Experimental Physics and
Industrial Control System

"Jeff Hill" <[email protected]> · Mon, 18 Apr 2011 16:28:55 -0600

> > or how
> > EPICS would deal with arrays approaching 2Gb.
>
> I think the only issue would be that there has to be enough RAM in the
> IOC for the multiple copies of the data that are made between the device, 
> record and CA server.
>

Currently another issue, on 64 bit hosts, will be no arrays larger than
roughly 4294967279 Bytes - which is actually pretty close to 2GB. 

Perhaps I shouldn?t be surprised that it didn?t take that long to get from
16kB to 4GB!

Jeff
______________________________________________________
Jeffrey O. Hill           Email        [email protected]
LANL MS H820              Voice        505 665 1831
Los Alamos NM 87545 USA   FAX          505 665 5107

Message content: TSPA

With sufficient thrust, pigs fly just fine. However, this is
not necessarily a good idea. It is hard to be sure where they
are going to land, and it could be dangerous sitting under them
as they fly overhead. -- RFC 1925

> -----Original Message-----
> From: [email protected] [mailto:tech-talk-
> [email protected]] On Behalf Of Andrew Johnson
> Sent: Wednesday, April 13, 2011 9:11 AM
> To: [email protected]
> Subject: Re: Python EPICS interface (pyepics 3.1.1) double waveform
> record (NLEM = 65535 ) record handling exception
> 
> 
> On Wednesday 13 April 2011 09:23:43 Matt Newville wrote:
> > > I am also interested to know how far we can go with the array size in
> > > EPICS and PyEPICS interface.
> >
> > There is no explicit limit on array size that doesn't exist in EPICS.
> > With Area Detector, it's not unusual for us to be transferring 8Mb
> > arrays.   I don't know what the largest verified array size is, or how
> > EPICS would deal with arrays approaching 2Gb.
> 
> I think the only issue would be that there has to be enough RAM in the
> IOC for
> the multiple copies of the data that are made between the device, record
> and
> CA server.  The record obviously has to have the storage for its copy,
> and the
> CA server reserves EPICS_CA_MAX_ARRAY_BYTES + a little bit for buffering
> on
> the way out.  Whatever device support you're using might want its own
> copy as
> well, but that might be avoidable depending on the actual code in use.
> 
> - Andrew
> --
> An error is only a mistake if you don't learn from it.
> When you learn something from it, it becomes a lesson.

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System