UCS-2 and SQL Server

Question

While researching options for storing mostly-English-but-sometimes-not data in a SQL Server database that can potentially be quite large, I\'m leaning toward storing most st

Answer 1

What do you mean by "swimming up the UCS-2 stream"?

Here are your options:

• Use the new 2012 _SC collations (https://msdn.microsoft.com/en-us/library/ms143726.aspx). This idea comes from srutzky. You should check out his answer. This is by far the best solution.

Not recommended but possible:

• Implement a UDT. This will be a lot of work and you will loose tolling support (OR mapping and certainly some SQL Server features that work on native types).
• Use varbinary(max): Requires you to do custom conversion code. No range indexing.
• Use nvarchar(N) and turn on row compression. Starting with SQL Server 2008 R2 this will use an encoding that is as compact as UTF-8. But this requires enterprise edition.

See the comments to read about the severe drawbacks that these approaches have.