the Peterman Pod

@ryanlpeterman

15.0K followers · usually 20.4K views a post

Views

53.7K

Likes

1.3K

Comments

134

Shares

49

SoundTikTok sound · original sound - ryanlpeterman

Why it worked

The video leverages the credibility of a "Turing Award Winner" and a "faster filesystem than Linux" hook to draw viewers into a technical discussion about operating system design and performance.

Summary

The video discusses the potential for a database-managed filesystem to be faster than Linux's native filesystem. The speaker explains that a well-designed scheduling engine can provide high availability without complex failover mechanisms, and questions why Linux hasn't adopted such improvements. He notes that operating system developers historically felt threatened by database approaches encroaching on their territory.

Structure

1Introduction of a faster filesystem concept
2Explanation of high availability through scheduling engines
3Questioning Linux's adoption of such improvements
4Historical resistance from operating system developers

Product placement

Linux (operating system): acts - it is the filesystem being compared to a new one; removing it would change the topic. Postgres (database): appears - mentioned as a creation of Mike Stonebraker; removing it would not change the topic.

Call to action

Search his name on my YouTube, Spotify or Apple Podcasts to find the full conversation

On-screen text

the Peterman Pod A faster filesystem than what Linux has Mike Stonebraker (Turing Award Winner)

Original caption

Can a database make filesystems faster? This is a clip from a longer conversation I had with Mike Stonebraker (Turing Award Winner and creator of Postgres). You can search his name on my YouTube, Spotify or Apple Podcasts to find the full conversation

More from @ryanlpeterman

More breakout videos from this creator.

Transcript

Will a filesystem written on top of a DBMS be faster than the Linux filesystem? The scheduling engine is competitive with other scheduling engines and the you can make everything fail over so you get high availability without having to do anything else. And so the answer is there, there's really no downside. Then why wouldn't Linux incorporate that and upgrade itself? You hope they would. In other words, you should keep all the device driver junk down at the bottom 'cause that's, there's a lot of it. And no one wants to do that and replace everything else. Implementation, is that something that you've mentioned to Linux people and what's their typical reaction? Back in the academic project, when I'd mentioned that to operating system folks, they would get very threatened, which is, this is the database guys trying to take over their turf.