A bug that results in significant increase in memory and CPU usage is somewhat concerning, especially with the sunset of the v201109 API fast approaching. A response from the AdWords API team on this would be much appreciated.
Cheers On Wednesday, 17 October 2012 13:10:31 UTC-5, j.e.frank wrote: > > Yes. All I care about is whether the account had any activity (i.e. > impressions) over a given date range, so an empty report means that it's > not active, for my purposes. I don't care if the account is "active" in > the sense of being available to run. > > On Wednesday, October 17, 2012 12:09:30 PM UTC-4, Oliver wrote: >> >> Thanks for sharing such a great optimization tip. >> >> Aside from the performance issue, I am curious to know how your're >> handling the account reports when there are no impressions. There is a >> known bug with the Account Performance Report whereby it comes back >> completely empty if there is no impression data. Are you parsing an empty >> report as non-active account? >> >> Oliver >> >> >> On Wednesday, October 17, 2012 3:13:16 PM UTC+1, j.e.frank wrote: >>> >>> I noticed this post that was also related to report serialization and >>> jaxb: >>> >>> https://developers.google.com/adwords/api/community/?place=msg%2Fadwords-api%2F2h_hhyThed0%2FN94XKOfR2GoJ >>> >>> I made my own version of the serializer for a ReportDefinition object, >>> that creates a single JaxBContext object and reuses it instead of making a >>> new one each time. According to some web searching that I did, the >>> JaxBContext is threadsafe so this should be fine, and in fact the >>> recommended approach is to use just one JaxBContext. This change made a >>> significant reduction in my CPU utilization, in addition to the permgen >>> memory issues that Eli mentions in the post above. >>> >>> On Tuesday, October 16, 2012 4:18:31 PM UTC-4, j.e.frank wrote: >>>> >>>> I'm upgrading from v201109 to v201209 in java. I have a method which >>>> figures out all of our "active" accounts over a given time frame, by >>>> running a simple report for each account and seeing if there are any >>>> impressions. I tested it out and it works fine but my CPU is pretty much >>>> pinned at 100% while that method is running. I went back to the earlier >>>> version and it doesn't get much above 20%. >>>> >>>> At first I thought it was a difference from the "old" java library to >>>> the "new", so I also tried it with v201209 in the "old" library and had >>>> the >>>> same issue. I haven't done profiling yet to figure out what the exact >>>> difference is, so I was wondering if anyone else has experienced a similar >>>> issue. My first theory is that it has something to do with >>>> serializing/deserializing, since I know that has changed significantly >>>> with >>>> jaxb. >>>> >>> -- =~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~ Also find us on our blog and discussion group: http://adwordsapi.blogspot.com http://groups.google.com/group/adwords-api =~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~ You received this message because you are subscribed to the Google Groups "AdWords API Forum" group. To post to this group, send email to adwords-api@googlegroups.com To unsubscribe from this group, send email to adwords-api+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/adwords-api?hl=en
