EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  <20042005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  <20042005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: RE: ? Gateway
From: "Liyu, Andrei" <[email protected]>
To: Jeff Hill <[email protected]>, [email protected]
Date: Thu, 09 Sep 2004 14:46:31 -0400
Jeff,

	Why will IOC not crash? IOC doesn't free memory and Real-time
IOC should not do swapping. So IOC will come to memory limit (not only
low memory block). Then ... I don't know. Maybe IOC will work but
anybody couldn't connect to it. But I think vxWorks
allocates/deallocates memory constantly. So it will stop.

	About Gateway. 
	I know idea of Gateway. I am interesting if client will not
deallocate resource that
	- Gateway takes care itself and IOC.
OR
	- Gateway will not take care itself and IOC.
There are other solutions Gateway takes care itself or IOC but it is
less real.
	
	Why am I interesting?
SNS accelerator has started and I don't like to stop IOC. If I will
connect/stop to 100 PVs and 100 times per day that I could come to IOC
memory limit.

Andrei.

-----Original Message-----
From: Jeff Hill [mailto:[email protected]] 
Sent: Thursday, September 09, 2004 2:03 PM
To: Liyu, Andrei; [email protected]
Subject: RE: ? Gateway


> 	If client connects to IOC then IOC allocates some resource for
> the client. Client should free resource. If a lot of clients wouldn't
> free that IOC should crash.

The IOC might run low on memory blocks of sufficient size, but it should
not
crash.

> 
> 	What will we have with Gateway? 

The GW's purpose is to transform the load of N external clients into an
effective inside IOC load similar to what is imposed by one client. This
is
much better for an overloaded IOC, but of course the load is transferred
to
the GW. A distributed system is converted to a centralized system. There
is
a single point of failure and activity levels can be high in the GW as
it is
a central bottleneck between M IOCs and N clients. This effect is
mitigated
somewhat by the GW's ability to run on cost effective SMP virtual memory
systems (high performance VME processors with enormous memories tend to
be a
bit pricey compared to Linux and windows boxes). You can also spread
load
across as many gateways as you have the time and inclination to install.

> Can I debug client and don't think about Gateway and IOC servers?

Almost, but note that depending on configuration the GW will forward all
write requests (and any associated load) to the IOC. Note also that if
you
don't want your tests to interfere with a GW essential to your
operations
then you could start up a private GW that is only used for testing
purposes.

Jeff




Replies:
RE: ? Gateway Jeff Hill

Navigate by Date:
Prev: RE: ? Gateway Jeff Hill
Next: RE: ? Gateway Jeff Hill
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  <20042005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
Navigate by Thread:
Prev: RE: ? Gateway Jeff Hill
Next: RE: ? Gateway Jeff Hill
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  <20042005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
ANJ, 10 Aug 2010 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·