Explain AWS NAT Gateway cost increases
Separate NAT Gateway-hour, processed-GB, and data-transfer charges so teams investigate the cost driver before changing network paths.
BASH4 min read
Cloud cost guides
Find answers to common questions about cloud and AI costs.
58 guides
Check AWS billing tag coverage before allocating costs
Review activation, historical gaps, and CUR 2.0 line-item values before using AWS resource tags to assign costs to teams.
Steps3 min read
Review Azure Files provisioned capacity and snapshot storage
Compare share-level utilization with snapshot storage in Azure Monitor before reviewing provisioned capacity. Keep account totals out of share-level decisions.
Steps3 min read
Check whether resource-based CUD sharing fits your billing setup
Match Compute Engine resource-based CUD scope to intended project coverage and verify shared reservation projects use the same Cloud Billing account before a move.
Steps3 min read
Review Compute Engine recommendations before resizing VMs
Review VM size suggestions against peak demand, compatibility and restart needs. Learn why estimated savings and empty recommendation lists need care.
BASH3 min read
Match Cloud Run functions to the right pricing model
Classify Cloud Run functions by documented version and console area before comparing costs. Keep pricing-family labels separate from API history and billed-SKU confirmation.
Steps3 min read
Check Blob lifecycle rules before changing access tiers
Check lifecycle rules before moving an individual Azure Block Blob out of Archive, and account for retrieval and early-deletion meters before approval.
Steps3 min read
Separate transfer costs before blaming compute or storage
Review AWS transfer usage types, Azure bandwidth categories, and Google Cloud network tiers before assigning a compute, storage, or network cost cause.
Steps4 min read
Check whether Redshift Serverless limits stop queries
Compare retained Redshift Serverless usage with capacity and RPU-hour limits, then confirm whether each action warns, logs, alerts, or stops query processing.
SQL4 min read
Review Blob Storage versions and soft-delete costs
Check whether an existing Azure Blob Inventory report covers retained versions and deleted objects, then compare its scope with Blob Capacity before planning storage changes.
Steps4 min read
Review BigQuery long-term storage billing
Check BigQuery partition tiers and last-write timestamps before forecasting storage costs. Separate logical and billable bytes without treating metadata as a full bill.
SQL3 min read
Separate AI requests from retries and evaluation work
Review AWS and Azure telemetry to separate application requests, model calls, retries, and evaluation activity before allocating AI spend.
Steps3 min read
Check Cloud SQL compute, storage, backup and network charges
Separate Cloud SQL charges from usage charts before choosing a cost change. Review one invoice month and include any separate MySQL enhanced-backup project.
Steps3 min read
Separate RDS billing meters before changing capacity
Break down Amazon RDS costs by usage type so instance hours, storage, IOPS, backups, and transfer are reviewed separately before capacity changes.
BASH3 min read
Check BigQuery autoscale charges by reservation minute
Review BigQuery autoscale slot-seconds from the last day and compare reservation limits and slot sharing before asking an owner to change capacity.
SQL3 min read
Measure AI cost per successful business outcome
Compare scoped AI spending with accepted business outcomes using a read-only worksheet that keeps retries, failed attempts and attribution gaps visible.
Steps4 min read
Review quiet SageMaker capacity before cutting it
List SageMaker endpoints and notebook instances, compare endpoint activity, and ask owners which resources can become candidates for a cost reduction.
BASH3 min read
Check Cloud Storage protections before counting deletion savings
Review Cloud Storage versioning, soft delete, and locked retention settings before treating deletion as storage savings.
Steps3 min read
Review Azure Machine Learning capacity floors
Check managed online deployment minimums and training-cluster node floors against workload needs, then use endpoint cost tags to prioritize owner reviews.
Steps3 min read
Compare Azure OpenAI token costs with provisioned capacity
Compare Azure OpenAI token charges with hourly PTU capacity costs, including idle hours, model-specific sizing, reservation coverage and unused commitments.
Steps4 min read
Check who can use your Azure commitments
Compare Azure reservation and savings plan scopes with eligible usage before widening access across an EA enrollment or MCA billing profile.
Steps3 min read
Choose models by cost per usable answer, not token price
Compare model candidates on shared tasks, quality limits and response times. Use a read-only worksheet to estimate cost per accepted result from evaluation records.
Steps3 min read
Check Cosmos DB hot partitions before reducing throughput
Compare Cosmos DB partition utilization and 429 responses before a throughput cut, then use existing logs to identify keys with concentrated demand.
KQL3 min read
Check Azure Front Door request and transfer costs
Review Standard and Premium profile fees, request counts, and transfer charges separately to identify which costs caching can reduce and which remain.
Steps3 min read