데이터 흐름 분석

catalog.execution_data_statistics SSISDB 데이터베이스 뷰를 사용하여 패키지의 데이터 흐름을 분석할 수 있습니다. 이 뷰는 데이터 흐름 구성 요소가 다운스트림 구성 요소에 데이터를 전송할 때마다 행을 표시합니다. 이 정보를 사용하여 각 구성 요소로 보내진 행을 자세하게 파악할 수 있습니다.

[!참고]

catalog.execution_data_statistics 뷰를 사용하여 정보를 캡처하려면 로깅 수준을 자세히로 설정해야 합니다.

다음 예에서는 패키지의 구성 요소 간에 보내진 행 수를 표시합니다.

use SSISDB
select package_name, task_name, source_component_name, destination_component_name, rows_sent
from catalog.execution_data_statistics
where execution_id = 132
order by source_component_name, destination_component_name

다음 예에서는 특정 실행 중 각 구성 요소가 보낸 밀리초당 행 수를 계산합니다. 계산되는 값은 다음과 같습니다.

  • total_rows - 구성 요소가 보낸 모든 행의 합계

  • wall_clock_time_ms - 각 구성 요소에 대한 총 실행 경과 시간(밀리초)

  • num_rows_per_millisecond - 각 구성 요소가 보낸 밀리초당 행 수

HAVING 절은 계산에서 0으로 나누기 오류를 방지하기 위해 사용됩니다.

use SSISDB
select source_component_name, destination_component_name,
    sum(rows_sent) as total_rows,
    DATEDIFF(ms,min(created_time),max(created_time)) as wall_clock_time_ms,
    ((0.0+sum(rows_sent)) / (datediff(ms,min(created_time),max(created_time)))) as [num_rows_per_millisecond]
from [catalog].[execution_data_statistics]
where execution_id = 132
group by source_component_name, destination_component_name
having (datediff(ms,min(created_time),max(created_time))) > 0
order by source_component_name desc

관련 태스크

데이터 흐름 디버깅

패키지 실행 문제 해결 도구

관련 내용

Integration Services 아이콘(작은 아이콘) Integration Services 관련 최신 정보 얻기

Microsoft의 최신 다운로드, 문서, 예제 및 비디오와 커뮤니티에서 선택된 솔루션을 보려면 MSDN의 Integration Services 페이지를 방문하십시오.


이러한 업데이트에 대한 자동 알림을 받으려면 해당 페이지에서 제공하는 RSS 피드를 구독하십시오.

참고 항목

개념

데이터 흐름의 데이터