Manuel,

It's an old problem I discovered one year ago. You can partially solve
it having more than one CPU, but when you have N simultaneous calls
where N is the same number of CPUs you have, the problem is the same.
A workaround is to place calls like Sleep() or PeekMessage() inside your
function code.
A better way to solve the problem is to use an async call to the
function code, leaving the Remoting interface only the role of listener.
Actually, for what I know, there is no automatic way to dispatch in
different threads long remoting calls.

Marco

-----Original Message-----
From: Moderated discussion of advanced .NET topics.
[mailto:[EMAIL PROTECTED]] On Behalf Of Manuel Costa
Sent: giovedì 30 maggio 2002 9.03
To: [EMAIL PROTECTED]
Subject: Re: [ADVANCED-DOTNET] singleton server problem


Hi Mike,

yes, my server side is doing a very intensive work with 100% of cpu
usage and no I/O. Is there any thing i can do to overcome this issue
without changing much my code structure? Does this mean that every time
your machine hosting the server is overloaded your application will no
longer works? If this is true, i'm basically screwed, given that i want
to maximize the cpu usage at the server side.

Manuel



----- Original Message -----
From: "Mike Woodring" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, May 29, 2002 10:02 PM
Subject: Re: [ADVANCED-DOTNET] singleton server problem


> Do your server-side methods do long-running cpu intensive operations?

> Or
do
> they do things that naturally include some blocking (like performing 
> I/O, grabbing locks, etc)?  I ask because the remoting infrastructure 
> uses I/O completion ports to manage the thread pool in such a way that
long-running,
> cpu intensive operations performed by server-side methods can cause 
> this behavior (especially if a server-side method gets into an 
> infinite loop
that
> never blocks).  But if you're threads are not doing long-running, cpu 
> intensive operations, than I don't want to go into the details if you 
> have
a
> different problem altogether :-)
>
> -Mike
> http://staff.develop.com/woodring http://www.develop.com/devresources
>
>
> ----- Original Message -----
> From: "Manuel Costa" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Wednesday, May 29, 2002 10:36 AM
> Subject: [ADVANCED-DOTNET] singleton server problem
>
>
> > Hi,
> >
> > Is there a limit for the number of simultaneous calls to a remote
object
> > registered as singleton?
> > I have an application with several servers and several clients per
server.
> > After some amount of simultaneous calls to the servers they stop
> responding,
> > blocking any client call. When a client invokes a method it will
stay
> > blocked and if i try to debug at the server side by putting a
breakpoint
> in
> > the called method, nothing will happen. Is there any reasonable
> explanation
> > for this?
> >
> > Manuel
> >
> > You can read messages from the Advanced DOTNET archive, unsubscribe
from
> Advanced DOTNET, or
> > subscribe to other DevelopMentor lists at
http://discuss.develop.com.
> >
>
> You can read messages from the Advanced DOTNET archive, unsubscribe
from
Advanced DOTNET, or
> subscribe to other DevelopMentor lists at http://discuss.develop.com.

You can read messages from the Advanced DOTNET archive, unsubscribe from
Advanced DOTNET, or
subscribe to other DevelopMentor lists at http://discuss.develop.com.

You can read messages from the Advanced DOTNET archive, unsubscribe from Advanced 
DOTNET, or
subscribe to other DevelopMentor lists at http://discuss.develop.com.

Reply via email to