tag:help.beatunes.com,2009-07-24:/discussions/questions/5205-avoid-resending-data-when-restarting-analysis-taskbeaTunes: Discussion 2017-06-05T20:49:02Ztag:help.beatunes.com,2009-07-24:Comment/426928472017-06-04T10:45:52Z2017-06-04T10:45:52ZAvoid resending data when restarting analysis task<div><blockquote>
<p>Analysis starts by sending data to the server.</p>
</blockquote>
<p>It only does that, if you have selected a task that uses online resources.</p>
<blockquote>
<p>When there are a lot of files this can take quite some time.</p>
</blockquote>
<p>Processing results from the server may actually start <em>before</em> all data is sent to the server.</p>
<blockquote>
<p>When analysis is resumed after a program restart, the data for the files in the task queue needs to be resend.</p>
</blockquote>
<p>Indeed.</p>
<blockquote>
<p>When analysis is paused this is not the case.</p>
</blockquote>
<p>Correct, because the connection to the server is kept alive. If it gets lost, some data from the server may simply be dropped.</p>
<blockquote>
<p>Would it be possible (desirable?) to keep the sent data available/alive through a program restart(even for a limited amount of time)?</p>
</blockquote>
<p>You mean, produce the data that is going to be sent to the server, save it to a file and re-use that same data, if the analysis is resumed?</p>
<p>How would we know, whether the data is still "fresh"?</p>
<p>Also, the issue may not be sending... it may be processing on the server side. I.e. how fast can the server actually send reasonable answers?</p>
<p>You can help the server a little by using <code>Submit Library to Server</code> under <code>Tools</code>. beaTunes will then send your song data to the server and connect your data to reference data, which is a big step of what's happening on the server side. Any subsequent analysis won't have to do this anymore. The connection is id-based, i.e. your username, name, email address etc. cannot be connected to this link.</p>
<p>I can understand the desire to speed up this part of analysis, but believe it's a little too late in the dev cycle to change something so fundamental. Also, I'm not sure how I'd want to change it.</p>
<p>Perhaps not using tasks with online resources is something you want to try?</p></div>hendriktag:help.beatunes.com,2009-07-24:Comment/426928472017-06-05T19:56:52Z2017-06-05T19:56:52ZAvoid resending data when restarting analysis task<div><p>Thanks for clarifying. Not a practical idea for the reasons you point out.</p>
<blockquote>
<p>How would we know, whether the data is still "fresh"?</p>
</blockquote>
<p>Indeed.</p>
<blockquote>
<p>You can help the server a little by using Submit Library to Server under Tools. beaTunes will then send your song data to the server and connect your data to reference data, which is a big step of what's happening on the server side. Any subsequent analysis won't have to do this anymore.</p>
</blockquote>
<p>Good to know, I submitted library to the server.</p>
<p>How about these msgs: "beaTunes was unable to identify the track xyz, because it is not yet in the database."<br>
I'd expect them not to occur anymore after submitting the library to the server, or is this something different?</p></div>ErikLtag:help.beatunes.com,2009-07-24:Comment/426928472017-06-05T20:48:59Z2017-06-05T20:48:59ZAvoid resending data when restarting analysis task<div><blockquote>
<p>How about these msgs: "beaTunes was unable to identify the track xyz, because it is not yet in the database." I'd expect them not to occur anymore after submitting the library to the server, or is this something different?</p>
</blockquote>
<p>Not all of your songs may have been processed yet. That said, the error message is shown, when you are trying to import data from the central database. Even if you submit your data to the database, it may still return that message, if there is no reliable reference data for your song. This can happen, when it is not in certain reference database yet and hasn't been submitted by multiple beaTunes user's yet.</p></div>hendrik