Hi Xiang For your question #2, assuming ShellBolt is used, the execute method just creates the message to send and stores it before exiting, hence the low latency shown on the UI - see https://github.com/apache/storm/blob/master/storm-core/src/jvm/org/apache/storm/task/ShellBolt.java#L148
You could specify a value for Config.TOPOLOGY_SHELLBOLT_MAX_PENDING which should hold the incoming tuples into the execute method until there is space in the message buffer. This will give you more realistic execute latency figures. HTH Julien On 7 April 2016 at 09:03, Xiang Wang <[email protected]> wrote: > Hi All, > > I got several questions regarding storm ui and multi-lang processing. > > 1. Is there any APIs to get the storm statistics, e.g., execute latency, > capacity, as shown in the storm UI? > I would like to get those statistics automatically using a program and > save them to a file such that I don't need to monitor the UI all the time. > It would be great if you could provide an example. > > 2. How does storm compute the statistics for multi-lang bolt (I use c++)? > The statistics of my multi-lang bolt in UI is very strange. > For example, the expected execute latency of my multi-lang is about 2ms > (which can be computed inside c++), while the actual value shown in the UI > is only 0.002ms, which is extremely small... > > Could someone give me some comments? Your thoughts will be much > appreciated. > Thanks. > Best, > Xiang > > > ------------------------------- > Xiang Wang PhD Candidate > Database Research Group > School of Computer Science and Engineering > The University of New South Wales > Sydney, Australia > -- *Open Source Solutions for Text Engineering* http://www.digitalpebble.com http://digitalpebble.blogspot.com/ #digitalpebble <http://twitter.com/digitalpebble>
