Feature Requests

The Looker Dashboards have a huge area at the top in which any global filters are placed. When placing a filter eg Cost Categories, is it necessary to have both the Cost Category filter and the cost bucket? This doubles the space used and puts useless items into the filters. When viewing on a laptop, the real estate is half the screen. This makes the rest of the Dashboards visualisations very hard to view. Scrolling down the Dashboard can also make a large panel of white appear at the bottom. Sometimes there are 3 levels of scroll bar in the Users view. There is a problem with the way Harness has embedded Looker, which is not allowing dynamic resizing in a good UX design way. Please review

It appears that the parameters by which Harness moves their Recommendations to "accepted" status is too rigid. It doesn't appear to be observing the ticket's resolution field, and is looking at the status, so anything outside of the actual "Done" status won't be accepted We'd like Harness to either allow us to customize the acceptance criteria, or broaden the range.

We would like the ability for our Annual reports to have the ability to set a budget alert based on the monthly budget numbers, and exceeding them beyond a certain level. Currently, the alert is only based on the Annual total cost being exceeded.

We would like a method to take recommendations that are made with CCM and move them into the ignore List via API or through UI. We'd like to mark wildcards and for it to be able to process it. For example, all machines starting with windows-*

When actioning a recommendation Engineers often find the recommendation too harsh or contravenes a SaaS requirement, but they are able to implement a partial action of the recommendation and estimate their own more appropriate amount for the Savings $$$ figure. The integrations with Jira and SNow need to allow for the setting of the new amount when closing the ticket, populating the recommendation in Harness with the new estimated savings amount. They do not do this currently

Allow a recommendation to remain as listed once it has been put into a new "In Progress Status". This gives the engineer a chance to return to the tool to register the action taken and provide a fair estimation of the actual savings gained. Sometimes it takes a few weeks before the savings can be seen in enough data to prove the value of the actioned recommendation. A5t the moment the Engineer only has until the next update, when the recommendation may just disappear! Mad!

The recommendations table can be split by account but not by any other real "value" adding field. Enabling tag values on recommendations would enable us to use tags either on the account or on the resources to meaningfully split recommendations. Ideally this would apply at the cloud account level so we can build different views around more consistent values. Cloud accounts can come and go and maintaining views for Cloud accounts is painful

Description: Requests are available to view, but there is no differentiation between a read request and a write request. This can determine what instance is causing a performance bottleneck. The CUR may already have such a field. Service: Amazon Simple Storage Service Usage Type: USW2-Requests-SIA-Tier1 Unit: Requests Line Item Description: $0.01 per 1,000 PUT, COPY, POST or LIST requests to Standard-Infrequent Access However, there's no direct field that clearly indicates whether a request is a read or write operation. Feature Request: It would be helpful to have an additional field or tag that explicitly categorizes each request type as either a read or write. This would enhance visibility and improve the accuracy of usage analytics and cost attribution across services. Use Case: This would help teams more easily break down and monitor storage interactions based on operation type (e.g., PUT vs. GET), which is critical for optimizing costs and understanding usage patterns.

Currently, the scheduled report email displays the total cost and the cost for the previous month. However, the report can be confusing due to a lack of clarity around the time ranges. To improve this, we suggest explicitly labeling the figures with their respective time ranges: The left-hand figure should be labeled “The last [duration]” The right-hand figure should be labeled “Previous [duration]” This will help users better understand the context of the reported costs.

This is for on-prem Harness, but may apply to SaaS as well. Problems: You can't middle click a recommendation to open it in a new tab. If you left click a recommendation, then navigate back in your browser, it sends you back to page 1 with all filters disabled. The UI should respect standard browser workflows and maintain the user's state. These two issues together make it extremely clunky to use Recommendations. You have to manually click the ... menu on a recommendation, click "Copy Link", then open a new tab on your own and paste in the URL, to get any sensible workflow in this UI. Clicking 'back' in your browser and having to manually reset all your filters and the page you were on is nonsensical.