Maintenance Release

CloudPouch 1.49.0 modernizes the desktop runtime and release foundation.

• Electron 43.0.0 with Chromium 150; builds use Node.js 24.18.0.
• Packaging and updates use electron-builder 26.15.3 and electron-updater 6.8.9.
• Improved macOS signing, notarization, and release reliability.
• Fixed loading animation compatibility.
• Minimum supported systems are macOS 12 and Windows 10.

Bug Fixes & Maintenance

• Fixed the savings percentage calculation in the EBS Volume Optimization insight and improved the accompanying summary message so it's clearer what the figure represents.

Bug Fixes & Maintenance (this release sharpens EBS rightsizing for idle volumes)

💽 Idle EBS Volumes Now Recommend Cold/Throughput HDD

CloudPouch's EBS Volume Type Advisor now spots idle, over-provisioned volumes and recommends moving them to Cold HDD (sc1) or Throughput Optimized HDD (st1) — the cheapest storage tier — instead of only trimming provisioned IOPS and throughput.

• Previously, a volume with almost no I/O was held back by a throughput safeguard meant to keep HDD away from busy random-access workloads. That safeguard no longer applies to effectively idle volumes (peak ≤ 50 IOPS), which have no such workload.
• The advisor still keeps HDD away from active low-throughput volumes, where random small-I/O latency matters.
• Example: a 2,580 GiB GP3 volume sitting idle but provisioned for 6,450 IOPS and 250 MiB/s was only being right-sized to GP3 baseline (~22.00 USD/month). It now correctly recommends SC1, cutting storage from ~228.00 USD to ~38.00 USD — about 190.00 USD/month per volume.

Why it matters: Storage capacity, not IOPS, is usually the largest line on an EBS bill. Surfacing the HDD option for genuinely idle volumes turns a modest provisioning trim into a major monthly saving — exactly where over-provisioned, forgotten volumes hide.

Features (this release is all about making EBS over- and under-provisioning visible — and giving you the cheapest fix)

⚙️ New Cost Insight: EBS Volume Type Advisor

CloudPouch now analyzes every attached EBS volume against 14 days of real CloudWatch usage and tells you the most cost-effective volume type and provisioning for the actual workload — across gp2, gp3, io1, io2, st1, sc1, and Magnetic.

• Compares provisioned IOPS and throughput against observed peak and average usage from CloudWatch, rather than guessing from volume type alone.
• Recommends the cheapest feasible volume type, including a move to Throughput Optimized HDD (st1) or Cold HDD (sc1) when the access pattern is sequential and IOPS are low.
• Right-sizes within the same type — for example, lowering provisioned IOPS or throughput on a gp3/io1/io2 volume without changing the type.
• Surfaces gp2 → gp3 savings even before CloudWatch data is available, since gp3 is cheaper per GB and can be configured to match gp2's guaranteed performance.
• Flags legacy Magnetic (standard) volumes to migrate away from.
• Each row shows the attached EC2 instance, observed IOPS and throughput (max / avg), the recommended type and provisioning, and a stacked Current Cost breakdown (storage + provisioned IOPS + provisioned throughput).

Why it matters: EBS is billed on three separate dimensions — capacity, provisioned IOPS, and provisioned throughput — so it is easy to keep paying for performance a workload never uses. The advisor turns two weeks of usage into a single, concrete "switch to X" action with the exact monthly difference, and the section header brags about how many volumes were reviewed even when most are already optimal.

📈 Cheapest-Path Performance Upgrades (cost-insight)

When a volume is genuinely saturated, CloudPouch now recommends the least expensive way to add headroom instead of always jumping to a premium type.

• A volume is flagged as near capacity when peak usage reaches roughly 85% of what it can deliver.
• CloudPouch sizes for peak usage plus a 20% safety buffer, then picks the cheapest configuration that satisfies it.
• Raising provisioned IOPS/throughput on the current type is compared head-to-head against migrating to a higher tier — and the cheaper option wins.
• For example, a gp3 volume sitting at its 3,000 IOPS baseline but briefly peaking just above it is told to raise gp3 IOPS for about 5.00 USD/month rather than migrate to io2 for roughly 268.00 USD/month — same headroom, a fraction of the cost.

