Catalog scans for a few hours (e.g. 6 hours) will have few duplicate catalog leaves. It might be faster to skip the "find latest" step for these catalog scans. For catalog scans spanning a single catalog page definitely don't need to "find latest" since the duplicates can be eliminated in memory when the page is read.
Catalog scans for a few hours (e.g. 6 hours) will have few duplicate catalog leaves. It might be faster to skip the "find latest" step for these catalog scans. For catalog scans spanning a single catalog page definitely don't need to "find latest" since the duplicates can be eliminated in memory when the page is read.