8:10 p.m.
-------- Original Message --------
Subject: ISPN-2808 - thread pool for incoming message [feedback]
Date: Wed, 27 Feb 2013 19:06:49 +0000
From: Pedro Ruivo <pedro(a)infinispan.org>
To: infinispan-core-dev(a)infinispan.org <infinispan-core-dev(a)infinispan.org>
Hi all,
I'm working on ISPN-2808 and I want some feedback about it (code is here
[1])
I'm starting to implement this feature but I know that Asynchronous
Invocation API is not totally finished in JGroups.
My idea in to use an executor service in CommandAwareRpcDispatcher
(CARD) and when a request (command) is received, it checks if it is
useful to move the command execution to another thread (in this line [2])
For now, I'm thinking to move all the write commands, lock control
command, prepare command and commit command to the executor service
(Note: commit command is only moved when in DIST mode and L1 is enabled).
first question: do you think it is fine to move the commands to the
executor service in CARD or should I move this functionally to the
InvoundHandler?
second question: do you have in mind other commands may block the
OOB/Regular thread and should be moved to a thread in the executor service?
Any other feedback is welcome.
thanks!
Cheers,
Pedro
PS. should I move this topic to ISPN-DEV?
[1] https://github.com/pruivo/infinispan/tree/ISPN-2808
[2]
https://github.com/pruivo/infinispan/commit/a267da0b2a4c785279141a9df1924...
11:09 a.m.
On 2/27/13 8:10 PM, Pedro Ruivo wrote:
-------- Original Message --------
Subject: ISPN-2808 - thread pool for incoming message [feedback]
Date: Wed, 27 Feb 2013 19:06:49 +0000
From: Pedro Ruivo <pedro(a)infinispan.org>
To: infinispan-core-dev(a)infinispan.org <infinispan-core-dev(a)infinispan.org>
Hi all,
I'm working on ISPN-2808 and I want some feedback about it (code is here
[1])
I'm starting to implement this feature but I know that Asynchronous
Invocation API is not totally finished in JGroups.
It is; I'll upload the 3.3.0.Beta1 version to Nexus in ca. 20 minutes...
My idea in to use an executor service in CommandAwareRpcDispatcher
(CARD) and when a request (command) is received, it checks if it is
useful to move the command execution to another thread (in this line [2])
For now, I'm thinking to move all the write commands, lock control
command, prepare command and commit command to the executor service
Can't a COMMIT or ROLLBACK be invoked directly ? Aren't they non-blocking ?
(Note: commit command is only moved when in DIST mode and L1 is
enabled).
first question: do you think it is fine to move the commands to the
executor service in CARD or should I move this functionally to the
InvoundHandler?
I thought we were going to implement this as *first* interceptor in the
chain, e.g. to allow thread locals to still work ? By the time a command
reaches the CommandAwareRpcDispatcher, we're already half-way through
the chain.
-- Bela Ban, JGroups lead (http://www.jgroups.org)
11:27 a.m.
On 02/28/2013 10:09 AM, Bela Ban wrote:
On 2/27/13 8:10 PM, Pedro Ruivo wrote:
>
>
> -------- Original Message --------
> Subject: ISPN-2808 - thread pool for incoming message [feedback]
> Date: Wed, 27 Feb 2013 19:06:49 +0000
> From: Pedro Ruivo <pedro(a)infinispan.org>
> To: infinispan-core-dev(a)infinispan.org <infinispan-core-dev(a)infinispan.org>
>
> Hi all,
>
> I'm working on ISPN-2808 and I want some feedback about it (code is here
> [1])
>
> I'm starting to implement this feature but I know that Asynchronous
> Invocation API is not totally finished in JGroups.
It is; I'll upload the 3.3.0.Beta1 version to Nexus in ca. 20 minutes...
>
> My idea in to use an executor service in CommandAwareRpcDispatcher
> (CARD) and when a request (command) is received, it checks if it is
> useful to move the command execution to another thread (in this line [2])
>
> For now, I'm thinking to move all the write commands, lock control
> command, prepare command and commit command to the executor service
Can't a COMMIT or ROLLBACK be invoked directly ? Aren't they non-blocking ?
the COMMIT can block is you are in DIST + L1 because it needs to sends a
sync L1 invalidation command.
> (Note: commit command is only moved when in DIST mode and L1 is enabled).
>
> first question: do you think it is fine to move the commands to the
> executor service in CARD or should I move this functionally to the
> InvoundHandler?
I thought we were going to implement this as *first* interceptor in the
chain, e.g. to allow thread locals to still work ? By the time a command
reaches the CommandAwareRpcDispatcher, we're already half-way through
the chain.
I'm not sure if I understood what you mean. However, I though that
this
thread pool is only for the incoming commands to avoid blocking the
OOB/Regular threads in JGroups thread pools. So, the
CommandAwareRpcDispatcher is the first component to know about the
remote command and can move them to another thread before they reaching
the interceptor chain.
-- Bela Ban, JGroups lead (http://www.jgroups.org)
_______________________________________________
infinispan-dev mailing list
infinispan-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/infinispan-dev
11:49 a.m.
On 2/28/13 11:27 AM, Pedro Ruivo wrote:
On 02/28/2013 10:09 AM, Bela Ban wrote:
> On 2/27/13 8:10 PM, Pedro Ruivo wrote:
>>
>> -------- Original Message --------
>> Subject: ISPN-2808 - thread pool for incoming message [feedback]
>> Date: Wed, 27 Feb 2013 19:06:49 +0000
>> From: Pedro Ruivo <pedro(a)infinispan.org>
>> To: infinispan-core-dev(a)infinispan.org
<infinispan-core-dev(a)infinispan.org>
>>
>> Hi all,
>>
>> I'm working on ISPN-2808 and I want some feedback about it (code is here
>> [1])
>>
>> I'm starting to implement this feature but I know that Asynchronous
>> Invocation API is not totally finished in JGroups.
> It is; I'll upload the 3.3.0.Beta1 version to Nexus in ca. 20 minutes...
>
>
>> My idea in to use an executor service in CommandAwareRpcDispatcher
>> (CARD) and when a request (command) is received, it checks if it is
>> useful to move the command execution to another thread (in this line [2])
>>
>> For now, I'm thinking to move all the write commands, lock control
>> command, prepare command and commit command to the executor service
>
> Can't a COMMIT or ROLLBACK be invoked directly ? Aren't they non-blocking ?
the COMMIT can block is you are in DIST + L1 because it needs to sends a
sync L1 invalidation command.
OK, makes sense then
>> (Note: commit command is only moved when in DIST mode and L1
is enabled).
>>
>> first question: do you think it is fine to move the commands to the
>> executor service in CARD or should I move this functionally to the
>> InvoundHandler?
>
> I thought we were going to implement this as *first* interceptor in the
> chain, e.g. to allow thread locals to still work ? By the time a command
> reaches the CommandAwareRpcDispatcher, we're already half-way through
> the chain.
I'm not sure if I understood what you mean. However, I though that this
thread pool is only for the incoming commands to avoid blocking the
OOB/Regular threads in JGroups thread pools. So, the
CommandAwareRpcDispatcher is the first component to know about the
remote command and can move them to another thread before they reaching
the interceptor chain.
OK, good
--
Bela Ban, JGroups lead (http://www.jgroups.org)
4326
days inactive
4327
days old
infinispan-dev@lists.jboss.org
3 comments
2 participants
participants (2)
-
Bela Ban -
Pedro Ruivo