I am very disappointed in your response. You said:
which makes it sound like the results you are getting from your script give you what you want, but is too slow to use to process 800 log files. But, with your sample input, it produces no output. If we convert the date in the grep in your code:

Code:

grep "2017-16-02" logfile1 | cut -c2-18 | uniq -c

which is in the format YYYY-DD-MM to the date found in your sample file:

Code:

2017-02-15 17:49:06 ... ... ...

which is in the format YYYY-MM-DD, we get the output:

Code:

   5 017-02-15 17:49:0

which in addition to truncating the year also truncates the time so it will give you counts of ten second intervals instead of one second intervals, and makes no attempt to find the maximum count and makes no attempt to include the filename in the output.

Are you looking for counts for each one second interval or for 10 second intervals?

Are you looking for counts only on a specific date, or are you looking for the maximum count in each file no matter what the date might be for that count?

You have not answered most of the questions I asked in post #2 in this thread:If you can't provide a clear specification of what it is that needs to be done, it will be very hard for us to help you find a solution to your problem!