scrapyd - sharing memory between separate processes?

Question

I run multiple spiders concurrently by posting new start_urls in scrapyd and it creates separate processes.

How can I get all of the crawled items in the mem