[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [saml-dev] acquiring | dereferencing an assertion using an artifact
william wrote: > a. approach #1 assumes some kind of persistence mechanism > (in-memory cache, file-system serialization, rdbms, etc...); > please, can anybody share the pros and cons of this approach and > the different possible persistence mechanisms? They both require the same degree of persistence because artifacts have to be bound to a particular SP, so you have to store that information at a minimum. Neither is particularly practical in a clustered environment that wants to avoid a single point of failure, particularly if your cluster is separated widely. That's why 2.0 uses indexed endpoints and ensures each artifact can be bound to a specific one. There are in-memory approaches such as JGroups that seem among the most practical way to deal with it for now. -- Scott
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]