How We Save You 40% on AVD Costs
One of the most common questions we get is: “How can you really save 40% on AVD costs?”
It’s a fair question. Azure Virtual Desktop is already more cost-effective than traditional VDI solutions, so claiming further significant savings requires explanation. Let us break down exactly how The Smart Scaler achieves these results.
Understanding AVD Costs
Before we dive into savings, let’s understand where your money goes in a typical AVD deployment:
- Compute costs — The VMs running your session hosts (typically 60-70% of total cost)
- Storage costs — OS disks, user profile disks, and data storage
- Network costs — Egress charges and bandwidth
- Licensing — Windows and Microsoft 365 licences
The Smart Scaler focuses primarily on compute costs — the biggest line item and the one with the most optimisation potential.
The Three Pillars of Cost Optimisation
1. Right-Sizing Based on Actual Usage
Most AVD deployments are provisioned based on peak capacity requirements. But peak usage typically occurs for only a few hours each day.
Traditional approach:
- Provision 100 session hosts
- Run all 100 session hosts 24/7
- Pay for 100 × 24 × 30 = 72,000 VM-hours per month
Smart Scaler approach:
- Analyse actual usage patterns
- Scale from 20 hosts (night) to 100 hosts (peak) to 60 hosts (standard)
- Pay for actual demand, typically 35,000-45,000 VM-hours per month
Savings: 35-50% on compute alone
2. Holiday and Calendar Awareness
This is where many organisations leave money on the table. Consider:
- Bank holidays — Usage drops 70-80% on public holidays
- Christmas period — Many organisations see minimal usage
- Company events — All-hands meetings, training days, etc.
The Smart Scaler includes:
- Pre-loaded public holiday calendars for 50+ countries
- Custom date support for company-specific events
- Regional awareness for multi-location organisations
Savings: 5-10% additional reduction
3. Proactive Scaling vs Reactive Scaling
The difference between proactive and reactive scaling is crucial for both cost savings and user experience:
Reactive scaling (traditional autoscaling):
- Users log in
- System detects increased demand
- New VMs start spinning up (3-5 minutes)
- Users wait or experience degraded performance
Proactive scaling (The Smart Scaler):
- Historical data predicts 9am Monday rush
- VMs start spinning up at 8:45am
- Users log in to immediately available resources
- Happy users, optimal resource usage
Proactive scaling also enables smarter shutdown decisions. Rather than keeping VMs running “just in case,” we can confidently scale down knowing demand won’t spike unexpectedly.
Real-World Example
Let’s look at a typical organisation’s numbers:
Before The Smart Scaler:
- 50 session hosts, D4s v3 (£0.192/hour)
- Running 24/7/365
- Annual cost: 50 × £0.192 × 24 × 365 = £84,096
After The Smart Scaler:
- Same 50 session hosts at peak
- Intelligent scaling based on usage
- Average utilisation: 55% of baseline
Actual usage:
- Weekdays 7am-7pm: Full capacity (50 hosts)
- Weekdays 7pm-7am: Reduced (15 hosts)
- Weekends: Minimal (8 hosts)
- Holidays: Minimal (5 hosts)
Annual cost: Approximately £49,000
Total savings: £35,000+ per year (42%)
Additional Cost-Saving Features
Beyond basic scaling, The Smart Scaler includes:
Spot Instance Integration
For non-critical workloads, use Azure Spot VMs for up to 90% discount on compute costs.
Reserved Instance Optimisation
We help you identify the right balance of reserved vs on-demand capacity.
Cost Reporting and Forecasting
Understand your spending trends and forecast future costs with our analytics dashboard.
Getting Started
Ready to see what you could save? Here’s how to get started:
- Sign up for a free account at thesmartscaler.com
- Connect your Azure environment (read-only access initially)
- Review your personalised savings estimate
- Enable scaling on your chosen host pools
Most organisations see their first savings within the first week.
Questions about how The Smart Scaler can help your organisation? Contact our team for a personalised demo.