Google Cloud Cost Optimization India: CUD and SUD Guide
Country Manager, India
AI, Manufacturing, DevOps, and Managed Services. 17+ years across Manufacturing, E-commerce, Retail, NBFC & Banking
How Does Google Cloud Pricing Work in India?
Google Cloud operates a Mumbai region (asia-south1) and a Delhi region (asia-south2), both offering competitive pricing for Indian enterprises. According to Google Cloud's pricing blog (2025), Committed Use Discounts (CUDs) save up to 57% on compute, while Sustained Use Discounts (SUDs) apply automatically and save up to 30%. These two mechanisms form the backbone of Google Cloud cost optimization for Indian workloads.
Key Takeaways
- Sustained Use Discounts (SUDs) apply automatically, no commitment needed, saving up to 30%
- Committed Use Discounts (CUDs) require 1- or 3-year commitments, saving up to 57% on compute (Google Cloud, 2025)
- Google's per-second billing and automatic discounts make it the most transparent pricing model among hyperscalers
- Indian enterprises should combine CUDs with right-sizing and active recommender use
Google Cloud's share of the Indian public cloud market is growing steadily, with IDC (2025) placing it as the third-largest provider behind AWS and Azure. The Delhi region launch in 2022 expanded local options for latency-sensitive workloads. Understanding how Google's discount mechanisms differ from AWS and Azure is essential for Indian CIOs evaluating multi-cloud cost strategies.
cloud cost optimization services
What Are Sustained Use Discounts and How Do They Work?
Sustained Use Discounts (SUDs) are automatic discounts that increase as you run a VM for longer portions of a billing month. Google Cloud applies SUDs without any commitment or upfront action. Google's documentation (2025) shows that running an N1 or N2 instance for an entire month saves roughly 30% compared to the on-demand hourly rate. This happens automatically on your bill.
How the Discount Tiers Work
Google divides each month into usage tiers. The first 25% of the month charges full price. From 25-50%, you get a small discount. From 50-75%, the discount increases. From 75-100%, you receive the maximum discount. The effective monthly rate for a full-month instance ends up about 30% below the sum of hourly on-demand charges. No action required from your side.
Which Instance Types Qualify for SUDs?
SUDs apply to N1, N2, N2D, and C2 machine types but not to E2, Tau (T2D/T2A), or Sole-Tenant Node instances. Importantly, SUDs don't apply to instances already covered by Committed Use Discounts. For Indian enterprises using E2 instances (popular for cost-conscious workloads), SUDs won't apply, making CUDs even more important for those families.
[CHART: Line chart - SUD discount tiers across a billing month showing progressive discounts from 0% to 30% - Google Cloud documentation]
Need expert help with google cloud cost optimization india: cud and sud guide?
Our cloud architects can help you with google cloud cost optimization india: cud and sud guide — from strategy to implementation. Book a free 30-minute advisory call with no obligation.
What Are Committed Use Discounts and When Should You Buy Them?
Committed Use Discounts (CUDs) are Google Cloud's equivalent of AWS Reserved Instances. You commit to a specific amount of vCPU and memory for one or three years and receive discounts up to 57% for three-year commitments. According to Google Cloud (2025), CUDs override SUDs on covered usage, and the CUD discount is always deeper than the maximum SUD rate.
Resource-Based vs Spend-Based CUDs
Google offers two CUD models. Resource-based CUDs commit to specific vCPU and memory quantities in a specific region. Spend-based CUDs commit to a minimum hourly spend on certain services like Cloud SQL, BigQuery, and Cloud Run. For Indian enterprises running compute-heavy workloads in asia-south1 or asia-south2, resource-based CUDs on vCPU and memory deliver the highest savings.
Calculating CUD Sizing for India Regions
Review your Compute Engine usage in the Cloud Billing console. Identify the steady-state vCPU and memory consumption across your India-region projects. Commit to approximately 70% of that baseline. The remaining 30% will benefit from SUDs if eligible or run at on-demand rates. This approach balances savings with the flexibility to handle traffic variations.
[IMAGE: Google Cloud console showing CUD purchase interface with vCPU and memory commitment options - google cloud cud purchase india]
How Do CUDs and SUDs Work Together?
[ORIGINAL DATA] Google's billing engine applies CUDs first, then SUDs on any remaining eligible usage. If you commit 100 vCPUs via CUD but actually run 140 vCPUs throughout the month, the first 100 get the CUD rate and the extra 40 benefit from SUDs. This layered approach means Indian enterprises capture discounts on virtually all compute usage without manual intervention.
Practical Example for an Indian E-Commerce Company
Consider an Indian e-commerce platform running 200 vCPUs steady-state in asia-south1, spiking to 350 during sale events. A 150-vCPU CUD covers the reliable baseline at 57% discount (three-year). The next 50 steady-state vCPUs earn SUDs at up to 30%. Sale-event spikes run on-demand. The blended effective discount across the year might reach 40-45% compared to pure on-demand.
Avoiding CUD-SUD Overlap Confusion
SUDs never apply to CUD-covered resources. The billing engine handles this automatically. You don't need to tag or separate instances. Simply purchase CUDs for your baseline, and Google's system applies the optimal discount to each vCPU-hour. This simplicity is a genuine advantage over AWS and Azure, where commitment management requires more manual oversight.
What Does Google Cloud's Active Assist Recommender Offer?
Active Assist is Google Cloud's built-in recommendation engine, similar to Azure Advisor or AWS Compute Optimizer. Google's Active Assist documentation (2025) states that its recommendations cover idle resource cleanup, right-sizing, and CUD purchases. For Indian enterprises, the idle VM recommender and the CUD recommender are the two most actionable tools.