Why it matters: "Fix performance" advice that defaults to the most expensive option quietly erodes both trust and budgets. Recommending the cheapest sufficient configuration keeps performance and cost aligned.

How to Use It

1. Open CloudPouch and select your AWS profile.
2. Choose Current Month or Previous Month.
3. Go to EC2 - Other and run Check Cost Insights.
4. Open the EBS optimization / rightsizing section.
5. Review the per-volume recommendations, estimated monthly savings, and upgrade suggestions — and export everything to TSV when you need to share it.

Other Features

• Cost breakdown in the table: Current Cost splits into storage, provisioned IOPS, and provisioned throughput with a total, so you can see exactly which dimension drives the bill.
• Attached instance at a glance: each volume shows the EC2 instance it is attached to, alongside region and age.
• Clear HDD sizing: st1/sc1 recommendations show the size-derived throughput they will deliver — and when a volume needs to grow to reach it — with a short "throughput scales with size" note.
• Idle volumes are right-sized, not ignored: volumes with zero I/O over the window are treated as real zero-usage data and right-sized to baseline instead of being dropped.

Bug Fixes & Maintenance

• CloudPouch now distinguishes a genuinely idle volume from one whose CloudWatch metrics could not be fetched, dramatically increasing how many volumes receive a recommendation instead of an "insufficient data" message.
• Removed the older, narrower GP2 → GP3 section and the standalone provisioned-IOPS section; both are fully superseded by the Volume Type Advisor, which covers the same volumes with more options and better pricing logic.
• Recommendation quality is validated against realistic volume and CloudWatch-metric fixtures, reducing the risk of noisy or misleading insights.

Features

This is mostly a maintenance release, with a few small but useful improvements for CloudPouch users.

🔁 Compare Reports With Merged Region Usage Types

Compare reports now include a Merge regions toggle for usage-type rows.

• Region-prefixed usage types such as EU-EBS:VolumeUsage.gp3 and USW2-EBS:VolumeUsage.gp3 can now be collapsed into one logical usage type.
• The toggle is enabled by default, so Compare reports start with a cleaner and easier-to-read view.

Why it matters: Merging regional usage types makes month-to-month changes easier to understand and helps you focus on the real cost driver instead of regional billing noise.

🧭 Permission Issues in Cost Insights

Cost Insights now shows IAM permission and opt-in problems instead of silently returning empty results.

• AWS API failures such as AccessDenied, OptInRequiredException, and similar errors are collected and shown in the UI.
• After Bulk Cost Insights, a collapsible Permission issues panel lists the affected service, region, API operation, and AWS error code.

Why it matters: Empty insight tables can mean either “nothing to optimize” or “CloudPouch was not allowed to check.” This release makes the difference visible, so you can fix IAM coverage before trusting the absence of savings opportunities.

💾 Conditional Snapshot Evaluation

EC2 Other now makes EBS snapshot analysis explicit with a new Evaluate EBS snapshots toggle.

• Snapshot evaluation is disabled by default.
• You can enable it before running Cost Insights when you want CloudPouch to scan for EBS snapshot cleanup opportunities.
• When snapshot evaluation is skipped, the results clearly state that snapshots were not evaluated instead of implying that no snapshot savings were found.

Why it matters: Some AWS accounts have very large snapshot inventories, and a full scan can take a long time. CloudPouch now avoids unexpected long-running analysis while still giving you the option to inspect snapshot waste.

Other Improvements

• CloudWatch metric batching: Cost Insights can fetch CloudWatch metrics more efficiently, especially in larger AWS accounts.

Bug Fixes & Maintenance

📊 Scoped Billing Reports You Can Trust

CloudPouch now uses a unified billing model across all places where costs are displayed, making scoped cost analysis more reliable and faster.

