2 publicaciones etiquetados con "performance"

At WHOOP, some of our most demanding infrastructure challenges arise when we need to run a computation at population scale. Earlier this year, we needed to simulate one of our ML models across a large internal dataset to support a time-sensitive research workload. Each simulation used several weeks of historical inputs, and the total came out to roughly 15.8 million inference tasks.

The tool we had on hand was built for a much smaller job a couple of years ago. At its current per-task latency, completing the new run would have taken more than two months, far too slow to get researchers the data they were waiting on. We brought that down to six days. This post covers the changes that made this possible.

Late last year, Google communicated with us on an observation they had with our App in the Play Store: our Android sessions were being flagged for excessive partial wake lock usage. We did not see significant impact on our build. Battery life wasn't trending in support tickets, and app reviews showed nothing unusual. But the signal was there, and we worked with Google to dig in and fix it.

At the time, we knew something was off. What we didn't know yet was which background workers were responsible, or how wake lock time was distributed across them. Finding out would make Android connectivity more robust and insights delivered more efficiently to members.

🥊 Wake Locks: 1 — WHOOP: 0. Google flags ~15% of sessions. It did not appear to affect production.