Idle VM and Idle Project Recommendations
Active Assist identifies VMs with consistently low CPU and network activity and recommends shutting them down. It also flags entire projects with no recent activity. For Indian development teams that spin up projects for proof-of-concept work and forget to decommission them, the idle project recommender is invaluable. Each flagged resource shows estimated monthly savings.
Right-Sizing Recommendations
The right-sizing recommender analyses CPU and memory usage to suggest smaller machine types. Like AWS Compute Optimizer, it can recommend moving between machine families, not just sizes. For Indian companies that started with N1 instances and haven't evaluated E2 or N2D options, the recommender often surfaces significant savings opportunities.
How Does Google Cloud Pricing Compare to AWS and Azure in India?
[UNIQUE INSIGHT] Google Cloud's per-second billing (minimum 1 minute) is more granular than AWS's per-second (minimum 1 minute) or Azure's per-minute billing. For Indian enterprises running short-lived batch jobs or bursty containers, this granularity matters. Google's automatic SUDs also reduce the urgency to purchase commitments immediately, giving new cloud users time to understand their usage before committing.
GCP's Price Advantage for Specific Workloads
Google Cloud often prices data egress lower than AWS for inter-region transfers within India. BigQuery's on-demand pricing model, where you pay per query rather than for always-on clusters, can be dramatically cheaper for analytics workloads with variable query volumes. Indian enterprises evaluating Google Cloud should benchmark their specific workload mix, not just VM prices.
The Multi-Cloud Cost Consideration
Many Indian enterprises run multi-cloud setups. Understanding each provider's discount mechanism prevents over-spending. Google's automatic SUDs mean less management overhead. AWS's Savings Plans offer cross-service flexibility. Azure's Hybrid Benefit stacks with reservations. The right strategy depends on workload distribution and internal expertise. Our cloud pricing comparison covers this in detail.
[CHART: Table - Feature comparison of GCP CUDs/SUDs vs AWS Savings Plans vs Azure Reservations for India regions - compiled from provider documentation 2026]
What Are Common GCP Cost Mistakes for Indian Companies?
The most frequent mistake Indian enterprises make on Google Cloud is ignoring CUDs entirely because SUDs "already save money." While SUDs provide a 30% baseline discount, CUDs add 20-27% more on top of that. According to the FinOps Foundation (2025), only 35% of Google Cloud customers purchase CUDs, leaving significant savings on the table. Don't let automatic discounts create a false sense of optimisation.
Ignoring Preemptible and Spot VMs
Google Cloud Spot VMs (formerly Preemptible) offer 60-91% discounts for workloads that can tolerate interruption. Batch processing, CI/CD, data pipelines, and rendering workloads are natural fits. Indian companies running nightly batch jobs on on-demand instances are overpaying dramatically. Spot VMs in asia-south1 deliver the same compute at a fraction of the cost.
Not Using Labels for Cost Allocation
Google Cloud uses labels (equivalent to AWS tags or Azure tags) for cost categorisation. Without consistent labels, cost attribution to teams, projects, or business units is impossible. Implement a labelling policy before scaling your Google Cloud deployment. The billing export to BigQuery combined with labels enables powerful cost analytics.
Frequently Asked Questions
Do Sustained Use Discounts apply automatically in India regions?
Yes. SUDs apply automatically to eligible machine types (N1, N2, N2D, C2) running in any region, including asia-south1 (Mumbai) and asia-south2 (Delhi). No action or configuration is needed. The discounts appear on your bill at the end of each month based on actual usage hours.
Can I cancel a Committed Use Discount?
No. CUDs are non-cancellable and non-refundable. Once you commit, you pay the committed amount for the full one-year or three-year term regardless of actual usage. This makes proper sizing critical before purchase. Start with a conservative commitment and add more CUDs as you gain confidence in your baseline.
Which is better for Indian startups: CUDs or SUDs?
Startups with unpredictable growth should rely on SUDs initially. They're automatic and risk-free. Once your workload stabilises for 3-6 months and you can predict baseline usage, layer CUDs on top. This staged approach avoids over-commitment during the rapid-growth phase when resource needs change frequently.
Does Google Cloud offer free cost management tools?
Yes. Cloud Billing reports, Active Assist recommender, and billing export to BigQuery are all free. The BigQuery export is particularly powerful, letting you run custom SQL queries against your billing data for analysis not available in the console. Combined with cloud cost optimization services practices, these tools give Indian enterprises full visibility.
Optimising Google Cloud Costs for Indian Workloads
Google Cloud's discount model rewards both passive use (through automatic SUDs) and active commitment (through CUDs). Indian enterprises should take advantage of both layers. Let SUDs cover your variable and short-term workloads. Purchase CUDs for stable, long-running baseline compute. Use Active Assist to continuously identify idle resources and right-sizing opportunities.
The combination of per-second billing, automatic SUDs, and flexible CUDs makes Google Cloud a compelling option for Indian enterprises that want cost optimisation with less management overhead. Start by enabling billing export to BigQuery. Review Active Assist recommendations weekly. And when your baseline is clear, commit through CUDs to lock in the deepest discounts available.
For hands-on delivery in India, see Opsio's google cloud platform practice.
Related Services
About the Author
Country Manager, India at Opsio
AI, Manufacturing, DevOps, and Managed Services. 17+ years across Manufacturing, E-commerce, Retail, NBFC & Banking
Editorial standards: This article was written by a certified practitioner and peer-reviewed by our engineering team. We update content quarterly to ensure technical accuracy. Opsio maintains editorial independence — we recommend solutions based on technical merit, not commercial relationships.