This long-overdue internal improvement gives CloudPouch a stronger foundation for representing cost data consistently across reports, scoped views, and Cost Insights.

Other

• Fixed Cost Insights behavior for large accounts where CloudWatch metric fetching or S3 insight processing could fail.
• Migrated EC2 and RDS pricing data to compact JSON assets.
• Added more tests for billing scope behavior, usage-type merging, AWS error classification, permission issue collection, CloudWatch metric batching, and EC2 Other snapshot evaluation.

Features

🆕 Bedrock Cost Breakdown by IAM Principal Tag (cost-insight)

CloudPouch now shows your Amazon Bedrock / Claude spend split by IAM principal tags — directly inside the existing service drill-down. When a single Bedrock line item used to hide many teams, applications, and chatbots, you can now see exactly who is burning the tokens.

This builds on the recently announced AWS IAM Principal-Based Cost Allocation for Amazon Bedrock, which surfaces caller identity (iamPrincipal/* tags) for every Bedrock API call in Cost Explorer and CUR 2.0. CloudPouch reads those tags from Cost Explorer and turns them into something you can actually scan in seconds.

• A new Costs by IAM tag tab appears in the service drill-down for every Claude … (Amazon Bedrock Edition) model present in your bill.
• A tag selector lets you switch between any activated IAM principal tag key.

• A stacked bar chart shows daily or monthly cost per tag value, with a tooltip that reports both the per-tag cost and the total.
• Per-region filtering is respected, so you can isolate which region (and which team in that region) is driving the spend.
• Untagged calls are not hidden — they are grouped under No tag key: <tagName> so you can spot gateway roles, untagged service accounts, or principals you forgot to tag.
• Series under 0.01 USD are filtered out automatically, and remaining tag values are sorted from highest to lowest cost so the biggest spenders sit at the top of the legend.
• Tax, refunds, credits, and EDP discounts are excluded from the IAM-tag view, matching how the rest of CloudPouch reports clean usage cost.

Why this matters: Bedrock used to appear in your bill as a single foundation-model line item, so questions like “is the customer-facing chatbot or the internal summarizer driving this spend?” required hand-rolled CUR queries. CloudPouch now answers them visually, by IAM principal, without any custom tooling — and makes shared-gateway architectures (where many teams reuse one role) immediately obvious because their spend collapses into one bar.

How to Use It

1. In AWS, tag the IAM users and roles that call Bedrock with low-cardinality keys such as department, costCenter, team, or project.
2. In the Billing and Cost Management → Cost Allocation Tags console, activate those tags (they appear as iamPrincipal/<key>).
3. In Data Exports, enable Include caller identity (IAM principal) allocation data on your CUR 2.0 export. Allow up to 24 hours for AWS to backfill the data.
4. Open CloudPouch, pick the AWS profile, select Current Month or Previous Month, and expand any Claude … (Amazon Bedrock Edition) row.
5. Switch to the Costs by IAM tag tab, pick a tag key from the dropdown, and review the per-principal breakdown.

Other Features

• AmazonQ icon: Amazon Q now renders with its proper service icon in the AWS services table instead of falling back to the generic placeholder.
• BUSINESS SUPPORT+ icon coverage: The new BUSINESS SUPPORT+ line item is now recognized and rendered with the matching support-tier icon.

Bug Fixes & Maintenance

• Cost Explorer calls for the new IAM-tag breakdown are issued sequentially per service and tag key to avoid Cost Explorer throttling, so cached refreshes stay reliable on accounts with many activated IAM principal tag keys.
• The IAM-tag fetch fails soft: if Cost Explorer rejects a single (service, tagKey) combination, the rest of the breakdown still loads and the error is logged per service/tag instead of breaking the whole drill-down.
• Recommendation and chart logic for the new feature is covered by unit tests against realistic Cost Explorer fixtures, reducing the risk of misleading per-tag totals.

Features

📂 New Cost Insight: Amazon EFS (cost-insight)

CloudPouch now includes a dedicated Cost Insight for Amazon EFS, so file system spend stops getting lost inside generic storage line items.

