Performance tuning of Realtime cube

Hello Experts,

I have some real-time info-cubes in system, and data is getting loaded into these cube multiple times a day.

Data is been getting loaded into these cube since some years now.

We are observing very low performance in Bex Queries on these cubes, and few reports take like for ever to to display the output.

I seek help here in few questions here.

1. Conventional performance tuning activities such as Aggregates, compression, partition are applicable to real-time cubes same as standard info cubes?

2. What will happen if Real-time cube data load and a process chain for aggregate or compression run in parallel at the same time ?

3. Are the any other ways to improve bex query performance for Real-time info cubes ?

Thank in advance to all you take effort to reply on this.

Thanks,

Tanveer