Skip to content
Lucas Lessa
Voltar para as notícias
SnowflakeAtualizaçãoImpacto: Médio

Improved data freshness for organization usage history views

Data latency dropped to under 15 minutes for selected premium views in the ORGANIZATION_USAGE schema.

O que mudou

For what it is used

Imagine you need to audit access, logins, consumption, and queries across multiple accounts in your organization almost in real time. Now, selected premium history views in the ORGANIZATION_USAGE schema update in under 15 minutes.

  • Query access history much faster.
  • Track login attempts with lower delay.
  • Monitor credit consumption almost instantly.
  • Analyze query history with up-to-date data.

How to use

SELECT *
FROM snowflake.organization_usage.query_history
WHERE start_time >= DATEADD('hour', -1, CURRENT_TIMESTAMP());

What happens in this code:

  1. Accesses the query_history premium view in the organization usage schema.
  2. Filters queries executed in the last hour, leveraging the new reduced data latency of under 15 minutes.

Limitations and requirements

Affected views are ACCESS_HISTORY, LOGIN_HISTORY, METERING_HISTORY, and QUERY_HISTORY. This applies to premium views in the ORGANIZATION_USAGE schema, which incur additional consumption costs based on processed records and require a capacity contract by default.

Ver nota oficial do Snowflake

Improved data freshness for selected ORGANIZATION_USAGE history views

Resumo e explicação gerados por IA a partir da release note oficial. Em caso de dúvida, a documentação oficial da plataforma prevalece.

Relacionadas