Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
RE: [cosmos-dev] Performance positioning of the COSMOS M2 vis-a-visP2Pintegrations

Jimmy,
 
We always expected that a consumer who needs to exchange large volumes of data would use a direct connection between the source and the consumer.  That doesn't mean that they need to reinvent all of the wheels that are required to discover, configure, query, and manage each other.  It just means that they would still need to open a private connection for the purpose of exchanging bulk data after they have discovered each other and negotiated their protocols and other details.  Likewise, a web service query through a COSMOS data manager that is located close to the management database may perform much better than a direct remote query directly to that database by reducing the volume of transmitted data.
 
I don't think that the issue is additional layers, it would be additional layers that add overhead without adding value.
 
Regards,
 
Don


The contents of this e-mail are intended for the named addressee only. It contains information that may be confidential. Unless you are the named addressee or an authorized designee, you may not copy or use it, or disclose it to anyone else. If you received it in error please notify us immediately and then destroy it.
From: cosmos-dev-bounces@xxxxxxxxxxx [mailto:cosmos-dev-bounces@xxxxxxxxxxx] On Behalf Of Mohsin, Jimmy
Sent: Monday, November 19, 2007 12:19 AM
To: Cosmos Dev
Cc: Craddock, Chris; Hayes, Ian
Subject: RE: [cosmos-dev] Performance positioning of the COSMOS M2 vis-a-visP2Pintegrations

Sorry for a type below… replace “entails” with “entail”…

 

Thanks,

Jimmy Mohsin

Cell   +1-609-635-1703

 

From: cosmos-dev-bounces@xxxxxxxxxxx [mailto:cosmos-dev-bounces@xxxxxxxxxxx] On Behalf Of Mohsin, Jimmy
Sent: Monday, November 19, 2007 12:09 AM
To: cosmos-dev@xxxxxxxxxxx
Cc: Craddock, Chris; Hayes, Ian
Subject: [cosmos-dev] Performance positioning of the COSMOS M2 vis-a-vis P2Pintegrations

 

All,

 

On more than one occasion, some of the reviewers of the COSMOS presentations have put the following questions on the table; and this usually happens when we describe the data manager….

 

“But you just added TWO layers to my data-integration problem that I solved using point-to-point integration… Specifically, you are now asking me to add Data Managers to BOTH of the products I just integrated via a P2P approach?”

 

Obviously, this view has its flaws, i.e. it fails to recognize that COSMOS offers a standards-based data integration building blocks / tools / approaches that could one day become the lingua franca for such integrations in this business space…  Eventually, COSMOS could alleviate the need for MANY (but definitely NOT ALL) point-to-point integrations.

 

Thus, there will ALWAYS be need for SOME point to point integrations, where blindingly fast and massive amounts of data exchange are key criteria.

 

Now the question: do we need to start to put some thresholds / guidelines in place that will enable adopters to determine if COSMOS is a viable solution for them?  These thresholds or guidelines or whatever could be used by a potential adopter to determine if they should use a COSMOS-based approach or go for a P2P integration.  Or do we say that, “COSMOS will impose minimal overhead due to its Data Managers and associated infrastructure, and should be considered for ALL data centric integrations; and P2P integrations must only be attempted as a last resort?”

 

Or in other words, how brave should we be J in positioning our product?  Interestingly, this question comes up VERY frequently from data centric audiences, who are our primary adopter target…  Seems the data folks are very much data performance focused, however, there are MANY use cases that do not entails large amounts of data transfer between products.

 

Thoughts / Feedback / Comments?,

Jimmy Mohsin

Cell   +1-609-635-1703

 



Back to the top