Large Data volume operation

1, the data volume of the table, such as currency Code, organization code, and so on when the client program starts loading into the DataTable, the access to these code information as far as possible not to trouble the server, in memory can be checked.
2, indexing, according to the user's operating habits, the data table most frequently accessed fields indexed, and rebuilt once a month.
3. Use the most level SELECT statement (less subqueries) to maximize the efficiency of the database itself (Oracle I use)

We don't see any bottlenecks in the amount of data we have tens of millions of a day. The key is that your index is properly built and indexed as much as needed. Of course, the performance after partitioning should also be greatly improved.
It is also necessary to construct the corresponding caching mechanism and data mining mechanism according to the specific business.
But you didn't say anything. I couldn't give you a more detailed solution.
Also increase the speed of the hard drive, for example, using a RAID array. Because there is disk IO operation even for index queries. And this value can be calculated.

Large Data volume operation