• List all EFS file systems with size, storage class mix (Standard / IA / Archive), throughput mode, performance mode, lifecycle policy state, and monthly cost.
• Per-region pricing tables and EFS-specific usage type interpretation, so the cost shown for each file system actually matches what AWS bills.
• Recommendations point out file systems with no lifecycle policy, misaligned storage class usage, no mount targets, and other configuration patterns that quietly leak money.
• The section degrades gracefully on sparse or noisy billing inputs — partial data is annotated rather than swallowed silently.

💰 Why it matters: EFS bills hide a lot of detail behind a single number. Lifecycle policies, infrequent-access transitions, and provisioned throughput can quietly cost more than the storage itself. CloudPouch now makes each file system, its class mix, and its likely cleanup action visible at a glance.

λ New Cost Insight: AWS Lambda (cost-insight)

CloudPouch now includes a full Cost Insight for AWS Lambda, with regional pricing tables, per-function estimates, and concrete optimization recommendations.

• List every function with memory size, average duration, invocation count, cost, and (when present) Provisioned Concurrency configuration.
• Per-region GB-second pricing tables drive accurate cost estimates per function, including the Provisioned Concurrency surcharge.
• Recommendations flag over-provisioned memory, idle Provisioned Concurrency, and other patterns where right-sizing or removing PC unlocks immediate savings.
• A separate path keeps Lambda functions with no recorded cost visible — useful for catching abandoned or stale functions that don't appear in normal cost reports.
• Small but non-zero Lambda rows stay visible in the main services table so low-cost-but-actionable Lambda spend isn't filtered away.

💰 Why it matters: Lambda cost is driven by memory configuration and Provisioned Concurrency, which are easy to set once and forget for years. CloudPouch now turns those configuration choices into a sortable, function-level cost story you can actually act on.

🔁 Auto Scaling Group Cost Recommendations (cost-insight)

EC2 Cost Insights now goes beyond just grouping Auto Scaling Group members and actively recommends ASG-level cost optimizations.

• ASG-aware right-sizing recommendations with detailed messaging that explains why a given instance type is suggested at the group level instead of per instance.
• A direct Auto Scaling Group details link opens the ASG in the AWS console for fast investigation.
• Region is now part of the grouping key, so an ASG name that exists in multiple regions stops collapsing into one misleading row.
• Improved cost calculation for ASG members handles missing-pricing edge cases gracefully instead of producing zero-cost rows.

💰 Why it matters: ASGs are the place where compute cost actually accumulates, but per-instance recommendations don't fit fleet behavior. CloudPouch now reasons at the group level and surfaces the savings story you would otherwise have to reconstruct by hand.

Other Features

• DocumentDB Graviton-aware savings: Potential savings now name the suggested Graviton instance type, so the recommendation is directly actionable instead of abstract.
• Glue failed job cost: The Glue insights section computes and displays cost for failed job runs (Ireland-only pricing, experimental), making failure cost visible alongside successful runs.
• EBS analysis polish in EC2 Other: Snapshot analysis now uses concurrency control and better error handling against the AWS API, so large accounts complete reliably instead of hitting throttling cliffs.
• More EC2 instance types: M7I-Flex Large, R7A Medium, and R7A Large added to the recommendation catalog with full specifications.
• Better Glue and CloudWatch usage descriptions: New descriptions for Glue Flex execution class and interactive sessions, plus improved generic matching for CloudWatch usage types — fewer "unknown usage type" rows in service drill-downs.
• Claude and Bedrock service icons: Both services now render with proper colored icons in the AWS services table instead of the generic placeholder.

Bug Fixes & Maintenance

• RDS Extended Support enrollment is now reported consistently across the current-month, previous-month, and future-enrollment views — eliminating cases where the same cluster appeared enrolled in one panel and not another.
• Lambda cost calculations handle null pricing entries gracefully, so insights still render in regions or for runtimes with incomplete pricing data instead of crashing the section.

