WhenAny

There are times when firing off the same query in parallel to two or more data stores is a good choice, but normally hitting one, waiting for the response, and only hitting the second if necessary is a better approach. Here I’ll show how to query Redis, and if the key is missing, query MS SQL.

There are times when firing off the same query in parallel to two or more data stores is a good choice, keeping the first response, and canceling the other requests. Here I show how a achieve that even if the Task<> types are different.

After some discussions on Twitter a few new ways of using WhenAny with Tasks of different types came up. Here is what I think will be my last post on this topic for a while, probably…

If you use WhenAny with different types of Task you may want to do something when the first finishes, but also wait for the second to finish. Here’s how.

Sometimes you want to start multiple tasks in parallel, and you only care about the first one that finishes. WhenAny is great in that scenario…except when the tasks are of different types. But with a bit of pattern matching, you can get it working.