Given a distributed database system for a bank which has $3$ datacenters located in NY$,$ LA$,$ and TEXAS. Each record contains an account number $($unique$)$ and a dollar amount.

Part $1.$ Assume: The data is fragmented with $\backslash (33\mathrm{.}33\backslash \%\backslash )$ of the data in each location with no overlapping data.

$($a$)$ You wanted to know the account number of the account which has the most money. How would the DBMS determine this?

$($b$)$You wanted to change account $123\text{'}$s balance to $500$ million. How would the DBMS accomplish this?

Part $2$: Assume: The data is fully replicated at each location.

$($a$)$ You wanted to know the account number of the account which has the most money. How would the DBMS determine this?

$($b$)$You wanted to change account $123\text{'}$s balance to $500$ million. How would the DBMS accomplish this?