shakedown.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A community for live music fans with roots in the jam scene. Shakedown Social is run by a team of volunteers (led by @clifff and @sethadam1) and funded by donations.

Administered by:

Server stats:

267
active users

#distributedsystems

0 posts0 participants0 posts today

Constellation update: the system is at the point of a bare-minimum end-to-end demo. I can submit a basic request to a peer, have it send to a consensus pool (if needed), then to a processor, then report the result back.

The "last mile" of code is janky and has lots of tech-debt to get to this point on a certain timeline, but it's there. The core architecture is mostly sound, though.

We're #hiring!

Two(!) full #professorships open in our department at WU Vienna (Vienna University of Economics and Business) under two complementary focus topics:

1) #Foundations of contemporary #InformationSystems, where we look for candidates who complement and strengthen the existing research at our department in areas such as:

· #ArtificialIntelligence: #AI Systems and Architectures
· #DataMining and #MachineLearning
· #DistributedSystems and #Decentralization
· #DistributedLedgers
· #Cloud and #Virtualisation
· #IoT and #EdgeComputing
· #DataGovernance for AI

2) #OperationsManagement with a focus on #DigitalTransformation, where the candidate’s expertise falls within one of the following research areas:

· #behavioural #operations
· AI application to #process improvements
· integrated #supplymanagement and #demandmanagement
· #ProductionPlanning and control
· #SupplyChain planning and control
· circular supply chains and sustainable supply chain management
· #tokenization in supply chains and new product development

Details at the link below... Please get in touch, if you want to know more!

wu.ac.at/en/isom/events/isom-n

www.wu.ac.atopen positions: 2 full professorships of Business Administration and Informaton SystemsWe are looking for applicants with an emphasis on either * contemporary information systems or * operations management and digital transformation.

Constellation update:

I resumed principal development around Christmas. The first effort was some technical debt resolution, to clean up after the mad dash to get it to MOC.

Finished that up this morning. This greatly improves some of the main APIs and puts me in a much better place moving forward.

Next batch of work will focus on core capabilities I'll need to get to initial operating capacity (IOC).

ITT: future-facing visions for a distributed OSS ecosystem.

So I've spoken a bit about Constellation, which is my (very) work-in-progress project to hopefully provide the infrastructure for where I see things like the Fediverse and the OSS movement going: towards a more decentralized, participatory alternative.

I'm a systems, infrastructure, and core problems person, so this reflects that perspective.

1/..

Pick Your Distributed Poison

"I see you enjoy the wonders of non-deterministic metastability that comes from adaptive concurrency controls. Oh, you don’t? So you have hard isolation between the two systems? I see. That gets you non-deterministic metastability but without needing adaptive concurrency controls. Fascinating innit?”

#distributedSystems

Great stuff from @hazelweakly

hazelweakly.me/blog/pick-your-

[edit: typo fix in quote]

Hazel WeaklyPick Your Distributed Poison | Hazel WeaklyOne of the hardest things for people to understand with distributed systems is that eventual consistency is the same thing as eventual inconsistency. The very...

This is a fantastic read about distributed systems: hazelweakly.me/blog/pick-your-

…but what's smacking me in the face right now is this bit:

"Reproducible and bootstrappable systems get a lot of love among neurodivergent people. For good reason: they’re very friendly to those with little working memory but vast amounts of working context”

I’m very bad at keeping things in my head but I'm very good at knowing where the thing I'm missing needs to be swapped in from.

Hazel WeaklyPick Your Distributed Poison | Hazel WeaklyOne of the hardest things for people to understand with distributed systems is that eventual consistency is the same thing as eventual inconsistency. The very...

Today in papers I love to make fun of: time, clocks, and the ordering of events in a distributed system.

lamport.azurewebsites.net/pubs

In classic Lamport fashion, introduces a foundational concept in distributed systems deserving of a Turing award with impressively little help to the reader to understand its implications, with bonus digressions to special relativity to show off how smart the author is.

But with such good insights underlying it.

Today is International Beaver Day, so there is no better time to tell you about my upcoming #osdi24 paper "Beaver: Practical Partial Snapshots for Distributed Cloud Services"... using pictures of cute beavers.

This project was led by Liangcheng Yu, along with his advisor Vincent Liu and other colleagues at UPenn. (1/6)

(photo from en.wikipedia.org/wiki/Beaver#/)

For part 2 of the Understanding Latency Series, we are doing a bit of mixing and matching.

Perhaps you have heard of "Tail At Scale", the CACM article that periodically made the rounds in the last decade. In this post, we provided interactive charts so you can play with that kind of setup and *see* for yourself the profound effect of simply issuing multiple requests at once.

iop.systems/blog/composite-lat

iop.systemsHedging or Scattering? A Mixed Bag of Distributed Latencies When a service depends on other servies, their latencies can come together in surprising ways to affect the overall user experience.
Hi there!

New account, new #introduction!

I'm Adrian, a french #dev, interested by everything #Elixir, #DevOps and #DistributedSystems

The past few months I've worked a bit on platform engineering for my current company, but before that I've developed a tool to send USSD messages with a Raspberry PI through an API, #Livebook or gRPC call using Elixir magic :D

I'm here to keep exchanging about #tech, keep learning and maybe teaching stuff!

I've also began a blog, but I have yet to write more posts :)

On the personal projects side, I have a few ideas but I've not started yet!

I will (try to) write in English here, but you may see me answer in French from time to time though!

#MyElixirStatus
genserver.socialAkkoma

Really interesting to see how fast, cheap content addressable stores are changing how we build CI and dev tools. If you can load or store just about anything a tool might produce in a few ms, knowing it's content digest only, you start to build differently.

Not just in build systems but anything that needs access to the artifacts: generated code, bytecode, logs, errors, object files, binaries.

This has been a trend for a while but I think the costs are so low now it becomes possible to build multiple layers of tools reusing the remote distributed cache

Federated link aggregators should not own individual topics. Topics should belong to the entire fedi.

Instance owned topics is just another form of centralization.

We should not only discourage topic diaspora (e.g., a "gaming" topic on beehaw is separate from a gaming topic on another instance) but engineer solutions to *allow* and default unification of content by topic across the fedi for link aggregation (but not force this).

(De)federation then becomes how we keep topics safe for our communities: blocking posts and comments from demonstrably unsafe instances.

This is a CAP distributed system problem. My instincts say that federated link aggregation is a different enough problem from federated blogging (mastodon) that we need federated instances to share content via a consensus algorithm and not the mastodon style of federation.

I wonder about the practicalities of making this work in the truly heterogenous system of the #fediverse.

Perhaps by having the federation members/instances themselves share capability and capacity metadata with the cluster so as to load balance intelligently? The alternative could be an unintentional DDOS due to a massive instance going offline, shedding traffic to instances too small to handle the load.