Data Locality — Learn free

Self-check

Questions 1–10 of 10

1. What is Data Locality?
Question 1A. Moving the computation close to where the data resides on the storage or data node, instead of moving large data to the computationB. Knowing the geographical location from which data was generatedC. Storing data in the country where it was generated to satisfy sovereignty lawsD. Copying all data from data nodes into the master nodes before queryingE. Separating compute and storage into different clusters
2. Traditionally, where was data fetched to for computation in analysis?
Question 2A. From master nodes into storage nodesB. From storage or data nodes into master nodesC. From the cloud into on-premises serversD. From a Data Room into a Data LakeE. From columnar storage into row storage
3. Why did fetching data into master nodes for computation become impractical?
Question 3A. Because master nodes cannot run SQLB. Because network links between nodes were removedC. Because semi-structured and unstructured data now runs to terabytes and petabytes on clusters of hundreds of data nodes, while master nodes are normally fewD. Because data must stay in the country where it was generatedE. Because master nodes only store metadata
4. With the introduction of Big Data, Hadoop and HDFS, in what shape is data stored on the data nodes?
Question 4A. As documentsB. As key-value pairsC. As graph nodes and edgesD. As blocksE. As spreadsheets
5. Under Data Locality, what is sent back from the storage or data nodes after the query is processed there?
Question 5A. The complete raw datasetB. Every block that the query touchedC. A copy of the whole cluster's metadataD. The query itself, for re-execution on the master nodeE. Only the processed data
6. In a nutshell, when is a query said to exhibit Data Locality?
Question 6A. When the query runs on the same node where the data is locatedB. When the query is written in standard SQLC. When the query returns results in real timeD. When the query is executed on a master nodeE. When the query uses columnar rather than row storage
7. Which newer concept is described as surfacing in recent times, in contrast to Data Locality?
Question 7A. Pulling whole datasets into master nodesB. Separating computing and storageC. Storing all data in a single Physical Data RoomD. Replacing HDFS with operating-system file systemsE. Merging master nodes and data nodes into one
8. How relevant is the literal reading of Data Locality as simply knowing the location of the data?
Question 8A. Entirely accurate: that is the whole meaningB. Entirely wrong: location has nothing to do with itC. Partially relevant: the name suggests location, but the concept is about moving computation to the dataD. Relevant only for relational databasesE. Relevant only for cloud storage
9. Which of the following is NOT part of the history of how data was pulled for processing before Data Locality?
Question 9A. Initially the whole data was pulledB. Later, blocks were pulledC. Later, the columnar pull concept came inD. Later, data was pulled as printed reportsE. Eventually the query was taken to the data instead
10. What was the critical requirement that led to the Data Locality concept?
Question 10A. To encrypt data on every data nodeB. To reduce the number of master nodes to oneC. To convert all data to structured formD. To keep data in the country where it was generatedE. To process queries where the data is residing