DistributedComputing

DISTRIBUTEDC.md

##Map Reduce

1. Map/Reduce Library partitions the input data into M pieces of typically 16-64 MB.

   True.

2. If there are M partitions of the input, there are M map workers running simultaneously.

   False. There are generally less worker nodes than partitions.

3. Reduce function accepts intermediate key and a set of values for that key and merges together these values to form possibly a smaller set of values.

   True.

4. Map and Reduce are user defined functions

   True.

5. MapReduce is a restricted programming model?

   True.

Following are multiple Choice Questions.

Question 1: Which of the following functions of cloud computing is optimized by MapReduce?

      ` e) All of the Above`

Question 2: The value of R, the number of Reduce workers, is determined by

 `b)  master program - number of partitions are defined by user though `

Question 3: In Map Reduce function, what can be said about the input key, output key and the intermediate key values?

C. The input keys are drawn from a different domain as output keys and the intermediate keys are drawn from the same domain as output keys.

Question 4: What is the relationship between intermediate keys and intermediate value? A 1:N

Question 5: Which of the following Key/Value pair is used by Map function in Map Reduce?

b) (k1, v1)--> List(k2 , v2)

Question 6: How is the intermediate Key/Value pair arranged and processed during partition of tasks?

a) Key/value pairs are processed in increasing key order

Question 7: Which is the reduce function in the example of Distributed Grep:

c) Copy the intermediate data to output

Question 8: Which is the reduce function in the example of Text indexing?

d) none of the above - count document IDs for each word

Question 9: Which of the following properties of Worker Machine is stored by MapReduce?

a) State b) Identity

Question 10: Map is phased in M pieces and Reduce is phased into R pieces. What is the maximum number of scheduling operations required to be performed?

`b)  O(M+R)`

Question 11: How is the network bandwidth maintained by Map Reduce in cloud computing?

a) Data is stored locally on disks on different machines with 64 MB blocks

Question 12: How MapReduce environment handle failures?

b) Master worker pings the client worker regularly to check the status

Question 13: MapReduce provides the following benefits:

a) Reducing the amount of data sent across the network b) Optimization the data by storing it locally c) Redundant execution to handle machine failures and data loss

Question 14: What is the difference between the Combiner function and Reduce function?

b) Output of combiner function is written to intermediate file and Reduce function to output file

Remote Procedure Call (RPC)

1. An RPC (remote procedure call) is initiated by the: client

2. RPC works between two processes. These processes may be: On the same computer and a different computer connected with a network.

Client and Server

1. The local operating system on the server machine passes the incoming packets to the: Server stub

Distributed Objects

1. _____is a framework for distributed objects on the Microsoft platform. DCOM

2. ____ is a framework for distributed objects using Borland Delphi. DDObject

3. ____ is a framework for distributed components using a messaging paradigm. Jt

4. ____ is a Sun specification for a distributed, shared memory. JavaSpaces

Map Reduce

1. ____ is a framework for distributed objects using the Python programming language. Pyro

2. The reduce function typically outputs a smaller set than what is input to it. True

3. If there are M partitions of the input, there are M map workers running simultaneously. True or False? False