Mailing-List: contact river-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: river-dev@incubator.apache.org
Received-SPF: neutral (nike.apache.org: local policy)
MIME-Version: 1.0
In-Reply-To: <4A5CE6C6.4060709@wonderly.org>
References: <F5DF5420-6DCA-4873-A8E4-8E9F501D5CB3@mac.com>
	 <4A4921FA.3030408@wonderly.org>
	 <510143ac0907030226i22b4d6e9tf70857d357f115f4@mail.gmail.com>
	 <e736790907030825m322488e0rcb75dfb347eab059@mail.gmail.com>
	 <30B045EF-D11D-4AB6-988E-3F97CAE8419B@gmail.com>
	 <4A4EA0AD.4080807@zeus.net.au>
	 <93890df00907031726k22aa0b50ib2306c2ef304c6ea@mail.gmail.com>
	 <4A4EDD74.3010400@zeus.net.au>
	 <93890df00907121605sbece8c8g1b2b3d9f8c603e8f@mail.gmail.com>
	 <4A5CE6C6.4060709@wonderly.org>
Date: Wed, 15 Jul 2009 11:59:09 -0400
Message-ID: <a2cfc5ef0907150859y6e35bf90j233f0c545ca1095e@mail.gmail.com>
Subject: Re: [Was: OSGi and Jini] Now -> Next Steps
From: Peter Jones <pcj@roundroom.net>
To: river-dev@incubator.apache.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Just to be clear, the "com.sun.jini.jeri.tcp.useNIO=3Dtrue" mode of the
JERI TCP endpoint implementation still uses multiple threads to
dispatch remote invocations-- RMI behavior would be drastically
impacted if it only used a single thread for that (for reasons like
you describe below).  It just uses a single thread for performing I/O
operations, at least those that "would block".

-- Peter


On Tue, Jul 14, 2009 at 4:12 PM, Gregg Wonderly<gregg@wonderly.org> wrote:
> The Jini 2.0 software includes the ability to use NIO in the TCP endpoint=
.
> =A0This is activated by system property "com.sun.jini.jeri.tcp.useNIO". I=
f you
> look at TcpEndpoint and TcpServerEndpoint, you'll not find much different
> going on. =A0The biggest issue with a typical network "distributed" syste=
m, is
> circular wait that can occur as systems "randomly" develop new connection=
s
> to other system. =A0Using a single thread to dispatch events in a server =
for
> "load handling" is not a good thing for any "work" that can have "externa=
l"
> contact. =A0So, the use of NIO for
> "scalability" is find for things that don't end up interacting circularly=
.
> =A0For things that you have no idea how they will interact, it's better t=
o
> make sure that you have "new inbound call =3D=3D new inbound thread" so t=
hat you
> don't get into problems. =A0It may be that there is some work that you ca=
n do
> in a single thread, but that usually is something that a specific
> application optimizes.
>
> One way to do that is to use inbound calls to "queue" work. =A0But then y=
ou
> need to use "callbacks" to notify the caller of the results of there
> request, unless you design a very custom invocation layer that allows the
> use of a "Future" or some other "signaling" mechanism to control when the
> result is sent back to the caller.
>
> Gregg Wonderly
>