Final Exam: Snowflake Performance Monitoring and Optimization

WHAT YOU WILL LEARN

• 

  Outline the defining aspects of snowflake and analyze the three layers in snowflake architecture
  

  choose between resizing warehouses and multi-cluster warehouses
  

  create a new compute warehouse and explore warehouse settings
  

  evaluate the benefits of large compute warehouses for more complex queries and greater concurrency
  

  configure a multi-cluster warehouse in auto-scale mode with an economy scaling policy
  

  contrast the performance of auto-scale mode clusters with standard scaling and maximized mode clusters
  

  utilize snowflake resource monitors to take action when usage of credits exceeds certain thresholds
  

  analyze how databases, schemas, tables, and partitions figure into snowflake's data model
  

  analyze the working of the query acceleration service (qas) and caching on snowflake
  

  instantiate a warehouse with the qas turned on and evaluate a query for eligibility
  

  analyze the results of an eligible query run with and without query acceleration
  

  load data into a table for benchmarking purposes
  

  analyze the impact of query result caching and local disk caching
  

  provide an overview of the intuition behind clustering depth and the number of overlapping partitions
  

  populate a table with data using put and copy into and use tablesample to shuffle the data in that table
  

  compare query performance on clustered and unclustered tables
• 

  resume clustering after cloning a clustered table
  

  turn on search optimization for a table and verify that the search access path is fully constructed
  

  compare the performance of queries using search optimization, clustering, and no optimization at all
  

  evaluate search optimization with in, and, and or clauses
  

  implement search optimizations for individual columns for both equality and substring matches
  

  work with variant, object, and array data loaded into a snowflake table
  

  leverage search optimization for working with variant, array, and object data
  

  create and query views
  

  use views and role-based access control to grant different users access to different subsets of a table
  

  create and query materialized views
  

  suspend and resume materialized views
  

  create secure views and contrast them with ordinary views
  

  create a google cloud platform (gcp) service integration object and assign gcp roles to the associated service account
  

  create an external table connected to an external stage
  

  create an iceberg table and configure appropriate cloud integrations