Centrally Controlled P2P Connectivity

[Former Member]

Hi All,

Came across this line in one of the XI overview docs: "The architecture allows Peer-to-peer connectivity (still centrally controlled) to avoid additional point of failures". Can anyone explain how this can be done?

Thanks in advance.

Regards,

Bharath

[Former Member · Accepted solution]

Hi Bharath,

>>"The architecture allows Peer-to-peer connectivity

>>>(still centrally controlled) to avoid additional >>point of failures".

it means that now multiple applications/systems can communicate with each other through a central hub that provides all the services required by each of the systems connecting to it. By having a central hub, you can now limit the failure points from inidivual connections between systems to the central hub(XI) that manages and controls all the connection to the peer systems.

CHeers,

Naveen

[Former Member · Accepted solution]

From the statement that I quoted, I thought for a moment that XI supported central monitoring of even point-to-point (old world) interfaces (i.e. systems connected directly to each other w/o any middleware in between). That's why I raised the question. But your reply made me realise that I just misinterpreted the statement.

Anyway, thanks Naveen for responding

[Former Member · Accepted solution]

> From the statement that I quoted, I thought for a

> moment that XI supported central monitoring of even

> point-to-point (old world) interfaces (i.e. systems

> connected directly to each other w/o any middleware

> in between).

That's also the case with XI, it manages all interfaces and end-points from a central point (i.e. RWB, INtegration directory, Monitor tool, adapter framework, etc..). All interfaces whether or not they are p-t-p or service oriented, they are maintained and managed by XI. Some XI adapters like the IDOC adapter and ABAP proxies support end-to-end monitoring (at technical and application level), ofcourse both application interacting in such an interface must understand acknowledgemnts etc..

Cheers,

Roberto

[Former Member · Accepted solution]

Roberto: I dont think you can monitor existing endto end interfaces that have not been developed/integrated with XI. The question posted was can we do end to end monitoring of existing interfaces that are configured directly to the end systems with out XI in middle.

Bharat:

>>I thought for a moment that XI supported central >>monitoring of even point-to-point (old world) >>interfaces (i.e. systems connected directly to each >>other w/o any middleware in between).

IN Xi you cannot do an end to end monitoring, of those interfaces taht dont exist in XI. Hope i have answered you question!!! if so please mark the post as solved.

cheers,

Naveen

[Former Member · Accepted solution]

> Roberto: I dont think you can monitor existing endto

> end interfaces that have not been

> developed/integrated with XI.

> The question posted was

> can we do end to end monitoring of existing

> interfaces that are configured directly to the end

> systems with out XI in middle.

>

> cheers,

> Naveen

Naveen, the initial question was:

> Hi All,

> Came across this line in one of the XI overview docs: >"The architecture allows Peer-to-peer connectivity (still > centrally controlled) to avoid additional point of >failures". Can anyone explain how this can be done?

> Thanks in advance.

> Regards,

> Bharath

My previous answer was based on that initial question and with the assumption that the interfaces are/were developed and maintained in XI. As you have stated here, non-XI interfaces cannot be monitored with XI which is logical.