Description

This curriculum spans the technical, operational, and governance dimensions of capacity planning, comparable in scope to a multi-workshop program embedded within an enterprise’s internal reliability engineering function, addressing real-world decisions from infrastructure sizing to incident response.

Module 1: Defining Capacity Requirements and Demand Forecasting

Selecting between time-series forecasting models and regression-based demand projections based on historical data availability and business volatility.
Establishing service-level thresholds that trigger capacity planning reviews, such as sustained CPU utilization above 75% for 14 consecutive days.
Integrating input from sales, product, and finance teams to align capacity forecasts with projected customer growth and feature launches.
Deciding whether to use peak, average, or percentile-based metrics (e.g., 95th percentile) when sizing infrastructure needs.
Adjusting forecast models to account for seasonality, such as end-of-quarter surges or holiday traffic in e-commerce platforms.
Documenting assumptions in demand models to enable auditability and recalibration during post-mortems or capacity incidents.

Module 2: Infrastructure Sizing and Resource Allocation

Choosing between vertical and horizontal scaling strategies based on application architecture and failover requirements.
Calculating redundancy requirements for high-availability systems, including active-passive versus active-active configurations.
Allocating reserved versus on-demand compute instances based on workload predictability and cost tolerance.
Determining memory-to-CPU ratios for database workloads using query execution patterns and buffer pool requirements.
Right-sizing storage tiers (SSD vs. HDD) based on IOPS requirements and data access frequency.
Validating virtual machine or container density limits to prevent noisy neighbor issues in shared environments.

Module 3: Performance Baselines and Monitoring Integration

Defining baseline performance metrics during normal operations to detect capacity degradation over time.
Configuring monitoring thresholds that distinguish between transient spikes and sustained capacity pressure.
Integrating capacity metrics into existing observability platforms without overloading data ingestion pipelines.
Selecting key performance indicators (KPIs) per system tier—such as queue depth for message brokers or p99 latency for APIs.
Automating baseline recalibration after major system changes, such as software upgrades or architectural refactoring.
Correlating capacity metrics with business events (e.g., marketing campaigns) to improve predictive accuracy.

Module 4: Capacity Modeling and Scenario Simulation

Building what-if models to evaluate the impact of doubling user load on current database connection pools.
Simulating failure scenarios to assess spare capacity availability in remaining nodes during outages.
Using load testing results to validate model assumptions before committing to hardware procurement.
Modeling the effect of data retention policies on storage growth and backup window expansion.
Comparing cost and performance trade-offs between cloud bursting and permanent on-premises expansion.
Updating simulation parameters quarterly to reflect changes in application efficiency or user behavior.

Module 5: Governance and Change Control in Capacity Decisions

Requiring capacity impact assessments for all change requests involving high-resource services.
Establishing approval workflows for capacity expansions that exceed predefined budget or risk thresholds.
Documenting capacity decisions in a centralized repository to support audit and compliance requirements.
Enforcing capacity review gates in the release management process for major application deployments.
Assigning ownership for capacity health at the service or application level to ensure accountability.
Coordinating capacity change schedules with maintenance windows to minimize operational disruption.

Module 6: Cloud and Hybrid Environment Capacity Strategies

Implementing auto-scaling policies with cooldown periods to prevent thrashing during traffic oscillations.
Managing cross-region capacity dependencies in multi-cloud architectures to avoid single points of failure.
Tracking reserved instance utilization to avoid underuse penalties and optimize renewal cycles.
Designing egress cost controls in cloud environments where data transfer impacts capacity economics.
Aligning cloud provider quotas with projected needs and initiating increase requests before constraints impact operations.
Integrating cloud cost APIs into capacity dashboards to expose financial implications of resource decisions.

Module 7: Capacity Optimization and Right-Sizing Initiatives

Conducting quarterly resource utilization reviews to identify and decommission underused instances.
Applying container resource limits and requests based on observed usage, not default configurations.
Renegotiating data center power and cooling SLAs when consolidating or retiring physical servers.
Implementing database archiving strategies to reduce active dataset size and improve query performance.
Using A/B testing to validate performance impact after downsizing over-provisioned systems.
Standardizing instance types across environments to simplify forecasting and reduce management overhead.

Module 8: Incident Response and Capacity-Related Outages

Activating pre-defined surge capacity protocols during unexpected traffic spikes or denial-of-service events.
Executing failover to secondary systems when primary capacity thresholds are breached.
Documenting root cause of capacity-related incidents to prevent recurrence through design changes.
Temporarily throttling non-critical services to preserve capacity for core business functions.
Engaging procurement teams on emergency hardware or cloud credits when expansion timelines are compressed.
Conducting blameless post-mortems to evaluate whether monitoring, forecasting, or governance gaps contributed to the incident.