build1 publisherOne report Mercari's DBRE team overrides its TiDB autoscaling rules by time of day and campaign, and a cluster with a 50% CPU target now averages 44%. Extending the same controller to TiKV storage meant designing around a TiDB Cloud lock that refuses a second scale operation while one is running.
Publishers:engineering.mercari.com
Reality
- Evidence40
- Adoption30
- Hype gap+5
- Incentives25
- Confidence50
build1 publisherOne report Sourcegraph's new migration agent scopes a codebase-wide change from one prompt and pauses at every stage by default. Fixing its own red builds is a separate configuration step that needs a token which can read your CI logs.
Publishers:sourcegraph.com
Reality
- Evidence38
- Adoption32
- Hype gap+33
- Incentives88
- Confidence60
build1 publisherOne report Mercari's DBRE team found that on tables with a composite or non-integer primary key, TiDB reads every matching index entry and then sorts. The upstream fix is in master, not in the current LTS.
Publishers:engineering.mercari.com
Reality
- Evidence72
- Adoption38
- Hype gap−12
- Incentives32
- Confidence61
build1 publisherOne report Region split size cut region count about 10% and TiKV CPU up to 20%, and the In Memory Engine took 0.5-1.5s queries to roughly 0.1s. Plan cache now runs on one cluster.
Publishers:engineering.mercari.com
Reality
- Evidence60
- Adoption55
- Hype gap−10
- Incentives40
- Confidence62
build1 publisherOne report Mercari's DBRE team found the query only after switching a batch read endpoint to TiDB. Query replay had run first and had not flagged it.
Publishers:engineering.mercari.com
Reality
- Evidence63
- Adoption58
- Hype gap−12
- Incentives42
- Confidence62