1951
Call Me Maybe: Chronos
Chronos is a distributed task scheduler (cf. cron) for the Mesos cluster management system. In this edition of Jepsen , we’ll see how simple network interruptions can permanently disrupt a Chronos+Mesos cluster
Topics /
Systems & Performance
The most popular blog posts about systems & performance on Hacker News. 6,942 posts.
1952
An eBPF loophole: Using XDP for egress traffic
XDP only works for ingress. We found a loophole that lets it work for egress. Here's how we did the impossible.
1953
epoll: The API that powers the modern internet (2022)
Linux's epoll API solved the C10K problem, enabling fast and afforable Internet services.
1954
Writing high performance F# code
While this post is addressed to F# .NET developers, it introduces much wider concepts starting from hardware architecture to overall .NET runtime and JIT compiler optimizations. It shouldn't be a surprise - optimizing the application performance requires us to understand the relationships between our high level code…
1955
1956
1957
Using load shedding to avoid overload
How Amazon uses load shedding to stay available under overload, rejecting excess work early so the system keeps serving requests it can handle.
1958
Running a Minecraft Server and more on a 1960s UNIVAC Computer
18 bit registers, 90kb of memory, ones’ complement, powered by RISC-V.
1959
DNSFS – Store files in others' DNS resolver caches
A while ago I did a blog post about how long DNS resolvers hold results in cache for, using RIPE Atlas probes testing against their default resolvers (in a lot of cases, the DNS cache on their modem/router).
1960
When TCP sockets refuse to die
While working on our Spectrum server , we noticed something weird: the TCP sockets which we thought should have been closed were lingering around. We realized we don't really understand when TCP sockets are supposed to time out!
1961
Making an embedded Linux computer
Designing and soldering a small ARM Linux board with 217 ball BGA package and 0402 passive components
1962
Four features that justify a new Unix shell
On a lobste.rs thread about the rationale for the Fennel language, I posted this summary of why Oil exists:
1963
Linux NILFS file system: automatic continuous snapshots
In this article, I present the Linux file system NILFS and its automatic continuous snapshoting system.
1964
Random number generator enhancements for Linux 5.17 and 5.18
The random number generator has undergone a few important changes for Linux 5.17 and 5.18, in an attempt to modernize both the code and the cryptography used. The smaller part of these will be released with 5.17 on Sunday, while the larger part will be merged into 5.18 on Monday, which should receive its first release…
1965
1966
1967
RISC-V OS using Rust: Filesystems
I have taught operating systems at my university, so I will link my notes from that course here regarding the virt I/O protocol. This protocol has changed over the years, but the one that QEMU implements is the legacy MMIO interface.
1968
Books: an immutable double-entry accounting database service
Tracking financial transactions at scale
1969
1970
Console, the vital tool that Apple abandoned
Console is the epitome of Apple’s problem with tools and support for its users, and is exactly what it needs to fix.
1971
1972
1973
1974
Hello, Youki
First release of youki, an OCI container runtime written in Rust
1975
PyPy-STM: first “interesting” release
Hi all, PyPy-STM is now reaching a point where we can say it's good enough to be a GIL-less Python. (We don't guarantee there are no more...
1976
WSL3 Performance is about 5-60% faster than WSL2 depending on the workload
With the release of the WSL 3.x stack, Microsoft bumped the default virtualization runtime and advanced the shipped guest kernel from the 6.6 LTS branch up to 6.18. WSL upgrades often claim generalized performance gains, but marketing bullet points rarely translate directly to developer workloads. To see what actually…
1977
1978
Announcing KataOS and Sparrow
We've open sourced several components for our secure operating system, called KataOS, on GitHub, as well as partnered with Antmicro
1979
Glory is only 11MB/sec away (2023)
One thing is certain in life: if you’re getting something for a steal, someone is getting screwed, and it’s probably you elsewhere. This isn’t an anti-capitalist rant, it’s just the truth. Companies don’t exist if they don’t make money. Unless there’s some innovation, rest assured that you’re overpaying somewhere else.
1980
Jepsen: TigerBeetle 0.16.11
TigerBeetle is an Online Transactional Processing (OLTP) database built for double-entry accounting with a strong emphasis on safety and speed. It builds on the Viewstamped Replication (VR) consensus protocol to offer Strong Serializable consistency. Unlike general-purpose databases, TigerBeetle stores only accounts…