Consider adding options to tune storage binding performance

https://docs.microsoft.com/en-us/azure/storage/common/storage-performance-checklist
says

> For more information, see our blog post [Nagle's Algorithm is Not Friendly](http://blogs.msdn.com/b/windowsazurestorage/archive/2010/06/25/nagle-s-algorithm-is-not-friendly-towards-small-requests.aspx) towards Small Requests, which explains why Nagle's algorithm interacts poorly with table and queue requests, and shows how to disable it in your client application.

The article shows upto 90% TCP throughput improvements. Although this article is old, I couldnt find newer recommendations aroudn this. Also didnt see anything related to it in our repos, so thought it might be interesting to investigate.

The setting can be applied in code or in [config ](https://social.msdn.microsoft.com/Forums/en-US/d84ba34b-b0e0-4961-a167-bbe7618beb83/net-and-adonet-data-service-performance-tips-for-windows-azure-tables?forum=windowsazuredata) file. `

`

Ideally, having the ability to configure on a per binding level would be nice. There are other suggestions in the second article for `expect100Continue` and `MergeOption.NoTracking`

Start by reviewing the linked Azure Storage performance guidance, then locate the storage binding configuration entry points; the issue names no repository files or tests. Done means determining the scope of per-binding tuning for Nagle's algorithm and the other listed settings, with the resulting support or design decision recorded.

Written by the indexing model from the issue text.