the Nodes Stats API in 0.19.x contains a lot of useful info, but I am still
missing something. If I read it correctly it can tell us how much of the
data has gone through transport layer (node / transport) and how much data
has been read or written to the disk (node / fs / data) but it does not
tell me how fast it has been able to do it.
In other words, I would love to see ES to be able to tell if underlying
network and general IO interface is fast enough to support fast search
responses. Is that possible?
How "fast" is basically the delta (simplest) or more complex computation
between one sample and the next. Its certainly something that can be
computed in ES as well, but stats is focused on providing the "raw" data
currently.
the Nodes Stats API in 0.19.x contains a lot of useful info, but I am
still missing something. If I read it correctly it can tell us how much of
the data has gone through transport layer (node / transport) and how much
data has been read or written to the disk (node / fs / data) but it does
not tell me how fast it has been able to do it.
In other words, I would love to see ES to be able to tell if underlying
network and general IO interface is fast enough to support fast search
responses. Is that possible?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.