0

As per Technet specifications for the crawl database:

Read/write characteristics: Read heavy.

Wouldn't crawling require rather write operations than read?

On the other hand for the link database:

Read/write characteristics: Write heavy during content processing.

Once the links are created (write operation) wouldn't reading them (users browsing, searching etc.) occur more frequently than writing new ones?

1

@Grisha,

I will try to give you a brief definitions of the concepts in order to help you to build a better overview about the search components.

Crawling is simply a process of gathering documents from various sources/repositories, making sure they obey by various rules and sending them off for further processing to the Content Processing Component.

Read during Gathering. Write towards Content Processing Component. You can follow the counters of mssdmn.exe process on the task manger performance monitor.

Content Processing Component:

The content processing component processes crawled items and sends these items to the index component. The content processing component performs operations such as document parsing and property mapping. It also performs linguistics processing such as language detection and entity extraction. The component transforms crawled items into artifacts that are included in the search index. The content processing component also writes information about links and URLs to the link database.

You can follow the counters and activity of noderunner.exe (PID- Task Manager) process associated to NodeRunnerContent (PID - ULS). The Reading is lower since it has the info already and it is mainly busy writing properties in to the indexer.

Users browsing is another different process associated to w3wp.exe SharePoint Search is a combination of NodeRunnerQuery and NodeRunner

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.