shantha13

12345678910
Across
  1. 1. The process of reorganizing a Pail's mixed data (edges, properties) into a structure matching the vertically partitioned format required by the master dataset
  2. 4. A precomputed dataset generated by processing the master dataset. These views are designed to answer specific queries efficiently.
  3. 6. A simple indexing method where each key maps directly to a value, used for metrics like bounce rates
  4. 8. An indexing strategy where related data (e.g., all pageviews for a URL) is stored together, improving latency and throughput
  5. 10. The level of time precision (e.g., hour, day, week, month) used in precomputing valuesThe level of time precision (e.g., hour, day, week, month) used in precomputing valuesThe level of time precision (e.g., hour, day, week, month) used in precomputing values
Down
  1. 2. A probabilistic data structure used to estimate the cardinality (number of unique elements) of a dataset. It offers high accuracy with low memory usage
  2. 3. Algorithms that update only the parts of the data that have changed instead of recomputing everything from scratch
  3. 5. The time taken to serve a single query from the serving layer.
  4. 7. A library for managing structured data on HDFS. It supports features like vertical partitioning, snapshots, and appends
  5. 9. A graph structure where edges indicate that two user identifiers refer to the same individual. It is used in user-identifier normalization.