Finding a reasonable recommendation that works for many use cases is hard, Splunk use varies too much for that. However, I've yet to come across a use case where more than 50 panels, let alone more than 50 rows were the best way to go.
If that 50 row limit is stopping you from building your dashboard to your requirements, check if there are any "natural" ways of splitting up the data. Common splits are on detail level - providing a wider view first and drilling into more detailed views -, on application level - providing a whole-system overview first and drilling into applications or systems -, on departmental concerns - again starting from an overview, and drilling into operations, development, CRM, marketing, whatever views -, or simply by turning a dashboard listing ALL the servers into a form that lets you select what servers you want to see.
As a bonus, it'll make the individual dashboards much faster to load, less of a strain on the browser, and increase the chance of the user actually getting to view the data he needs rather than scrolling for miles to find it.
Are you seeing this in advanced xml or simple xml dashboards?
Simple XML Dashboards.
Finding a reasonable recommendation that works for many use cases is hard, Splunk use varies too much for that. However, I've yet to come across a use case where more than 50 panels, let alone more than 50 rows were the best way to go.
If that 50 row limit is stopping you from building your dashboard to your requirements, check if there are any "natural" ways of splitting up the data. Common splits are on detail level - providing a wider view first and drilling into more detailed views -, on application level - providing a whole-system overview first and drilling into applications or systems -, on departmental concerns - again starting from an overview, and drilling into operations, development, CRM, marketing, whatever views -, or simply by turning a dashboard listing ALL the servers into a form that lets you select what servers you want to see.
As a bonus, it'll make the individual dashboards much faster to load, less of a strain on the browser, and increase the chance of the user actually getting to view the data he needs rather than scrolling for miles to find it.