Kata Containers; LosslessCut; SQL or Death?

We continue our contained enthusiasm, and pair it with a sleek and speedy FFmpeg-based GUI video editing app, and a series that’s guaranteed to level up your SQL ops (or die trying).

In other news, to augment the work of a fine data journalist, I’ve got a small app that tracks changes to DNS records and HTTP/HTTPS / of a curated list of U.S. government web sites.

And, 47 Watch has a new search option and another budding feature we’ll talk about in a future Drop to showcase a cool Typst package.

---

TL;DR

(This is an AI-generated summary of today’s Drop using Ollama + llama 3.2 and a custom prompt.)

• An open source secure container runtime using lightweight VMs for stronger workload isolation through hardware virtualization, supporting multiple architectures and hypervisors (https://katacontainers.io)
• A cross-platform FFmpeg GUI tool for fast, lossless video/audio editing through direct stream copying rather than re-encoding (https://github.com/mifi/lossless-cut)
• CMU Database Research Group’s seminar series exploring SQL optimization and potential replacements, featuring talks from industry experts through Spring 2024 (https://db.cs.cmu.edu/)

---

Kata Containers

Kata Containers (GH) are different approach to container runtime implementation that bridges the gap between traditional containerization and hardware virtualization. Originally known as Clear Containers, it provides an OCI-compatible runtime that significantly differs from standard container implementations by leaning on virtualization technology to level up host isolation.

The architecture of Kata Containers is interesting from a systems perspective. They have multiple moving parts that all work together. The kata-agent operates as a supervisor within the virtualized guest environment, while the kata-runtime handles OCI specification compliance and manages container lifecycle events.

Unlike the containers you are likely most familiar with that share the host kernel, Kata Containers creates a lightweight VM for each container or pod (I stll loathe that term), using a specialized Linux kernel and custom-built images. This approach provides yugely upgraded isolation guarantees, but still remains compatible with container orchestration platforms.

From an implementation standpoint, Kata Containers integrates with major container engines like Docker and Podman through runtime configuration. The system is highly configurable, and supports multiple hypervisors including qemu-desktop, firecracker, and cloud-hypervisor.

There’s a whole other component of Kata Containers when it comes to how they work with Kubernetes, but this is not a Kubernetes blog, and most readers likely do not have large enough, or complex enough workloads to warrant digging in to that.

Ubuntu support for version 2 is still a bit thin, so we’ll forego a gnarly setup walkthrough and look forward to another few alternative container environments later in the week.

---

lossless-cut

LosslessCut (GH), in their own words:

aims to be the ultimate cross platform FFmpeg GUI for extremely fast and lossless operations on video, audio, subtitle and other related media files. The main feature is lossless trimming and cutting of video and audio files, which is great for saving space by rough-cutting your large video files taken from a video camera, GoPro, drone, etc. It lets you quickly extract the good parts from your videos and discard many gigabytes of data without doing a slow re-encode and thereby losing quality. Or you can add a music or subtitle track to your video without needing to encode. Everything is extremely fast because it does an almost direct data copy, fueled by the awesome FFmpeg which does all the grunt work.

LosslessCut fundamentally differs from traditional video editors by since it performs direct data stream copying rather than re-encoding files. While other editors may require hours of processing and inevitably degrade quality through repeated encoding, LosslessCut performs edits in seconds while maintaining the original quality.

Some things you can do with it include:

• seamlessly combining audio and video tracks from separate recordings
• extracting and manipulating specific media streams without quality loss
• converting container formats (like MKV to MP4) without re-encoding
• managing complex metadata, including GPS tracks and chapter markers
• performing scene detection and automated segmentation

The app is freely available on GitHub, but it’s also offered through app stores for a fee to help support the project (which I did since these are hard times, but it works the same if you can’t spare the coin).

---

SQL or Death?

Again, in the resource owner’s (CMU) own words (no blockquote this time)…

Suppose somebody has been rubbing gasoline on their body since the 1970s. Would you marry that person even if they smelled terrible? But suppose this person starts showering every day so they smell a little better. They also famously get along with nearly everyone on the planet and makes a lot of money. What about marrying this person now? This is the question that we face today in the world of databases. SQL is 50 years old. It started off smelling kind of funny, but it has gotten better over the years. There have been many attempts at replacing it, but none have succeeded.

The Carnegie Mellon University Database Research Group is exploring this question with the SQL or Death? Seminar Series. The talks in this series will present ideas on either (1) making SQL go as fast as possible or (2) replacing SQL with something better.

All talks are on-line and open to the public via Zoom. You do not need to be a current CMU student to attend. Random people off of the internet are especially welcome. Videos will be posted on the CMU-DB YouTube Channel after each talk.

Here’s the schedule:

• Feb 10: James Cowling — “Larry Ellison was Right (kinda)! TypeScript Stored Procedures for the Modern Age”
• Feb 17: Viktor Leis, Thomas Neumann — “Towards Sanity in Query Languages”
• Feb 24: Yash Mayya, Gonzalo Ortiz — “Apache Pinot Query Optimizer”
• Mar 03: Lloyd Tabb — “Malloy: A Modern Open Source Language for Analyzing, Transforming, and Modeling Data”
• Mar 10: Jeff Shute — “GoogleSQL Pipe Syntax”
• Mar 24: Tobias Brandt — “PRQL: Pipelined Relational Query Language”
• Mar 31: Kaisen Kang — “StarRocks Query Optimizer”
• Apr 07: Ben Naecker — “OxQL: Oximeter Query Language”
• Apr 14: Michael Widenius — “MariaDB’s New Query Optimizer”
• Apr 21: Michael Sullivan, “EdgeQL with EdgeDB”

---

FIN

Remember, you can follow and interact with the full text of The Daily Drop’s free posts on:

• 🐘 Mastodon via @dailydrop.hrbrmstr.dev@dailydrop.hrbrmstr.dev
• 🦋 Bluesky via https://bsky.app/profile/dailydrop.hrbrmstr.dev.web.brid.gy

Also, refer to:

• this post, and
• this post

to see how to access a regularly updated database of all the Drops with extracted links, and full-text search capability. ☮️