Dan Crankshaw

Results 27 comments of Dan Crankshaw

What is the current status of this? @ryanhoque were you able to get this finished before the end of the semester?

Ahh this is an excellent point. Thanks for bringing this to our attention.

@chester-leung is going to try to reproduce this on OSX.

Unfortunately this design doc has not been written. The authoritative source for our Redis configuration schema is in [redis.cpp](https://github.com/ucbrise/clipper/blob/develop/src/libclipper/src/redis.cpp), specifically in the `add_*` methods (add_model, add_container, etc).

@daminig is working on this. She has an implementation already and is just ironing out the testing. We'll have a PR up next week.

Yes it does. You can create [multiple replicas of a model](http://docs.clipper.ai/en/v0.3.0/clipper_connection.html#clipper_admin.ClipperConnection.set_num_replicas) and Clipper will automatically load-balance requests between them. However, Clipper does not yet support deploying models on GPUs automatically....

Agreed. Using something like memcached is likely the right design choice here. As @dbczumar points out, the query frontend is pseudo-stateful right now, which is not a desirable property of...