That's definitely a place where I'd explore other integration options then, Jarl. IMO, Web Services isn't a good bulk transfer mechanism but it's really handy and flexible for transactional stuff. Just the overhead of using an XML parser (SAX or DOM, though DOM is worse) makes it more bloated than other access mechanisms.
Add the Mid-Tier to the mix, and you're just looking for trouble. :) -Chris -----Original Message----- From: Action Request System discussion list(ARSList) [mailto:[EMAIL PROTECTED] Behalf Of Jarl Grøneng Sent: Monday, June 11, 2007 10:07 AM To: [email protected] Subject: Re: Hypothetical A few Kb was just a joke. But with messages 100kb ++ and 10-15000 messages a day the server did malloc quite often... This was on solaris with oracle. -- Jarl On 6/11/07, Grooms, Frederick W <[EMAIL PROTECTED]> wrote: > Jarl, > What platform are you on? I routinely have 60 - 100 Kb XML transactions with no memory errors. (I am on Sun with Oracle) > > Fred > > -----Original Message----- > From: Action Request System discussion list(ARSList) [mailto:[EMAIL PROTECTED] On Behalf Of Jarl Grøneng > Sent: Saturday, June 09, 2007 12:57 PM > To: [email protected] > Subject: Re: Hypothetical > > AR Server as middleware? Huh, it cant handle larger xml than a few Kb. > Storing XML in a database as tables and fields(like its done in AR > System) are not the prefered method when talking about performance. > > We all love the Malloc 300 errormessage when using webservices.... > -- > Jarl > > > On 6/9/07, Chris Woyton <[EMAIL PROTECTED]> wrote: > > Here's an opposing thought worth considering... > > > > Going back to the spirit of ARS being a Rapid Development Platform, > > why would BMC encourage development of the *same thing* that's out > > there already, regardless of who produced it? Many have lost sight of > > ARS as a development medium because it's been perceived as "just a > > Help Desk" for quite some time - and adding 50 more flavors of IT > > Request/Service Management won't do much to fix that perception. > > > > Requiring partners to produce products that are in non-competition is > > certainly part of the goal - money drives everything, as they say. > > However, it may also be construed as pushing the horizontal boundaries > > of the platform - pushing ISV's to take the product and move it into other arenas. > > There's obviously some interest in taking advantage of this facility, > > so instead of ITSM-esque applications, how about Fleet Management, > > Document Management, Middleware (Web Services + ARDBC + Workflow > > Engine is a dynamite combo for this), Financial Applications, etc. > > > > IMHO, those things add value to the platform - another ITSM product doesn't. > > A bigger pie provides revenue to BMC, no doubt, but it also gives the > > ISV a chance at more than crumbs. > > > > -Chris Woyton > > ATS, TuringSMI > > > > ps with regards to Robert's comment on CMDB, another thought comes to > > mind - I've often pondered using the OBJSTR sub-system as a > > development medium all on its own. Imagine this - you build a core set > > of Classes for a particular use, for example, > > Middleware/Data-Transfer. When a new Data Source becomes available, > > specialized a Sub-Class for it. Consumers of the data can then point > > to the specific Sub-Class or the root Parent Class (or at any point in > > the tree) depending on what data they need to use. Or, in a Request > > Management application, rather than providing different "Views" of an > > app to suit different groups, specialize a Sub-Class for that Group > > such that common data is shared, but specific data is segmented. Data > > sets could be used to support Tenancy in a model like this and the Recon Engine could facilitate inter-application integration (as well as exta-application). > > > > Maybe one of you hyper-motivated young guns can play with that idea > > (Reinfeldt already busts my chops for the 30 or so half-written emails > > to him I haven't had time to finish, so no way would I commit to > > prototyping that stuff..hehehe) :) > > > > -----Original Message----- > > From: Action Request System discussion list(ARSList) > > [mailto:[EMAIL PROTECTED] Behalf Of Robert Molenda > > Sent: Thursday, June 07, 2007 9:27 PM > > To: [email protected] > > Subject: Re: Hypothetical > > > > > > Axton - you think too much outside the box :) Just like so many of us > > on this list :) :) We need more of this thinking again!!! > > > > I have actually been wondering about this for some time now, > > especially in the area of CMDB and 'Re-development' or 'Module > > Integration' so to say. > > > > The BMC CMDB while being 'OK' (not to take this completely off topic) > > is such an overhead that a much simpler and "customer fitting design" > > would be so much more performant to the ARSystem and other applications... > > (none the less cheaper and easier to maintain at times!) > > > > At what point will BMC begin to limit customizations? Imagine if the > > install of say Incident Management installed all objects in "Locked > > Mode"... > > > > I wonder at times if BMC forgot the first envisioned cause for ARS... > > Rapid Application Development, Flexible Workflow, ... > > > > Robert > > > > -----Original Message----- > > From: Action Request System discussion list(ARSList) > > [mailto:[EMAIL PROTECTED] On Behalf Of Rick Cook > > Sent: Thursday, June 07, 2007 6:28 PM > > To: [email protected] > > Subject: Re: Hypothetical > > > > I don't know what BMC's criteria are for approval, but I do know that > > there are already competing Service Management products out there, > > what's the point of a few more, unless someone thinks they've > > architected the code better than BMC does? > > > > Rick > > > > -----Original Message----- > > From: Action Request System discussion list(ARSList) > > [mailto:[EMAIL PROTECTED] On Behalf Of Axton > > Sent: Thursday, June 07, 2007 4:24 PM > > To: [email protected] > > Subject: Re: Hypothetical > > > > hmmm... probably if you write it first and big brother likes it, > > you're SOL. > > Prepare to be bought or dropped (aka, prepare to be boarded)? I guess > > there's money to be made there, but geez, what a disappointment... > > > > Axton Grams > > > > On 6/7/07, patrick zandi <[EMAIL PROTECTED]> wrote: > > > ** > > > Woo, So first inventor win's ? as long as you pay and have it locked. > > > > > huh .. > > > Land Grab.. > > > > > > > > > On 6/7/07, Axton <[EMAIL PROTECTED]> wrote: > > > > > > > > Just a hypothetical question. > > > > > > > > Deployable applications, which include the ability to enforce user > > > > fixed/floating licenses, are available to partners/ISVs. > > > > > > > > Partners are not allowed to write competing products. > > > > > > > > Does this mean that companies/people attempting to write apps that > > > > are similar in nature to those that Remedy offers are in a catch22 > > > > situation? > > > > > > > > Axton Grams > > > > > > > > > > > > > > -- > > > Patrick Zandi > > ____________________________________________________________________________ ___ > UNSUBSCRIBE or access ARSlist Archives at www.arslist.org ARSlist:"Where the Answers Are" > ____________________________________________________________________________ ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org ARSlist:"Where the Answers Are" _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org ARSlist:"Where the Answers Are"
