> For the complete documentation index, see [llms.txt](https://help.pump.co/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.pump.co/pump-save/infrastructure.md). # Infrastructure The Infrastructure recommendations engine analyzes your running cloud resources and identifies opportunities to reduce waste. It surfaces specific, actionable changes you can make to your infrastructure, each with estimated monthly savings and implementation commands. This is an advisory feature. Pump does not modify your infrastructure. You review each recommendation and decide whether to apply it. ### What it analyzes Pump evaluates utilization data across the following AWS resource types: | Resource type | What Pump checks | | ----------------- | --------------------------------------------------------------------------------- | | EC2 instances | CPU utilization, memory utilization, network throughput relative to instance size | | RDS instances | CPU utilization, database connections, storage throughput | | EBS volumes | IOPS usage, throughput, attachment status | | ElastiCache nodes | CPU utilization, memory usage, connection count | For each resource, Pump classifies it into one of three categories: | Classification | Meaning | | ----------------------- | ----------------------------------------------------------------------------------------------------------------- | | **Idle** | The resource has minimal or no utilization. It may be safe to terminate. | | **Oversized** | The resource is running well below its capacity. A smaller instance type would handle the workload at lower cost. | | **Appropriately sized** | Current utilization is within a healthy range for the instance type. No action needed. | ### What a recommendation includes Each recommendation provides: 1. **The resource.** Service type, instance ID, region, and current configuration. 2. **The finding.** What Pump detected (idle, oversized, or a specific inefficiency). 3. **The recommended action.** Resize to a specific instance type, terminate, or modify a configuration. 4. **Estimated monthly savings.** The projected cost difference between current and recommended configurations. 5. **Implementation command.** A copy-pasteable AWS CLI command you can run to apply the change. ### How to use recommendations Navigate to the Infrastucture page in Pump under Save. For each recommendation: 1. Review the resource details and current utilization data. 2. Evaluate the recommended change against your operational requirements. Pump bases recommendations on utilization patterns, but lacks the workload context needed to fully assess usage trends (upcoming traffic spikes, seasonal patterns, compliance requirements). 3. If you agree with the recommendation, copy the provided CLI command and run it in your cloud environment. 4. If you disagree, dismiss the recommendation. ### How this differs from the Commitment Planner The Commitment Planner and Infrastructure Recommendations target different layers of cloud cost: | | Commitment Planner | Infrastructure Recommendations | | ----------------- | ---------------------------------------------------------- | -------------------------------------------------- | | What it optimizes | Pricing (paying less for the same resources) | Sizing (using fewer or smaller resources) | | How it works | Purchases discount instruments (RIs, SPs, CUDs) | Surfaces resize/terminate actions for you to apply | | Automation | Can execute automatically once recommendation is requested | Advisory only. You apply changes manually | | Risk profile | Financial commitment with defined term | Infrastructure change that may affect performance | Both mechanisms contribute to your overall savings. Commitment-level savings reduce the per-unit price of what you run. Infrastructure savings reduce what you run in the first place. ### Troubleshooting **Q: Why don't I see any recommendations yet?**\ A: Infrastructure recommendations require sufficient utilization history to generate meaningful findings. If you connected recently, allow a few days for data collection before recommendations appear. **Q: Why does this recommendation seem wrong for my workload?**\ A: Pump bases recommendations on observed utilization patterns. If your workload has characteristics that the utilization data doesn't capture (planned growth, periodic batch jobs, compliance minimums), dismiss the recommendation. Dismissed recommendations are excluded from future cycles. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://help.pump.co/pump-save/infrastructure.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.