Calum Chamberlain

On that note, while I'm playing around with it, I may have a look at memory usage for writing QuakeML files too - currently it seems like quite an expensive...

Okay, I have done some timing, and initial playing, with little success. The results are in the gist I posted earlier - re-linked [here](https://gist.github.com/calum-chamberlain/a455a6f38a8dc6360ed0e52498d7799e). The main point is - calls...

Ah, fair enough - will not pursue that strand any further then. I think some more reading on my part of useful ways to read in large xml files needs...

So after talking with @krischer at SSA, I finally had time to try a different implementation that converts the xml to a dictionary immediately, then makes the objects from the...

> Can you also try to pickle and un-pickle the same event on the same machine? I am curious how the un-pickling performance compares. Not sure I know what you...

@d-chambers Ah, yes, I was confused, I think because there are objects called `Unpickler` and `Pickler` in `io.quakeml.core` that do the read/write operations. I will test and report back. Also...

So I ran some profiling, comparing my hack to obspy 1.1.0, and to raw pickling. The notebook is [here](https://gist.github.com/calum-chamberlain/5687bfab1890c7385dab4bb8e4fd10a6). I initially worked on a larger catalog, but ended up not...

Thanks for those comments @flixha I agree and will get onto them! I'm not fond of setting the process spawning method, even in a context manager, mostly to ensure that...

An additional speed-up for party reading could be to read the parts of the party (Tribe, detection catalogs, detection stats) in separate processes, before finally reconstructing the party at the...

Are you running that as a script or from an iPython session? In general when running scripts that use multiprocessing libraries you should always encapsulate your code in an `if...