Papershelf
Research papers I’m reading to build better intuition around backend systems, distributed systems, databases, and infrastructure.
Paper
A readable path into consensus, leader election, replicated logs, and the edge cases hidden inside “understandable.”
Leader election, log replication, safety rules, and failure timing.
Paper
A clean model for splitting large jobs across machines while still handling scheduling, retries, stragglers, and locality.
How a simple programming model hides the hard parts of distributed execution.
Paper
A storage paper about tablets, locality, compaction, and the shape of a distributed database built on top of a file system.
Tablet serving, compaction, data locality, and structured storage at large scale.
Paper
A practical look at building a distributed file system for large sequential workloads, failures, chunk servers, and master metadata.
Fault-tolerant storage, chunk placement, master metadata, and large-scale file access patterns.
Paper
A paper about availability-first storage, quorum tradeoffs, conflict resolution, and what happens when uptime is the main constraint.
Availability-first design, replication tradeoffs, conflict handling, and operational complexity.