System Architecture

Wherein I use LEGO to talk about organizing your code for ease of use

Companies spend a lot of money on runtime logging. But are we getting a lot of value? This blog post examines some common logging pitfalls and provides examples and best practices for how you can i…

A case study by Nils Hyoma, Jan Schneider and Wuk Petrovic.

The scalability of traveling salesperson problem (TSP) algorithms for handling large-scale problem instances has been an open problem for a long time. We arranged a so-called Santa Claus challenge and invited people to submit their algorithms to solve a TSP problem instance that is larger than 1 M nodes given only 1 h of computing time. In this article, we analyze the results and show which design choices are decisive in providing the best solution to the problem with the given constraints. There were three valid submissions, all based on local search, including k-opt up to k = 5. The most important design choice turned out to be the localization of the operator using a neighborhood graph. The divide-and-merge strategy suffers a 2% loss of quality. However, via parallelization, the result can be obtained within less than 2 min, which can make a key difference in real-life applications.

(Or, “regularized greedy algorithms and their applications.”) James Gilles “Algorithm” is one of those words that sounds fancier than it really deserves. We hear all the time about the “facebook algorithm”, the “youtube algorithm”, “high frequency trading algorithms”, “artificial intelligence algorithms”, and so on. What are these, really? The “facebook algorithm” is a large blob of software that Facebook uses to try to get people to stay on their website longer.

Have you ever been to an event where you had to present ID to prove that you were the person who bought the ticket? The act of asking for…

“After building like eight personal wikis/information management systems/hypertext databases (and becoming exceedingly efficient at it) I'll say the problem is that 99% of their uses decompose into three disjoint usecases: 1. Todo list+journal 2. Learning 3. Reference archive”

The kernel is, in many ways, a marvel of scalability, but there is a longstanding pain point in the memory-management subsystem that has resisted all attempts at elimination: the mmap_lock. This lock was inevitably a topic at the 2022 Linux Storage, Filesystem, Memory-Management and BPF Summit (LSFMM), where the idea of using per-VMA locks was raised. Suren Baghdasaryan has posted an implementation of that idea — but with an interesting twist on how those locks are implemented.

Minimalistic data structure sketches

We’ll build a web app to solve every developer's most common problem – finding an excuse to justify our messy work! And will do it with a single config file that covers the full-stack app architecture plus several dozen lines of code. In the quickest possible way, so we can’t excuse ourselves from building it!

A reverse-proxy Phishing-as-a-Service (PaaS) platform called EvilProxy has emerged, promising to steal authentication tokens to bypass multi-factor authentication (MFA) on Apple, Google, Facebook, Microsoft, Twitter, GitHub, GoDaddy, and even PyPI.

An execution environment is the combination of hardware and software components that can be used to execute and support applications. Traditionally, computing is effected in execution environments that not only permit the loading and execution of arbitrary programs but may also themselves be manipulated in arbitrary ways, such as by attaching a hardware debugger to observe and change the internal state. Trusted Execution Environments (TEEs) address this by providing an isolated environment for user applications with well-known properties.

Posted on 2022-09-05

Huffman coding is a lossless coding scheme that assigns variable-length codes to data symbols. The codes are assigned such that the shorter codes are used

You can download all the source code for this site and browse it completely offline from GitHub.

USB4 vs. Thunderbolt 4—and everything else to know about the newest USB standard.

For the past years, we’ve seen an increasing interest in using safe applications and making them as reactive as possible. For this, there…

A new report from Datadog reveals that serverless computing could be entering the mainstream. This has already been the case in my world for the past few years. Results show that more than half of all organizations are using serverless on one of the three major public cloud providers: Amazon,

CodeSee's enterprise platform is based on its codebase visualization and mapping tools and makes it easier for developers to navigate.

This fifth and final part of our blog series exploring Git's internals shows several strategies for scaling your Git repositories that match related database sharding techniques.

Farcaster, a decentralized social network

Hyperconverged infrastructure (HCI) is all well and good for companies that only need onsite hardware which supports business processes done in a standard way. But what if you need a solution with more flexibility and faster scalability?

Yes, and it’s GitHub. GitHub is an explorer and a search engine for the git protocol, and so most of the data I submit to it or edit is replicated locally. I edit most of my work on my computer