Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

SI - Site Forums : Silicon Investor - Welcome New SI Members! -- Ignore unavailable to you. Want to Upgrade?

To: Green Receipt who wrote (18910)	6/26/2003 1:35:38 PM
From: RockyBalboa	Respond to of 32871

agreed. The usage of blobs can be the problem. I have not too good memories on such things (a trading application using sybase databases and storing a whole lot of deal info in a "blob").

To: Green Receipt who wrote (18910)	6/26/2003 1:45:33 PM
From: SI Bob	Read Replies (1) \| Respond to of 32871

That's pretty close. Full-text searches are handled by a separate box. All of the machines rebooted sometime this week (probably an electrical outage that the IPS's worth UPS's didn't cover), and, being a Windoze machine with only a couple of services and one app to load, the full-text search box finishes booting before the message database box. When that happens, searches don't work because it didn't see a live message box during bootup. It'll be fixed shortly.

To: Green Receipt who wrote (18910)	6/26/2003 2:10:28 PM
From: SI Bob	Read Replies (3) \| Respond to of 32871

A little more in-depth explanation of how message search (currently) works. Messages are stored in CLOB fields on the main database server. A Perl script running on a separate machine (the one Laurin's going to try to take to the ISP today) parses the words in messages and stores them on yet another box. A Windoze machine running MySQL. When you submit a search, it goes to the MySQL box to get the message numbers then goes to the main database server to retrieve those messages. It's looking more likely that when I'm done, the search will end up working in a similar way. On iHub, we use the full-text indexing feature of W2K/SQLServer2K, with searching handled by the main (only) database server. That's fine on a site with about a million messages. I don't think it will be on a site with nearly 20 times as many. The cache would be in a nearly perpetual polluted state doing it that way.