Re: [wsbpel] Can Serializable Scope protect concurrent access within a<flow> ?

[Ron Ten-Hove <Ronald.Ten-Hove@Sun.COM>]

Ricky,

    I believe it is best to look at serializable scopes as modelled on 
the database concept of serialized isolation, rather than as a 
serialized access mechanism for a multi-threaded program. I believe the 
spec makes an explicit reference to this, although it does not elaborate 
on this. Do we need to expand on this section?

-Ron

Ricky Ho wrote:

> I believe the intention of "serializable scope" is to provide some 
> mechanism to protect shared data under concurrent access.  This 
> mechanism relies on a clear demarcation point for the "critical 
> section" (typically there won't be new thread of execution created 
> within the scope).  I guess the intention is for protection concurrent 
> execution of message handlers.
>
> However, it is not very clear to me if the same mechanism can be used 
> within a <flow> to protect shared data under concurrent access.  (In a 
> DAG model, the start and end of a new thread of execution is not 
> easily identifiable and so specifying the scope is hard).
>
> Therefore, is there any other mechanism to protect shared data within 
> a <flow> against concurrent access ?
>
> Rgds, Ricky
>
>
> To unsubscribe from this mailing list (and be removed from the roster 
> of the OASIS TC), go to 
> http://www.oasis-open.org/apps/org/workgroup/wsbpel/members/leave_workgroup.php. 
>
>