Tooling
hyperfine Will Tell You the Build Got Slower and By How Much
One cold cache run says the build is slow, and the cold cache run is the outlier. Statistical benchmarking with warmup and repeats turns 'it feels slower'…
Deep Dive
Why an Index Only Scan Still Read the Heap
The plan said Index Only Scan, the buffers said the heap was read anyway, and the win never arrived. Index only is conditional on the visibility map, and…
Bug Hunt
The Memory Leak That Lived in the Metrics Pipeline
The scraper's memory grew without bound and no application leaked. Someone put a user ID in a Prometheus label, and every unique value became a permanent…
Performance
Event Loop Lag Is the Metric Nobody Alerts On
CPU at forty percent, memory fine, and requests timing out. The Node process is not busy, it is blocked, and the only metric that shows the difference is…
Tooling
You Are Recording the DevTools Performance Profile Wrong
A profile recorded with a warm cache, no throttling and a mid session state measures none of the conditions your users meet, and the optimisation built on…
Performance
The Web Font That Made Your Page Invisible for Two Seconds
With font display block, the browser renders nothing where the text should be until the font arrives, and on a slow network the page is a blank skeleton…
Tooling
Ten bpftrace One Liners That Replace a Week of Guessing
Latency without CPU, stalls without errors, syscalls nobody counted. The kernel already observes all of it, and bpftrace turns the observation into a one…
Performance
Two Hundred Redis Calls, and Every One Was a Round Trip
Each Redis command took a tenth of a millisecond to execute and a full round trip to deliver, and two hundred of them in a loop made the request slow while…
DevOps
A 1.2 GB Image That Should Have Been 90 MB
The image shipped the compiler, the package cache and the build tools to production, because every layer ships.
Code Review
Reviewing a Cache Layer for the Four Ways It Will Hurt You
A cache is reviewed for correctness on the happy path and fails on the edges: the cold start, the hot key, the poison entry and the deploy.
Performance
The Endpoint That Spent Nine Hundred Milliseconds Building JSON
The query took forty milliseconds and the endpoint took nine hundred and forty. The difference was serialisation: building and stringifying forty megabytes…
Bug Hunt
One Cache Key, Two Customers, and a Very Bad Morning
Two accounts selling the same product slug saw each other's cached pricing. The key was built from the thing that was shared, and the thing that differed…