Greg Twyford wrote:
> Ian Cheong wrote:
>> http://www.clusterworld.com/value_cluster.shtml
>>
> 
> Ian,
> 
> I think you've missed Summernats.
> 
> Greg

I agree this is overkill for even the most CPU-intensive data mining of
 data from a single practice, but might not be overkill for analysis of
collated data from hundreds or thousands of practices collected over a
long period. All we need are some widely used data standards...

Actually, you also need to use analytical techniques which are readily
parallelisable, and only some are. However, a Web-based open source
exploratory data analysis engine which we have been working on is
potentially very parallelisable - preliminary work on a compute cluster
at ANU several years ago yielded promising results. However, even on a
single CPU, it works in real-time with data sets of up to 10 or 15
million rows.

Tim C


_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk

Reply via email to