You appear to have truncated the email thread, so I am working on memory here. I think your's was the system with a progressive degradation in the cache hit rate, and that the statistics that you sent me when the instance was fairly new showed about a 90% hit rate against the default buffer pool, good hit rates on the other buffer pools, and a tunable cache miss rate of about 3%. These statistics show that the hit rate against the default buffer pool has fallen to about 70% now, but that the tunable cache miss rate has stayed at about 3%. That means that the reason that your cache hit rate degrades progressively is that you do a lot of table scans (not direct reads because the buffer_pool_miss_rates.sql script excludes those). To tune this system, you need to find and fix (if appropriate) those full table scans.

I am sending you the status of tunable_cache_miss_rate,buffer_pool_miss_rates, and db_hit queries results attached. Pl suggest me what to do.

Thanks. I presume that it is the hit ratio for the DEFAULT pool that gets worse. To check whether you have a real degradation, please run the tunable_cache_miss_rate.sql script regularly for a few days (whenever you run your current script) so that we can see whether the figures correlate.