Created on 06-12-2017 05:20 AM - edited 09-16-2022 08:22 AM
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
Created 06-12-2017 05:39 AM
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.
Created 06-12-2017 05:39 AM
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.
Created 06-12-2017 06:07 AM