> An ODBMS will not solve your problems. 900M is quite a bit of records, > so I can see that you'll quickly run into problems there. I would > recommend approaching this problem with the usual optimization tricks.
We're wandering further off-topic for either flash or open source, but I just wanted to second Evert's suggestion. Once past a million records I often find I'd rather keep the data in csv (or custom binary) format, and just have summaries and views in the SQL DB. The downsides are having your data in two formats and needing to write scripts to convert between those formats. It can be a messier solution. But much, much faster. Some concrete examples. If the records are finance ticks you might keep 1 month of ticks in the DB, then the previous 6 months are just kept as 1 minute bars, the previous 5 years are 1 hour bars, and anything older is daily bars. (But still keep the ticks as disk files.) If the records are web server logs, then each 24hr log file can be analyzed once it is over 6hrs old, archived, and only the analysis stored in the DB. (The 6hr wait is to handle visitor analysis for visits that cross the midnight boundary.) Darren -- Darren Cook, Software Researcher/Developer http://dcook.org/mlsn/ (English-Japanese-German-Chinese-Arabic open source dictionary/semantic network) http://dcook.org/work/ (About me and my work) http://dcook.org/blogs.html (My blogs and articles) _______________________________________________ osflash mailing list [email protected] http://osflash.org/mailman/listinfo/osflash_osflash.org
