DEV Community

#distributedsystems

Topics related to systems where components are on different networked computers.

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Consistent Hashing: The System Design Interview Concept Everyone Gets Wrong

Consistent Hashing: The System Design Interview Concept Everyone Gets Wrong

1
Comments
2 min read
sync = true protects one JVM, not the cluster

sync = true protects one JVM, not the cluster

1
Comments
1 min read
The Bug Where a Customer Got Charged Twice for Nothing They Did Wrong

The Bug Where a Customer Got Charged Twice for Nothing They Did Wrong

Comments 1
7 min read
Dynamic machine ID leases in Elixir

Dynamic machine ID leases in Elixir

Comments
5 min read
Dev log #21 Hardening WebRTC-Direct and solving SCM leaks

Cross-language WebRTC and GitHub API polling fixes

Dev log #21 Hardening WebRTC-Direct and solving SCM leaks

39
Comments 5
5 min read
Your Order Fulfillment Workflow Is One 24-Hour Wait Away From Chaos

Your Order Fulfillment Workflow Is One 24-Hour Wait Away From Chaos

Comments
3 min read
Distributed Locking in Practice: Guarantees, Failure Scenarios and Better Alternatives (3/4)

Distributed Locking in Practice: Guarantees, Failure Scenarios and Better Alternatives (3/4)

Comments
6 min read
WebRTC at Scale

WebRTC at Scale

2
Comments
10 min read
The wait queue is just a channel: building a small distributed lock server in Go

The wait queue is just a channel: building a small distributed lock server in Go

1
Comments
7 min read
Distributed Locking in Practice: Guarantees, Failure Scenarios and Better Alternatives (2/4)

Distributed Locking in Practice: Guarantees, Failure Scenarios and Better Alternatives (2/4)

Comments
8 min read
Why HTTP 429 Should Not Count Against a Job's Retry Budget

Why HTTP 429 Should Not Count Against a Job's Retry Budget

Comments
2 min read
The Outbox Pattern Is Not Enough

The Outbox Pattern Is Not Enough

Comments
7 min read
Your Database Said "Success." Your Message Broker Said "Try Again."

Your Database Said "Success." Your Message Broker Said "Try Again."

Comments
10 min read
My idempotency library had one job. A dropped connection made it run the payment twice.

My idempotency library had one job. A dropped connection made it run the payment twice.

Comments
8 min read
A backend has to fail twice, and one failure is not an outage

A backend has to fail twice, and one failure is not an outage

Comments
3 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.