On Tue, 2006-01-17 at 18:40 -0600, Greg Woodhouse wrote:
> --- Kevin Toppenberg <[EMAIL PROTECTED]> wrote:
>
> > Looks like the job manager is allowing each task only one
> instruction
> > cycle, then moves on. Since each job is identical, it seems
> > reasonable that they would show a strict a
On Jan 17, 2006, at 8:55 PM, Chris Richardson wrote:
You guys are making it so complicated. Why not just have each copy
of the
process do this;
S ^RRCTST($H,$J)=$G(^RRCTST($H,$J))+1
Then you can run as many copies of the routine as you wish for as
long as
you wish and in post processing
On Jan 17, 2006, at 4:40 PM, Greg Woodhouse wrote:
--- Kevin Toppenberg <[EMAIL PROTECTED]> wrote:
Looks like the job manager is allowing each task only one instruction
cycle, then moves on. Since each job is identical, it seems
reasonable that they would show a strict alternating pattern.
coarse, make a call to the OS to get the current
millisecond count and replace the $H reference.
- Original Message -
From: "Gary Monger" <[EMAIL PROTECTED]>
To:
Sent: Tuesday, January 17, 2006 8:15 PM
Subject: RE: [Hardhats-members] Interleaving
> For that kind of
: [Hardhats-members] Interleaving
On Jan 17, 2006, at 7:07 PM, Gary Monger wrote:
> Wouldn't the HL7 link monitor be an example of this? The background
> processes write something about their state to a global and some other
> process monitors the global and displays it.
I don&#
On Jan 17, 2006, at 7:07 PM, Gary Monger wrote:
Wouldn't the HL7 link monitor be an example of this? The background
processes write something about their state to a global and some other
process monitors the global and displays it.
I don't think so. The time scales are very different: The HL
ed to lock.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Gregory
Woodhouse
Sent: Tuesday, January 17, 2006 9:50 AM
To: hardhats-members@lists.sourceforge.net
Subject: Re: [Hardhats-members] Interleaving
On Jan 15, 2006, at 7:03 PM, Gregory Woodhouse wrote:
>
--- Kevin Toppenberg <[EMAIL PROTECTED]> wrote:
> Looks like the job manager is allowing each task only one instruction
> cycle, then moves on. Since each job is identical, it seems
> reasonable that they would show a strict alternating pattern.
>
> Kevin
it surprises me. Switching between task
Looks like the job manager is allowing each task only one instruction
cycle, then moves on. Since each job is identical, it seems
reasonable that they would show a strict alternating pattern.
Kevin
On 1/17/06, Greg Woodhouse <[EMAIL PROTECTED]> wrote:
> --- Steven McPhelan <[EMAIL PROTECTED]>
--- Steven McPhelan <[EMAIL PROTECTED]> wrote:
> Mostly likely J1 would
> run to
> completion before J2 even started. With most processes today, I
> doubt 15000
> would be sufficient.
>
Curiouser and curiouser...After 158,108 (on this run) there was a
switch, but then the two jobs proceeded in
--- Steven McPhelan <[EMAIL PROTECTED]> wrote:
> Your increment example would show the interleave by looking at the
> subscripted global. Why not do this as:
>
> S ^GLB=""
>
> J1 ;
> F I=1:1:100 S ^GLB=$G(^GLB)_"A"
> Q
>
>
>
> J2 ;
> F I=1:1:100 S ^GLB=$G(^GLB)_"B"
> Q
>
>
Your increment example would show the interleave by looking at the subscripted global. Why not do this as:
S ^GLB=""
J1 ;
F I=1:1:100 S ^GLB=$G(^GLB)_"A"
Q
J2 ;
F I=1:1:100 S ^GLB=$G(^GLB)_"B"
Q
Wouldn't that show interleaving within standard M without using the expli
Interesting. Try making the jobs each do 1000 or 5000 writes. It may
be that after the command J RUN("A") is completed and before then next
line J RUN("B") is interpreted and compiled, during this time RUN("A")
has time complete. But maybe that was what you ment regarding
"pre-empted".
Kevin
Also, it turns out that (unless I put a HANG in the loop), this code
doesn't demonstrate any interleaving anyway (probably because the
background jobs finish before they are pre-empted:
BAY>ZL ZZGREG1 ZP
EN ; ; 1/17/06 6:04pm
K ^XTMP("GREG")
S ^XTMP("GREG")=0
J RUN
--- Kevin Toppenberg <[EMAIL PROTECTED]> wrote:
> Greg,
>
> I'll be dumb here. I was not aware of the $increment function. I
> just look here:
> http://www.jacquardsystems.com/Examples/
>
> and don't see it in the list of functions. What am I missing?
>
> Also, could you modifiy my code so t
Greg,
I'll be dumb here. I was not aware of the $increment function. I
just look here:
http://www.jacquardsystems.com/Examples/
and don't see it in the list of functions. What am I missing?
Also, could you modifiy my code so that if a lock was encountered, it
would pause briefly and try again
On Jan 15, 2006, at 7:03 PM, Gregory Woodhouse wrote:
On Jan 15, 2006, at 4:28 PM, Kevin Toppenberg wrote:
How about writing to a global:
+L set ^GLOBAL(^GLOBAL(0))=VALUE,^GLOBAL(0)=^GLOBAL(0)+1 -L
Kevin
That won't work (because of the lock). What I'd like to do is, in
some sense, the
On Jan 15, 2006, at 4:28 PM, Kevin Toppenberg wrote:
How about writing to a global:
+L set ^GLOBAL(^GLOBAL(0))=VALUE,^GLOBAL(0)=^GLOBAL(0)+1 -L
Kevin
That won't work (because of the lock). What I'd like to do is, in
some sense, the opposite of what you usually want to do with
concurrent
How about writing to a global:
+L set ^GLOBAL(^GLOBAL(0))=VALUE,^GLOBAL(0)=^GLOBAL(0)+1 -L
Kevin
On 1/15/06, Gregory Woodhouse <[EMAIL PROTECTED]> wrote:
>
>
> In Scheme, I can have multiple threads writing to the same terminal,
> but not in MUMPS. True, I can have multiple terminals open and s
On Jan 15, 2006, at 1:43 PM, Gregory Woodhouse wrote:
In Scheme, I can have multiple threads writing to the same
terminal, but not in MUMPS. True, I can have multiple terminals
open and see the output in each, but what I'd like to see is the
output of one job *interleaved* with the other.
On Jan 14, 2006, at 11:40 PM, Jim Self wrote:
Gregory wrote:
How can you demonstrate the interleaving of jobs without disturbing
it with explicit synchronization?
JOB BODY("A")
JOB BODY("B")
where BODY is just a FOR loop
BODY(X) ;
FOR I=1:1:25 WRITE X
QUIT
But, of course that's not all
Gregory wrote:
>How can you demonstrate the interleaving of jobs without disturbing
>it with explicit synchronization?
>JOB BODY("A")
>JOB BODY("B")
>
>where BODY is just a FOR loop
>
>BODY(X) ;
> FOR I=1:1:25 WRITE X
>QUIT
>
>But, of course that's not allowed, because background jobs can't
>wr
22 matches
Mail list logo
