Per-Second Billing: A Comparative Study

Understanding how different cloud providers charge for their services is essential for any software engineer. Let's delve into the per-second billing practices of Azure and Google Cloud, and explore how this approach impacts developers and businesses.

Azure's Approach to Per-Second Billing

• Introduction in Late 2019: Azure began offering second-wise charges without long-term commitments since late 2019.
• Not for All Instances: Azure's per-second billing is not available for all instances and mainly focuses on container-based instances.
• Billing Every Minute: Despite the per-second approach, many instances are still billed every minute.
• Focus on Containers: Microsoft's decision to focus on containers is due to their belief that per-second billing is more suitable for containerized environments.

Google Cloud's Journey to Per-Second Billing

• Evolution from Per-Hour to Per-Minute: Google Cloud initially billed per hour, then shifted to per minute, similar to AWS.
• Following AWS's Footsteps: When AWS moved to per-second billing, Google Cloud saw an opportunity and followed suit.
• A Better Offering than AWS?: Google Cloud's per-second billing is considered superior to AWS, as it is available for all VM-based instances, not just Linux-based ones.
• Enhancements in Invoicing: Google Cloud has refined their invoicing capacities to operate on the granularity of seconds, reducing the burden on consumers.

Quick Fact

• Billing in Seconds: Instances are billed by the number of seconds. If a virtual machine operates between 30 seconds and 1 minute, it is still billed for 1 minute of usage.

Serverless Services

• Providers: AWS Lambda, Azure Functions, and Google Cloud Functions are leading serverless services in the market.
• Charging Method: These providers charge in 100-millisecond increments, aligning costs with actual computing power usage.
• Benefits for Developers: This approach allows developers to focus on coding and event triggers, while the provider handles everything else.
• Cost Savings: With serverless computing, you pay for the time your code operates, not for reserving CPU cores and RAM. This approach makes billing more aligned with actual usage, eliminating unnecessary costs.

The per-second billing strategy in cloud services is an evolving and complex subject. By understanding the different approaches and offerings from Azure and Google Cloud, developers and businesses can make informed decisions that align with their specific needs and budget. The shift towards more granular billing practices, such as serverless computing, represents an industry trend towards efficiency and cost-effectiveness that every future software engineer should be aware of.