Hi Trent,
As always we welcome new SF updates and can't wait to see what is new or fixed in this release.

Here are some questions

1. I noticed new assemblies for LZMA compression and don't know which of these if any should be included in the project and distributed to customers.  Here are there names:
• MicroFour StrataFrame Lzma.dll
• MicroFour.StrataFrame.Lzma.x64.Net2.dll
• MicroFour.StrataFrame.Lzma.x64.Net4.dll
• MicroFour.StrataFrame.Lzma.x86.Net2.dll
• MicroFour.StrataFrame.Lzma.x86.Net4.dll
2. On DDT there are 2 Compression options: GZIP and LZMA, I don't recall which one was the default used by SF.  If I build a DDT package how would the DataMigration class would know which compression was used to properly read the DDT package when I release new updates to customers?

Thanks!

Edhy Rijo

Good questions, Edhy.

1. The only file that you will need to distribute with your app is MicroFour StrataFrame Lzma.dll.  This is the C# library that is 100% .NET.  The other assemblies were when we were using C++ for all of the compression and were left only for backup and will be removed from distribution in the future.
2. By default, SF uses the GZIP, but I highly recommend using LZMA now as it is far more powerful in regards to compression.  It is slower compressing, but you will not notice a difference on the decompression, when it really matters anyway.  As for the package file, there is intrinsic logic that can detect and read old packages.  However, I highly recommend updating your DDT packages to start using LZMA.  You will find this setting on the properties of the DDT project.
   

Hi Trent,

Thanks for the prompt response.  LZMA will be my default from now on.

Edhy Rijo

Yeah, it is amazing how much LZMA shrinks it down.  It reduced our distribution  data by 60-70% or so.  Major improvements!

Hi Trent,

Well, I am officially a believer on LZMA, testing new LZMA compression on one of my DDT package and I got 40% better compression than GZIP and it is worth to know that this
package only contain the SF Security deployment data not other lookup data which I would assume the bigger the DDT the better the compression we would get, so thanks again for this improvement.

As far as the rest, so far doing quick test everything seems to work just fine, but without more details on what has been changed/improved no realistic way for me to compare.

Edhy Rijo

I understand, Edhy.  We have been running around here crazy lately so the docs are usually the last thing.  I wanted to push out the build instead of waiting.  But we will publish a more replete list soon.