Dazhong Xia
Dazhong Xia
### Overview Right now, every time we start a new worker process (i.e. for every op), we have a several-second lag between when the process starts and when all the...
### Success Criteria How will we know that we're done? * [x] measured initial peak memory usage: * [x] identified memory usage hotspots * [ ] resolved or decided to...
### Overview This asset is marked as `"memory-usage": "high"`. ### Success Criteria How will we know that we're done? * [x] measured initial peak memory usage: * [x] identified memory...
### Overview This asset is marked as `"memory-usage": "high"`. ### Success Criteria How will we know that we're done? * [x] measured initial peak memory usage: * [x] identified memory...
### Overview This asset is marked as `"memory-usage": "high"`. ### Success Criteria How will we know that we're done? * [x] measured initial peak memory usage: * [x] identified memory...
### Overview This asset is marked as `"memory-usage": "high"`. ### Success Criteria How will we know that we're done? * [x] measured initial peak memory usage: * [x] identified memory...
For the purposes of our initial performance push, #4618 is sufficient. But it would be good to be able to continue to find our worst performance offenders without having to...
### Overview When updating the ETL, it is slow or impossible to rematerialize the affected assets locally. This makes ETL updates take much longer than necessary. The current workaround for...