Hello people. Is it really 10 months since my last post? I really need to pull my finger out. Sorry guys!

Anyway, onto the subject at hand. Why is the XML Serializer so piss poor?

Well, apart from being text-based meaning it’s quite large and bulky to begin with, it also has a massive overhead trying to work out what is what, what is not and how it’s going to store everything. We were working on a large(ish) quantity of data (about 150mb) and getting load times of 10+ seconds and save times of up to 40 seconds. Not ideal for a product that needs almost instant response.

A little bit of googling and I came across the following article in Code Project:

http://www.codeproject.com/Articles/15700/Optimizing-Serialization-in-NET

Ok, it means a little more work in understanding what the guy has done but my goodness! We got data serialization down from 40+ seconds to around 180ms! That was jaw-dropping!

The key is using the binary serializer and the library abstracts the detail quite simply so you literally tell the serializer what you want to save, what order and let it do its thing. The author deserves an award. The code isn’t rocket science and the implementation is very straight forward but the results are simply amazing.

Be careful though, you need to pay attention to the order and structure of your data but once written, it simply does it! Brilliant work Simon Hewitt!

VN:F [1.9.22_1171]
Rating: 0.0/10 (0 votes cast)
VN:F [1.9.22_1171]
Rating: 0 (from 0 votes)