Regards, Jerry Schwartz The Infoshop by Global Information Incorporated 195 Farmington Ave. Farmington, CT 06032
860.674.8796 / FAX: 860.674.8341 www.the-infoshop.com >-----Original Message----- >From: Scott Haneda [mailto:talkli...@newgeo.com] >Sent: Friday, October 16, 2009 2:44 PM >To: Jerry Schwartz >Cc: Ray; <mysql@lists.mysql.com> >Subject: Re: insert random number into table > >I always maintain a timestamp in my random numbers. As long as my >precision is higher than my requests per second, wouldn't I be safe >from collisions? Assuming a time machine is not invented. > [JS] "As long as..." is very dangerous. Even if your mean transaction arrival rate is okay, you can't rule out the possibility of two transactions arriving at the same time (within the precision of your clock). Admittedly, we're getting theoretical here; but I always like to come at things from the theoretical side, since it is so easy to overlook your assumptions if you come at it from the opposite direction. >-- >Scott >Iphone says hello. > >On Oct 16, 2009, at 11:29 AM, "Jerry Schwartz" <jschwa...@the-infoshop.com > > wrote: > >> JS] Just remember that the pseudo-random numbers that come out of a >> computer, >> regardless of where or how they are generated, are not guaranteed to >> be >> unique. Although the odds of a collision are slim, my philosophy has >> always >> been "if it's possible, it will happen - in the middle of your >> vacation". -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/mysql?unsub=arch...@jab.org
