage discount across all services; additional discount on S3 storage
Azure Enterprise Agreement (EA) Negotiated rates across Azure services
GCP Google Negotiated Pricing Committed use discounts and negotiated rates

Pricing Calculators

Before starting a project, estimate your monthly spend using the official calculators:

Note that the calculators show list prices. Your actual bill will reflect UC enterprise discounts.

Pricing Models

Model Description Best For
On-demand / Pay-as-you-go Billed by the second or hour, no commitment Variable workloads, development, testing
Reserved / Committed use Pay upfront or commit to 1–3 years for a discount Steady, predictable production workloads
Spot / Preemptible / Interruptible Deeply discounted but can be terminated with short notice Batch jobs, fault-tolerant compute

Research Credits

All three providers offer programs that award cloud credits or research funds to academic researchers. If you receive an award, use a Campus Cloud account to redeem it — either create a new account or migrate an existing personal account.

Provider Program What You Get Link
AWS Amazon Research Awards (ARA) Unrestricted funds + AWS Promotional Credits Apply for ARA
GCP Google Cloud Research Credits Up to $5,000 in GCP credits (faculty, PhD students, postdocs) Apply for GCP Research Credits
GCP Google Academic Research Awards (GARA) Unrestricted gifts for computing & technology research (professors) About GARA
Azure Microsoft Research Fellowship Azure compute access for collaborative research (faculty, PhD students, postdocs) About the Microsoft Research Fellowship
Azure Accelerating Foundation Models Research (AFMR) Azure AI services grants for foundation-model research About AFMR

Teaching & Classroom Credits

Faculty can apply for cloud credits to give students hands-on experience in coursework. These programs provide credits to the instructor for distribution to the class — they are not self-service student accounts.

Provider Program What You Get Link
GCP Google Cloud Teaching Credits Faculty apply per-course; GCP credits for students to use in coursework Apply for GCP Teaching Credits
Azure Azure Education Hub Faculty-managed Azure credits and lab environments for classroom use About Azure Education Hub

AWS does not offer a direct credits-to-faculty program; AWS Academy provides free curriculum with a Learner Lab sandbox but requires institutional membership and uses pre-built courses.

Programs change frequently — check each provider’s site for current eligibility, deadlines, and award amounts.


Cost Optimization Tips

  • Reserved Instances / Committed Use Discounts: If you have stable, predictable workloads (e.g., a long-running research cluster), committing to 1–3 years can reduce compute costs significantly.
  • Spot / Preemptible VMs: For fault-tolerant batch workloads, Spot VMs are 60–80% cheaper than on-demand.
  • Rightsizing: Review provider recommendations (AWS Compute Optimizer, Azure Advisor, GCP VM Manager) to downsize over-provisioned resources.
  • Storage lifecycle policies: Move infrequently accessed data to cheaper storage tiers (S3 Glacier, Azure Cool/Archive, GCS Nearline/Coldline).