Re: [ogsa-bes-wg] Management Functions in BES
+2 At 09:39 AM 7/24/2005 -0400, Tom Maguire wrote:
+1 Karl
Freyâs Law: âEvery 5 years the number of architecture components double and the ability to comprehend them halvesâ
Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.  Antoine de Saint-Exupery
T o m  M a g u i r e
STSM, On Demand Architecture
Poughkeepsie, NYÂ 12601
Karl Czajkowski <karlcz@univa.com > To Sent by: Steven Newhouse owner-ogsa-bes-wg <s.newhouse@omii.ac.uk> @ggf.org cc ogsa-bes-wg@ggf.org Subject 07/24/2005 01:13 Re: [ogsa-bes-wg] Management AM Functions in BES
On Jul 23, Steven Newhouse modulated:
Discussion: Sections 3.4, 3.5 & 3.6. Having 3 management operations in the BES specification until generic OGSA management structure defined?
The operations in 3.4 & 3.5 (Start & Stop AcceptingNewActivities) I see as being a service specific management function (i.e. they effectively alter the behavior of one specific operation within the service).
While its semantics references a BES operation, it seems to me that it is a more general management pattern for any service that has a stateful, factory-like interface: stop accepting/creating new work items but continue processing the existing ones (whatever they are). The BES-specific meaning could just as easily be specified by calling out this work acceptance model so that it can be related to an external management mechanism.
I think it is wrong to burden BES implementations with these kinds of management functions that are specific to a hosting environment. I think the right path here is to engage in management standards for web services and create profiles to compose these with a basic execution service if you require both.
I cannot see these functions as being more important interop points than any of the other security and policy management functions which everyone seems to agree are out of scope. If you must use localized mechanisms to do practical management of a BES installation, I am not sure what these functions buy you.
karl
-- Karl Czajkowski karlcz@univa.com
_______________________________________________________________ Ian Foster www.mcs.anl.gov/~foster Math & Computer Science Div. Dept of Computer Science Argonne National Laboratory The University of Chicago Argonne, IL 60439, U.S.A. Chicago, IL 60637, U.S.A. Tel: 630 252 4619 Fax: 630 252 1997 Globus Alliance, www.globus.org
+3 Ian Foster wrote:
+2
At 09:39 AM 7/24/2005 -0400, Tom Maguire wrote:
+1 Karl
FreyâEUR^(TM)s Law: âEURoeEvery 5 years the number of architecture components double and the ability to comprehend them halvesâEUR?
Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.  Antoine de Saint-Exupery
T o m  M a g u i r e
STSM, On Demand Architecture
Poughkeepsie, NYÂ 12601
Karl Czajkowski
<karlcz@univa.com
To
Sent by: Steven Newhouse owner-ogsa-bes-wg <s.newhouse@omii.ac.uk>
@ggf.org cc
ogsa-bes-wg@ggf.org
Subject 07/24/2005 01:13 Re: [ogsa-bes-wg] Management AM Functions in BES
On Jul 23, Steven Newhouse modulated:
Discussion: Sections 3.4, 3.5 & 3.6. Having 3 management operations in the BES specification until generic OGSA management structure defined?
The operations in 3.4 & 3.5 (Start & Stop AcceptingNewActivities) I see as being a service specific management function (i.e. they effectively alter the behavior of one specific operation within the service).
While its semantics references a BES operation, it seems to me that it is a more general management pattern for any service that has a stateful, factory-like interface: stop accepting/creating new work items but continue processing the existing ones (whatever they are). The BES-specific meaning could just as easily be specified by calling out this work acceptance model so that it can be related to an external management mechanism.
I think it is wrong to burden BES implementations with these kinds of management functions that are specific to a hosting environment. I think the right path here is to engage in management standards for web services and create profiles to compose these with a basic execution service if you require both.
I cannot see these functions as being more important interop points than any of the other security and policy management functions which everyone seems to agree are out of scope. If you must use localized mechanisms to do practical management of a BES installation, I am not sure what these functions buy you.
karl
-- Karl Czajkowski karlcz@univa.com
_______________________________________________________________ Ian Foster www.mcs.anl.gov/~foster <http://www.mcs.anl.gov/%7Efoster> Math & Computer Science Div. Dept of Computer Science Argonne National Laboratory The University of Chicago Argonne, IL 60439, U.S.A. Chicago, IL 60637, U.S.A. Tel: 630 252 4619 Fax: 630 252 1997 Globus Alliance, www.globus.org <http://www.globus.org/>
+4 to Karl. Thanks for pointing this out! Michel On 24 Jul 2005, at 18:38, Andrew Stephen McGough wrote:
+3
Ian Foster wrote:
+2
At 09:39 AM 7/24/2005 -0400, Tom Maguire wrote:
+1 Karl
Frey’s Law: “Every 5 years the number of architecture components double and the ability to comprehend them halvesâ€
Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.  Antoine de Saint-Exupery
T o m  M a g u i r e
STSM, On Demand Architecture
Poughkeepsie, NYÂ 12601
Karl Czajkowski <karlcz@univa.com
To
Sent by: Steven Newhouse owner-ogsa-bes-wg <s.newhouse@omii.ac.uk>
@ggf.org cc ogsa-bes-wg@ggf.org S ubject 07/24/2005 01:13 Re: [ogsa-bes-wg] Management AM Functions in BES
On Jul 23, Steven Newhouse modulated:
Discussion: Sections 3.4, 3.5 & 3.6. Having 3 management operations in the BES specification until generic OGSA management structure defined?
The operations in 3.4 & 3.5 (Start & Stop AcceptingNewActivities) I see as being a service specific management function (i.e. they effectively alter the behavior of one specific operation within the service).
While its semantics references a BES operation, it seems to me that it is a more general management pattern for any service that has a stateful, factory-like interface: stop accepting/creating new work items but continue processing the existing ones (whatever they are). The BES-specific meaning could just as easily be specified by calling out this work acceptance model so that it can be related to an external management mechanism.
I think it is wrong to burden BES implementations with these kinds of management functions that are specific to a hosting environment. I think the right path here is to engage in management standards for web services and create profiles to compose these with a basic execution service if you require both.
I cannot see these functions as being more important interop points than any of the other security and policy management functions which everyone seems to agree are out of scope. If you must use localized mechanisms to do practical management of a BES installation, I am not sure what these functions buy you.
karl
-- Karl Czajkowski karlcz@univa.com
Ian Foster www.mcs.anl.gov/~foster Math & Computer Science Div. Dept of Computer Science Argonne National Laboratory The University of Chicago Argonne, IL 60439, U.S.A. Chicago, IL 60637, U.S.A. Tel: 630 252 4619 Fax: 630 252 1997 Globus Alliance, www.globus.org
participants (3)
-
Andrew Stephen McGough -
Ian Foster -
Michel Drescher