I tweet/retweet interesting stuff about #DistributedSystems and #compsci. Suggest links/papers/conversations via DM! Tag for retweets. Run by @federico_ponzi
Joined January 2017
- Tweets3.6K
- Following223
- Followers6.4K
- Likes789
Big List of 20 Common Bottlenecks
highscalability.com/big-list…
In Zen And The Art Of Scaling, Sullivan offered an interesting conjecture: there are 20 classic bottlenecks. In practice we all know bottlenecks come in infinite flavors, all tasting of sour and ash.
Life Beyond Distributed Transactions
queue.acm.org/doi/10.1145/30…
Practical approaches used in the implementation of large-scale mission-critical applications in a world that rejects distributed transactions.
Git at any scale
cursor.com/blog/git-at-any-s…
Twenty years after Linus Torvalds created it, Git has become an industry standard. The average company rely on a centralized host. And hosting a Git repository, it turns out, is an incredibly hard thing to do.
On building scalable control planes
allthingsdistributed.com/202…
On the surface, the control plane looks quite boring: it records what should exist and reconciles that with what actually does.
Making 768 servers look like 1
planetscale.com/blog/making-…
A single database server cannot handle such demand, so we must spread the queries and data out across many servers with database sharding.
Client-Side Load Balancing at a Million Requests Per Second
engineering.zalando.com/post…
Aurora DSQL: Scalable, Multi-Region OLTP
brooker.co.za/blog/2026/07/1…
Paper: arxiv.org/abs/2607.13276
How TLA+ Caught a Silent Data Divergence Bug in Postgres’s pg_rewind
multigres.com/blog/tla-pg-re…
Queues Don't Fix Overload
ferd.ca/queues-don-t-fix-ove…
To oversimplify things, most of the projects I end up working on can be visualized as a very large bathroom sink. User and data input are flowing from the faucet, down 'till the output of the system[..]
Simple and Correct Snapshot Isolation
remy.wang/blog/si.html
Snapshot isolation (SI) is a popular approach to concurrency control in databases systems. It avoids many forms of anomalies, yet provides a high degree of concurrency, especially for read-heavy workloads.