Experimental Physics and
Industrial Control System

"Lawrence T. Hoff" <[email protected]> · Thu, 15 Feb 2007 12:29:27 -0500

	I hope this topic is not too specific for the general
tech talker....
	If memory serves, RTI was primarily a consulting
company with focus on robotics and VxWorks (hence Dave
Thompson's and Kay Kasemir's experience). RTI developed
a suite of tools for their own purpose, which was later
"productized" for sale by RTI, then repackaged for sale
via WRS.
	My personal experience was using "RTILib", purchased
from RTI. Among other things, it included a set of Heap
Analysis tools - useful for detecting memory leaks and
similar errors.
	This was back in the VxWorks 5.2 days, but their
product still works under VxWorks 5.5 (except for the 24-bit
relocation issue with PowerPCs).

Claude Saunders wrote:
I only looked at a few of the QoS capabilities implemented by ORTE, but
the RTPS specification is an interesting read in this area

[snip]
I think DDS makes many of these concepts more explicit.

Interesting. Thanks again for sharing your insight. It
seems to me that just identifying QoS as a concept provides
some value. Every now and then, I see someone construct a
system that depends on receiving data *in the sequence the
data was published*, else "bad things" (tm) happen. Under
certain circumstances, CA will discard intermediate values
in favor of publishing the latest (most up to date) value.
While these circumstances are rare, they can be surprising.
In theory, QoS could be used to distinguish between channels
where maintaining order is more important than most recent
data and vice versa.
Of course, this is all "pie in the sky". I am disappointed to
hear that DDS is not likely to be the "magic bullet" that
solves these issues.
-- Larry

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System