New

The latest blog posts to take off on Hacker News. 4,389 posts.

Systems & Performance
Any year
1688

The DynamoDB Paper

This week at USENIX ATC’22, a group of my colleagues 1 from the AWS DynamoDB team are going to be presenting their paper Amazon DynamoDB: A Scalable, Predictably Performant, and Fully Managed NoSQL Database Service . This paper is a rare look at a real-world distributed system that runs at massive scale.

1697

RETRO is fast

When I first read Google’s RETRO paper, I was skeptical. Sure, RETRO models are 25x smaller than the competition, supposedly leading to HUGE savings in training and inference costs. But what about the new trillion token “retrieval database” they added to the architcture? Surely that must add back some computational…

1699

How does rsync work?

This post is the third article in a series of blog posts about rsync, see the Series Overview. With rsync up and running, it’s time to take a peek under the hood of rsync to better understand how it works. How does rsync work? When talking about the rsync protocol, we need to distinguish between:

1700

RISC-V is getting MSIs

Message signaled interrupts or MSIs describe a way to signal an interrupt without a dedicated interrupt request pin (IRQ). One of the most prevalent uses for MSIs is the PCI bus, and the PCI specification defines the MSI and MSI-X standards. The potential benefits may include: (1) reduced number of direct wires from…