I have data as follows:
| SnapshotDate | User | Last Modified | Is X Done | Is Y Done |
| 1/1/2016 | A | 1-Jan | No | Yes |
| 1/8/2016 | A | 1-Jan | No | Yes |
| 1/15/2016 | A | 1-Jan | No | Yes |
| 1/1/2016 | B | 1-Jan | No | No |
| 1/8/2016 | B | 8-Jan | Yes | No |
| 1/15/2016 | B | 15-Jan | Yes | Yes |
| 1/1/2016 | C | 1-Jan | No | No |
| 1/8/2016 | C | 1-Jan | No | No |
| 1/15/2016 | C | 15-Jan | Yes | No |
Snapshots are taken once per week for all users, A, B, and C. Every time a User updates the responses to questions "Is X Done" and "Is Y Done" the data gets changed and the "Last Modified" reflects the date that the change was made.
I have data like this coming in from Salesforce. There are currently about 20,000 rows of data excluding snapshots. If we add snapshots, there will be a snapshot per week and it's desired to have a history of 5 weeks. This would translate to 100,000 rows (20,000 * 5). In the next few years the data is expected to increase by 1200%, which, if my math is correct would correspond to roughly 1.2 million rows of data including snapshots. It currently takes me about 30 seconds to refresh all the data for those 20,000 rows. Is it correct to assume that the refresh speed is going to slow down by 1200% as well?
If this is the case, is there a better way to deal with this situation?