What's correctness of a system?
By what properties can we guarantee correctness?
#distributedsystem #systemdesign #java
What's the difference between reliability and fault tolerance?
18
Search
Your #AIAgent Is Already a #DistributedSystem — You Just Didn’t Design It That Way by Zenefa Rahaman, PhD medium.com/data-science-c…
11
纯 Go实现的轻量级的分布式消息队列 miKago 🚀。 没有 JVM，抛弃 ZooKeeper，完全依赖内置的 Raft 算法手搓共识集群； 支持原汁原味的 Kafka 客户端二进制协议、Segment 磁盘稀疏索引，并打通底层 sendfile 实现了真正的 Zero-Copy 零拷贝下发！#Golang #Kafka #DistributedSystem⚡️ 代码如下github.com/ldehai/miKagoU
1
241
Real-time voice AI is a systems engineering challenge.
Diving into #Pipecat has highlighted how much work goes into the orchestration of a live conversation. You’re essentially building a real-time #distributedSystem where every millisecond counts.
#VoiceAI #SoftwareEngineering
2
1
2
24
Replaced etcd with embedded Raft in Danube Messaging using openraft + redb.
No external metadata service, simpler ops, faster failure detection.
danrusei.github.io/posts/migrate_…
#OpenSource #DistributedSystem #PubSub #Streaming
7
The output buffer of one agent becomes the input of another. Authority moves horizontally through the network. No user prompt box involved. This is agent-to-agent injection operating beneath conventional monitoring.
…pendent-research-systems-modeling.com/moltbook-incid…
#SystemArchitecture #DistributedSystem
12
This blog cover Caching Strategies in Distributed System as a part of my System Design Cohort.
Link: cache-strategies.hashnode.dev
#systemdesign #caching #distributedsystem
@piyushgarg_dev @Hiteshdotcom @nirudhuuu
@ChaiCodeHQ
2
21
Bayou (1995) embraced this instead of fighting it allowing updates on disconnected replicas, then reconciling later using version histories, dependency checks, and application-defined conflict resolution.
#DistributedSystem #Technology
1
8
What are the 3 fundamental ideas every dev should know if they're unfortunate enough to work on a #DistributedSystem?
@samnewman breaks down the essentials:
1️⃣ Timeouts – giving up
2️⃣ Retries – trying again
3️⃣ Idempotency – making it all a bit safe
🎥bit.ly/4lGml2meF
27
185
10K
If your distributed system relies on wall-clock time, it’s already wrong.
Clocks drift across machines, so “last write wins” becomes luck.
Systems like DynamoDB use logical/vector clocks to track causality — who caused what — not timestamps.
#DistributedSystem #SoftwareArchitect
10
2PC taught us a core lesson of distributed systems: safety without liveness doesn’t scale.
#distributedsystem #database
1
23
If #Loki feels heavy for a small cluster, keep it boring. Use lightweight log shippers like #FluentBit or Vector to push logs straight to a managed backend or simple storage. Fewer components, less tuning, faster signal. You don’t need #distributedsystem to read yesterday’s logs.
#Loki can feel heavy for small #Kubernetes clusters. It shines at scale, but for tiny setups memory, storage tuning, and ops overhead can outweigh benefits. If you’re spending more time babysitting logs than fixing issues, simpler log shipping might be the smarter move. #Grafana
3
157
Synchronization Enhances Cross-Node Coordination
Improved timing supports stable distributed operation
✓ Adaptive synchronization rhythm
✓ Reduced cross-node timing gaps
✓ Consistent system performance
🌐h5.67exus.com/#/homec
#67EX #NodeSynchronization #DistributedSystemG
7
21
4K