3. Select one or more of the following statements about scaling up and scaling out that are true.

Some computational problems are so big that scaling out is the only feasible way to solve them.

Creating a grid to solve a large computing problem is an example of scaling up

An organization that wants to be able to easily redeploy computing resources is more likely to scale out rather than scale up

Scaling out is a better choice than scaling up when the speed of communication between processors or processing nodes is a critical performance factor.

4. Which of the following statements correctly describe the operation, purpose, or use of the PUSH and POP instructions?

The value stored in the stack pointer must be changed as a result of PUSH instruction but should be unchanged after executing a POP instruction

Executing a PUSH operation can result in a stack overflow

An interrupt handler executes a PUSH instruction just before it exits

POP adds the current machine state to the bottom of the stack

PUSH adds the current machine state to the top of the stack

PUSH and POP enable the CPU to pause work on one program, execute another program, and resume execution of the previous program from the point of interruption.

5. Which of the following statements is true about the impact of a cache on storage device read performance?

		Performance improvements for read operations will probably be greater than for write operations
		Reading data that was recently written to the storage device may occur more quickly since the cache may still contain the data.
		If the cache already contains the data being read, it can be delivered more quickly because access to the cache is faster than access to the underlying storage device.
		Performance will increase if the cache controller can accurately guess which data the CPU will ask for next and fetch it into the cache before the CPU asks for it.