Archives of Support Questions (Read Only)

This is an archived board for historical reference. Information and links may no longer be available or relevant
Announcements
This board is archived and read-only for historical reference. To ask a new question, please post a new topic on the appropriate active board.

Oryx 2 recommenadtion || increase Throughput

avatar
Frequent Visitor

Hi,

 

We wanted to increase the Throughput of the oryx server layer by using the sample rate(LSH) value 0.3 . Currently our configuration have the value of sample rate 1.0. Will it affect the recommendation of the users.

 

Total Unique Item count:1.7 lakh

Total Unique user count:16.8 Million

1 ACCEPTED SOLUTION

avatar
Master Collaborator

Will it affect recommendations? yes. However, it doesn't affect the results much until you turn the LSH value down a lot. For example, you may find that 0.1 still yields good recommendations.

 

With 1.7M items however, you should find it's already pretty fast. See http://oryx.io/docs/performance.html . Even with 250 latent features, at LSH=0.3, you could probably serve ~100qps on one modern server with latency ~15ms.

View solution in original post

2 REPLIES 2

avatar
Master Collaborator

Will it affect recommendations? yes. However, it doesn't affect the results much until you turn the LSH value down a lot. For example, you may find that 0.1 still yields good recommendations.

 

With 1.7M items however, you should find it's already pretty fast. See http://oryx.io/docs/performance.html . Even with 250 latent features, at LSH=0.3, you could probably serve ~100qps on one modern server with latency ~15ms.

avatar
Frequent Visitor

@srowen

 

Thank you so much for the quick response.