Description

This curriculum spans the technical and organisational complexity of multi-workshop advisory engagements, addressing the full lifecycle of capacity forecasting from data collection and model selection to governance and adaptive management in hybrid and cloud environments.

Module 1: Foundations of Capacity Resource Forecasting

Selecting between reactive and proactive capacity planning based on historical incident patterns and service-level agreement (SLA) requirements.
Defining resource units (e.g., CPU core-hours, storage TB-months, FTE labor days) for cross-system comparability in forecasting models.
Mapping business drivers (e.g., product launches, seasonal demand, regulatory changes) to technical capacity requirements.
Establishing data retention policies for performance metrics to balance forecasting accuracy with storage costs.
Choosing appropriate forecasting horizons (short-term vs. long-term) based on procurement lead times and budget cycles.
Integrating capacity forecasting with IT financial management to align resource projections with cost models.

Module 2: Data Collection and Performance Baseline Establishment

Configuring monitoring tools to collect granular utilization data without introducing performance overhead on production systems.
Normalizing performance data across heterogeneous environments (cloud, on-prem, hybrid) for consistent baseline development.
Determining sampling intervals and aggregation methods that preserve signal fidelity while reducing data volume.
Identifying and filtering outlier data points caused by transient spikes or monitoring anomalies.
Establishing service-specific baselines that account for diurnal, weekly, and seasonal usage patterns.
Validating baseline accuracy through back-testing against known historical capacity events.

Module 3: Forecasting Methodologies and Model Selection

Choosing between time-series models (e.g., ARIMA, exponential smoothing) and regression-based approaches based on data availability and business drivers.
Implementing trend decomposition to separate growth, seasonality, and irregular components in utilization data.
Deciding when to apply machine learning models versus statistical methods based on data volume and interpretability requirements.
Calibrating model parameters using walk-forward validation to avoid overfitting to historical data.
Handling missing or incomplete data in forecasting inputs through imputation or model adjustments.
Documenting model assumptions and limitations for auditability and stakeholder transparency.

Module 4: Scenario Planning and Demand Simulation

Developing high, medium, and low demand scenarios based on business growth projections and market volatility.
Simulating the impact of infrastructure consolidation or migration projects on future capacity needs.
Modeling the effect of application refactoring or optimization initiatives on resource consumption trends.
Assessing the capacity implications of disaster recovery failover configurations during peak loads.
Quantifying the resource overhead of security controls (e.g., encryption, logging) in capacity projections.
Coordinating with application teams to incorporate planned feature releases into demand forecasts.

Module 5: Resource Allocation and Provisioning Strategies

Setting threshold levels for auto-scaling policies that balance responsiveness with cost efficiency.
Deciding between over-provisioning and just-in-time scaling based on workload criticality and procurement constraints.
Allocating buffer capacity for shared services while preventing resource hoarding by individual teams.
Implementing chargeback or showback mechanisms to influence resource consumption behavior.
Coordinating hardware refresh cycles with forecasted demand to avoid mid-cycle capacity shortfalls.
Managing cloud reserved instance commitments in alignment with long-term utilization forecasts.

Module 6: Governance and Cross-Functional Integration

Establishing a capacity review board to adjudicate conflicting resource requests across business units.
Integrating capacity forecasts into change management processes to assess impact of new deployments.
Defining escalation paths for forecast deviations exceeding predefined tolerance thresholds.
Aligning capacity planning cycles with enterprise budgeting and capital planning timelines.
Documenting capacity decisions and rationale for compliance with internal audit requirements.
Coordinating with procurement teams to ensure vendor lead times are reflected in provisioning schedules.

Module 7: Monitoring Forecast Accuracy and Adaptive Management

Calculating forecast error metrics (e.g., MAPE, RMSE) to evaluate model performance over time.
Triggering forecast model retraining based on sustained prediction drift beyond operational thresholds.
Adjusting forecasts in response to unplanned events (e.g., mergers, outages, policy changes).
Conducting root cause analysis on forecast misses to improve data inputs or modeling assumptions.
Updating capacity plans in response to changes in service architecture or technology stack.
Archiving outdated forecasts and models to maintain clarity in decision records.

Module 8: Advanced Topics in Distributed and Cloud Environments

Forecasting capacity across multi-cloud environments with varying performance characteristics and pricing models.
Modeling the impact of network latency and data egress costs on distributed workload placement.
Accounting for shared tenancy effects in public cloud environments when predicting performance at scale.
Projecting container density and orchestration overhead in Kubernetes-based platforms.
Estimating cold-start and warm-pool requirements for serverless computing workloads.
Integrating sustainability goals (e.g., carbon footprint) into capacity decisions and forecasting KPIs.