Subscribe
Sign in
Home
Caching
System Design
Database Internals
Breaking Prod
Archive
About
Latest
Top
Discussions
How Facebook built Cassandra
What happens when availability and throughput becomes the system's top priorities?
Aug 9
•
Sahil Sarwar
17
July 2025
Breaking Prod #2 - Building Search at Discord
Understanding the architecture that enables search at global scale
Jul 26
•
Sahil Sarwar
17
4
Eventually Consistent #1 - Learning vs Understanding
We don't have a learning problem, we have a understanding problem!
Jul 19
•
Sahil Sarwar
15
2
How gRPC works: Rebuilding gRPC from First Principles
A deep dive into the design, internals, and real-world performance of modern RPC systems
Jul 12
•
Sahil Sarwar
22
Why gRPC Is Fast: The Real Reason Is HTTP/2, Not Just Protobuf
How HTTP/2 enables true parallelism over a single connection
Jul 5
•
Sahil Sarwar
138
June 2025
Breaking Prod #1 - How Meta Manages Millions of Shards
Meta’s internal system for shard lifecycle, failover, and load rebalancing
Jun 28
•
Sahil Sarwar
16
4
Why RocksDB, Cassandra, and LevelDB Don’t Use B+ Trees
Comparing LSM Trees and B+ Trees Through Real-World Systems and Workload Patterns
Jun 21
•
Sahil Sarwar
23
When We Cross 1000 Subscribers
Where this newsletter is going — and why I might stop writing if it ever changes
Jun 18
•
Sahil Sarwar
4
Time and Order in Distributed Systems
Breaking down the concept of time in a network of machines
Jun 14
•
Sahil Sarwar
18
2
Why grep is Fast: NFA, DFA, and Regex Simplified
My most technical post, ever!
Jun 7
•
Sahil Sarwar
12
2
May 2025
Databases and Connection Pools
Why more open connections can degrade your queries?
May 31
•
Sahil Sarwar
53
2
500 Subscribers: A Quiet Reflection
Thank you to everyone joining along!
May 24
•
Sahil Sarwar
4
2
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
