Damian has been taking advantage of the Parallel capabilities using a number of different technologies. Some enable parallel computing on a massive scale. Damian writes:
Check out links below. Very informative.
http://blog.sqlauthority.com/2011/02/06/sql-server-cxpacket-parallelism-usual-solution-wait-type-day-6-of-28/
http://blog.sqlauthority.com/2010/03/15/sql-server-maxdop-settings-to-limit-query-to-run-on-specific-cpu/
We have also being using the new parallel options within .net 4.5 in our applications to speed up web request calls, loop processing (massive performance gains) and they are really simple and much easier to use that previous versions (threading options).
SQL Database wise we have found reducing parallelism for OLTP operations can help but again like everything it really depends on your work load. I guess use it as a tool where can otherwise not.
Check Microsoft Azure new services which is in preview. Cloud based parallel processing for developers with no hassle about servers ;-). I’m a developer and I can use this without the complicated task of dealing with servers.
http://azure.microsoft.com/en-us/documentation/services/batch/
Amazon AWS have also just launched a similar service http://aws.amazon.com/lambda/ and I’d also recommend checking outhttp://www.iron.io/worker who have offer parallel task processing for a number of years now. They are behind the very successful www.sendgrid.com so you can imagine that sending millions of emails would require a very good parallel tasking system ;-).
Check out links below. Very informative.
http://blog.sqlauthority.com/2011/02/06/sql-server-cxpacket-parallelism-usual-solution-wait-type-day-6-of-28/
http://blog.sqlauthority.com/2010/03/15/sql-server-maxdop-settings-to-limit-query-to-run-on-specific-cpu/
We have also being using the new parallel options within .net 4.5 in our applications to speed up web request calls, loop processing (massive performance gains) and they are really simple and much easier to use that previous versions (threading options).
SQL Database wise we have found reducing parallelism for OLTP operations can help but again like everything it really depends on your work load. I guess use it as a tool where can otherwise not.
Check Microsoft Azure new services which is in preview. Cloud based parallel processing for developers with no hassle about servers ;-). I’m a developer and I can use this without the complicated task of dealing with servers.
http://azure.microsoft.com/en-us/documentation/services/batch/
Amazon AWS have also just launched a similar service http://aws.amazon.com/lambda/ and I’d also recommend checking outhttp://www.iron.io/worker who have offer parallel task processing for a number of years now. They are behind the very successful www.sendgrid.com so you can imagine that sending millions of emails would require a very good parallel tasking system ;-).
Thanks for the update Damian. Feel free to write in your with your contribution of favorite tools or implementations to btaylor@sswug.org, or simply leave your comments here.
Cheers,
Ben