RE: Latency in asynchronous mode

[Ibrahim <i....@newcastle.ac.uk>]

Hi Michael,

Sorry for late replay.

Your question is:

What does the 392 represent?
The number 392 represents the highest latency when Zookeeper using group
commit (fsync). For example, assume we send writes requests (r1, r2,
r3,...., r1000), also assume zookeeper does fsync per 1000 requests. In this
case, all incoming requests will queue until the queue size reaches to 1000
(batch size), then the requests will flush into disk and continue complete
the process. This means, each request will wait until the batch size become
1000. This can lead to increase the latency (but improve the throughput). 
To conclude, such high latency (392 ms) causes by batching. As example, the
time 392 generates by subtracting the response and submitted time of r1
(request 1). 

Does this make sense?

Looking forward to hearing from you.
Thank you

Ibrahim

The interesting thing is the min values of the latency numbers of the async.
1ms?
But that's a different issue.

So lets start there. 



--
View this message in context: http://zookeeper-user.578899.n2.nabble.com/Latency-in-asynchronous-mode-tp7580446p7580556.html
Sent from the zookeeper-user mailing list archive at Nabble.com.