Grant Fritchey (5/27/2015)

---

The execution time is the same? I'm not surprised the reads are the same. You're still scanning 45,000 rows.

There just isn't anything to tune in the query here.

Wild shot in the dark, maybe if you set up a filtered, nonclustered index on the same key column as your cluster, filtered for 'RESO' and use the three columns from the output as INCLUDE. You'll still get an index scan, but it might be a smaller number of reads since you'll possibly have more data per page (maybe, can't see the rest of your structure). I'm not sure that will help at all.