As I understand it, Goolge Spanner is built using atomic clocks to synch data updates around the cluster, F1 is built on Google Spanner. This alone is going to make take up of the technology very restricted, unless Google offer access to F1 at a reasonable price.
Posts by acobley
4 publicly visible posts • joined 2 Nov 2012
Google goes back to the future with SQL F1 database
Big Data versus small data: Unpicking the paradox
Wednesday 6th February 2013 16:45 GMT
Re: Eh?
A string is none atomic because you might like to search inside it. So if you have a table, one column stores 140 character tweets, you might like to search for all tweets that contain mention of "the Register" and get all the rows that contain that condition. In that case the string we are interested in, can appear anywhere in the field we are looking at and the position that it appears is unimportant. For sentiment analysis we may be interested in the position of the string, so we might want to write a query that returns all the tweets with "the register" at the end of the tweet, because these tend to be more positive. Again we are looking inside the field and are interested in the properties that the text has.
Numerical values will normally be compared against external values (count >10) or against a value in another field. You would very rarely in the real world (and I struggle to find an example) want to return all rows where the 10's value is greater than the units value, I.E compare within the decimal number. Nor would we typically look for numbers that contain a value anywhere in the number. So would we ever write a query that returns all the stores that held a stock of oranges where the number of oranges had 12 in it ? I.e 12 oranges, 120 oranges, 312 etc ?
Biting the hand that feeds IT
