System Architecture

by Aditya Mavlankar, Zhi Li, Lukáš Krasula and Christos Bampis

Border Gateway Protocol and Generic Routing Encapsulation allow rerouting of traffic to scrubbing centers where harmful data is filtered out.

WARNING for Raspberry Pi 5 NAS users: **If you use btrfs to set up a Raspberry Pi 5 as a NAS, your filesystem will not be mountable from almost all other machines unless you use `mkfs.btrfs -s 4096`**. The Raspberry Pi 5 uses 16K pages by default, and as of today this is the default block size for mkfs.btrfs upstream on such systems. **btrfs does not support mounting 16K block size filesystems on x86 machines, or any machine with a 4K page size**. We knew this was going to be a problem for Asahi Linux (which also uses 16K pages), so Fedora ARM64 has long carried a patch to change the default to 4K. Unfortunately, even though we submitted it upstream a long time ago, the btrfs-progs maintainers have chosen not to apply it at this time. If you format a btrfs filesystem on a Raspberry Pi 5 normally, it will work, but you will be setting yourself up for a terrible disaster recovery scenario: If the Pi ever stops working, you will have to *find another one* (or an Apple Silicon laptop) to gain access to your data again - and no replacing/upgrading the Pi with any other machine (except a Mac) without a full disk reformat.

Active measurements over the public Internet can target either collaborating parties or non-collaborating ones. Sometimes these measurements, also called "probes", are viewed as unwelcome or aggressive. This document suggests some simple techniques for a source to identify its probes. This allows any party or organization to understand what an unsolicited probe packet is, what its purpose is, and, most importantly, who to contact. The technique relies on offline analysis of the probe; therefore, it does not require any change in the data or control plane. It has been designed mainly for layer 3 measurements.

#25: Learn More - How Distributed Systems Gossip (6 minutes)

Over the past few years, the Arm architecture has made steady gains, particularly among the hyperscalers and cloud builders. But in the HPC community, Arm remains under-represented. But perhaps not for long. The “Fugaku” system at RIKEN Lab in Japan is without a doubt the largest and best known Arm supercomputer, and blazed the technological

What is Flutter? This article explores how the mobile UI framework works, its key features, and the skills needed to use it.

As containerisation of applications becomes more popular, could we build a modern mainframe with new hardware features?

Explore and visualize data. Share and publish your insights. Discover and be inspired.

What started as lighthearted iconoclasm, poking at the bear of SOLID, has developed into something more concrete and tangible. If I do not think the SOLID principles are useful these days, then what would I replace them with? Can any set of principles hold for all software? What do we even mean by principles?

Microprocessor-scale transistors detect and respond to biological states and the environment. Your phone may have more than 15 billion tiny transistors packed into its microprocessor chips. The transistors are made of silicon, metals like gold and copper, and insulators that together take an elec

MirageOS is a programming framework for building type-safe, modular systems.

Below is an excerpt from Street Coder, where Sedat Kapanoglu talks about opacity in software.

A new way of understanding climate change and other phenomena.

Do you feel lost? Alone? Powerless in the face of forces beyond your control? Timothy Morton can help—if you’re ready to have your reality blown apart.

A detailed comparison of CPU, RAM, storage and network performance between Kubernetes clusters on virtual machines and bare metal.

Restoration project headed up by Steve Kaisler is gathering steam

#23: Read Now - Algorithm to Build a Planet-Scale Distributed Cache (4 minutes)

Nebulon has developed a DPU specifically for edge servers, enabling their CPUs to fully focus on running applications.

This document specifies the "SVCB" ("Service Binding") and "HTTPS" DNS resource record (RR) types to facilitate the lookup of information needed to make connections to network services, such as for HTTP origins. SVCB records allow a service to be provided from multiple alternative endpoints, each with associated parameters (such as transport protocol configuration), and are extensible to support future uses (such as keys for encrypting the TLS ClientHello). They also enable aliasing of apex domains, which is not possible with CNAME. The HTTPS RR is a variation of SVCB for use with HTTP (see RFC 9110, "HTTP Semantics"). By providing more information to the client before it attempts to establish a connection, these records offer potential benefits to both performance and privacy.

This note describes additional optional resource records for use with the Domain Name System (DNS). These optional resource records are for use with the Identifier-Locator Network Protocol (ILNP). This document is a product of the IRTF Routing Research Group. This document defines an Experimental Protocol for the Internet community.