2 ms·
A Terabyte of data can be still vast, depending on how different and unstructured they are. If you have thousands of different logfiles-formats, datadump-format
by PurpleRamen 3y ago
A Terabyte of data can be still vast, depending on how different and unstructured they are. If you have thousands of different logfiles-formats, datadump-formats and various complex fileformats, then you need a very different solution than you would need to handle a terabyte of uniform data in a csv-file or similar.
- hobs 3y agoSpark wont inherently help you solve that problem though, and neither will most of the big data solutions.