Here is the paper that elaborates how Facebook exploit memcached to handle millions of requests per second: Scaling Memcache at Facebook
apr
I am confused about the relationship between increase in UDP request throughput and latency of responses. Why is it that the latency increases when trying to increase the throughput?
On the link at the source, it says "Since we’ve made all these changes, we have been able to scale memcached to handle 200,000 UDP requests per second with an average latency of 173 microseconds. The total throughput achieved is 300,000 UDP requests/s, but the latency at that request rate is too high to be useful in our system." -- is total throughput referring to the throughput when we add up all types of requests - both those which go to memcached and those which query the db directly?
UDP - User Datagram Protocol
Here is the paper that elaborates how Facebook exploit memcached to handle millions of requests per second: Scaling Memcache at Facebook
I am confused about the relationship between increase in UDP request throughput and latency of responses. Why is it that the latency increases when trying to increase the throughput? On the link at the source, it says "Since we’ve made all these changes, we have been able to scale memcached to handle 200,000 UDP requests per second with an average latency of 173 microseconds. The total throughput achieved is 300,000 UDP requests/s, but the latency at that request rate is too high to be useful in our system." -- is total throughput referring to the throughput when we add up all types of requests - both those which go to memcached and those which query the db directly?