Write-Behind Events in AWS ScaleOut StateServer

[thejaka]

Does AWS ScaleOut StateServer suports Write-Behind/Erase-Behind Events?
If not, is this in your roadmap?
Is there any other alternative way we could achieve the same functionality?

 

[markw]

Yes, Write-Behind/Erase-Behind is currently supported on AWS--the core of this functionality does not depend on a specific on-prem/cloud environment. If you share what kind of backing store you intend to write to (and which ScaleOut client library you intend to use) then we may be able to offer some specific tips and implementation guidance.

 

[thejaka]

Thank you for quick reply.
Backing store initially would be SQL Server 2017 (standalone) version and later it will be changed to RDS latest stable version. ScaleOut client library going to be used is version 5.8.9.
Do you think the above combinations would work well with SOSS AWS latest stable version? If not, what is your recommendation.

Additionally, at the initial stage all web/application servers will be with fix IPs and later it will be changed to use auto scaling. Would SOSS AWS latest stable version work well with both the cases as well?
Would there be any complications with higher loads in a high available setup?

 

[markw]

Yes, SQL Server will work fine: data access to SQL Server is handled entirely in the Write-Behind event callback that is defined in your code, so the ScaleOut service is not directly involved in database communication--it's handled entirely by your application, allowing you to use whatever database you need.

More information can be found in these links:

• .NET Core APIs
• Traditional .NET Framework APIs
• Java

The front-end clients that access the ScaleOut service can be configured to auto scale, but the ScaleOut service itself does not yet support auto scaling. It's on our roadmap (some preliminary work has been done), but I don't have a date for it yet. If you'd like to discuss your timeline and requirements offline then go ahead and contact us through this form and we can set up a time.