Features (this release is all about usability and keyboard speed)

🚀 Bulk Cost Insights (Cmd/Ctrl + Enter) (usability-first)

Start bulk Cost Insights Analysis to generate optimization recommendations for all services (only available for Current Month or Previous Month timeframes)

• Press Cmd/Ctrl + Enter (or use Analysis → Start Cost Insights Analysis) to analyze every eligible service in one go.
• Live service badges show Done, Failed, or Queued so you always know what finished and what needs attention—even after the run.
• Smart toasts explain why analysis isn’t available (wrong timeframe, Org view, job already running), cutting trial-and-error clicks.

Why it matters: This is the fastest path from “open CloudPouch” to “I have answers.” Fewer clicks, more insight.

🔎 Quick Profile Switcher (Cmd/Ctrl + K) (usability-first)

Find and switch AWS profiles with a modern command palette.

• Press Cmd/Ctrl + K to open Search AWS profiles…; type to filter, ↑/↓ to navigate, Enter to select, Esc to close.
• Search matches profile name making large lists instantly manageable.
• Works with named profiles, SSO, and role-chaining from your ~/.aws/config and credentials sources.

Why it matters: Multi-account work is now frictionless—jump profiles in seconds without leaving the keyboard.

⌨️ Keyboard Shortcuts Cheat Sheet (usability-first)

Your productivity, now muscle-memory friendly.

• Open Help → Keyboard Shortcuts to view the in-app overlay anytime.

Why it matters: CloudPouch is now keyboard-centric - you’ll move faster with fewer context switches.

Other Features

• Snapshot Cleanup Playbooks for EC2: At-a-glance totals for snapshot counts, archiveable volume, and estimated monthly savings, plus per-volume “delete vs. archive” guidance and stakeholder-ready breakdowns.
• Smarter S3 Storage Guidance: Trustworthy tallies that exclude uninspectable buckets; an engine that weighs lifecycle rules vs. Intelligent-Tiering to surface the single next best action; security gaps remain visible alongside savings.

Bug Fixes & Maintenance

• RDS snapshot reports now show accurate spend for previous-month reviews, eliminating overstated savings in historical analyses.
• IPC listeners are cleaned up when layouts change, keeping long-running sessions responsive and avoiding ghost triggers.

🎬 I’ve recorded a short demo showing it all in action - check it out below 👇
https://www.linkedin.com/posts/pawelzubkiewicz_aws-finops-costoptimization-activity-7381267858103717888-V8KI

Features

⚙️ New Cost Insight: Amazon EKS Extended Support Detection

CloudPouch now includes a dedicated Cost Insight for Amazon EKS, giving you full visibility into your EKS clusters and their lifecycle status.

With this update, CloudPouch can:

• 🧾 List all EKS clusters, along with region, version, and monthly cost.
• 🏷️ Highlight clusters in Extended Support mode, which incur an additional 365.00 USD per month if running on deprecated Kubernetes versions.
• 🔎 Show when each cluster entered Extended Support and estimate historical and monthly support charges.

This helps you identify and upgrade outdated clusters before they silently start generating high ongoing costs.

💰 Why It Matters

AWS charges hefty support fees for EKS clusters running on unsupported Kubernetes versions. These charges are often hidden in the total EKS bill.
CloudPouch makes them visible, traceable, and actionable.

If you're paying 438.00 USD for a cluster that should cost just 73.00 USD, this feature will tell you exactly why.

Bug Fixes & Maintenance

• none

Features

📊 Improved Compare chart

Month-to-month comparison charts now embed per-service % deltas right inside each bar, so you can instantly spot which services went up or down, and by how much.

• A bold green arrow highlights the overall savings, making the scale of change crystal clear.
• There is also a symmetric red arrow in case your costs went up.

💰 Why It Matters

A screenshot is all it takes to showcase the impact of your FinOps initiatives. The chart speaks for itself - perfect for leadership updates, slide decks, or celebrating wins with the team.

Bug Fixes & Maintenance

• none