RE: XML versus SOAP/WSDL Performance

["Durham, David" <david.durham@intel.com>]

Yes, I agree stats and status require some further analysis... Though it is
interesting to note that operators at the iab nm workshop mentioned they
regularly get a stats dump using CLI and then screen scrape the output to
find what they need. Compared to this, even SOAP over XML over HTTP etc.
would be a dramatic improvement.

The main thing to avoid is a 3-way TCP handshake for each and every
consecutive query, latency is worse than a little extra header garbage on
the wire... But this is really just an implementation issue. 

-Dave

> -----Original Message-----
> From: Remco van de Meent [mailto:remco@vandemeent.net]
> Sent: Friday, September 20, 2002 10:33 AM
> To: xmlconf@ops.ietf.org
> Subject: Re: XML versus SOAP/WSDL Performance
> 
> 
> Durham, David wrote:
> > I don't think performance is the issue for configuration management.
> > Where as XML is already being used for voluminous ecommerce
> > transactions, I don't think the comparatively minuscule amount of
> > device configuration data should raise anyone's eyebrows.
> > Architecturally, it seems that the choice of sending either (for
> > example):
> > 
> >     A. The entire XML configuration for every minor config 
> update vs.
> >     B. Delta updates to Named Instances of XML configuration data
> > 
> > Is, IMHO, a much better discussion to have rather how long it takes
> > to parse a bunch of tagged data.
> 
> It's a separate discussion, I think, although very interesting. ;-)
> 
> Performance *is* an issue for frequent operations such as monitoring,
> and as people still seem to think that monitoring and configuration
> management operations should be combined in one single management
> protocol, this list seems quite suited to discuss monitoring using XML
> as well, I'd say.
> 
> So the question is whether the extra features of SOAP over HTTP over
> TCP over IP make the protocol stack so bloated that the extra delay
> incurred by SOAP versus 'an XML thing over something, specific to
> management operations' in unacceptable for frequent operations.
> 
> Some simple  measurements I did this afternoon led me to the very
> preliminary conclusion that, say, a simple Get operation using SOAP
> (over ..) is not that much slower than the good old snmpget. 
> Didn't try
> complicated things (yet). Neither does it address the 
> original question
> of 'SOAP over ..' vs 'just XML'.
> 
> 
> regards,
> Remco.
> 
> --
> to unsubscribe send a message to xmlconf-request@ops.ietf.org with
> the word 'unsubscribe' in a single line as the message text body.
> archive: <http://ops.ietf.org/lists/xmlconf/>
> 

--
to unsubscribe send a message to xmlconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/xmlconf/>