aboutsummaryrefslogtreecommitdiff
path: root/pipermail/nel/2001-February/000217.html
diff options
context:
space:
mode:
authorneodarz <neodarz@neodarz.net>2018-08-11 20:21:34 +0200
committerneodarz <neodarz@neodarz.net>2018-08-11 20:21:34 +0200
commit0ea5fc66924303d1bf73ba283a383e2aadee02f2 (patch)
tree2568e71a7ccc44ec23b8bb3f0ff97fb6bf2ed709 /pipermail/nel/2001-February/000217.html
downloadnevrax-website-self-hostable-0ea5fc66924303d1bf73ba283a383e2aadee02f2.tar.xz
nevrax-website-self-hostable-0ea5fc66924303d1bf73ba283a383e2aadee02f2.zip
Initial commit
Diffstat (limited to 'pipermail/nel/2001-February/000217.html')
-rw-r--r--pipermail/nel/2001-February/000217.html138
1 files changed, 138 insertions, 0 deletions
diff --git a/pipermail/nel/2001-February/000217.html b/pipermail/nel/2001-February/000217.html
new file mode 100644
index 00000000..abdd65c1
--- /dev/null
+++ b/pipermail/nel/2001-February/000217.html
@@ -0,0 +1,138 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
+<HTML>
+ <HEAD>
+ <TITLE> [Nel] Dynamic load balancing?</TITLE>
+ <LINK REL="Index" HREF="index.html" >
+ <LINK REL="made" HREF="mailto:jmark4%40home.com">
+ <LINK REL="Previous" HREF="000232.html">
+ <LINK REL="Next" HREF="000229.html">
+ </HEAD>
+ <BODY BGCOLOR="#ffffff">
+ <H1>[Nel] Dynamic load balancing?</H1>
+ <B>Jared Mark</B>
+ <A HREF="mailto:jmark4%40home.com"
+ TITLE="[Nel] Dynamic load balancing?">jmark4@home.com</A><BR>
+ <I>Mon, 19 Feb 2001 15:43:59 -0600</I>
+ <P><UL>
+ <LI> Previous message: <A HREF="000232.html">[Nel] Dynamic load balancing?</A></li>
+ <LI> Next message: <A HREF="000229.html">[Nel] Dynamic load balancing?</A></li>
+ <LI> <B>Messages sorted by:</B>
+ <a href="date.html#217">[ date ]</a>
+ <a href="thread.html#217">[ thread ]</a>
+ <a href="subject.html#217">[ subject ]</a>
+ <a href="author.html#217">[ author ]</a>
+ </LI>
+ </UL>
+ <HR>
+<!--beginarticle-->
+<PRE>Okay, so what you're saying is basically this...
+
+By seperating the services, you can have a cluster of machines, each running
+different services, and thus, keeping the load down across the board. You
+would have a single system, for example, handling just combat resolution,
+and you would have another system that would keep track of the database...
+or on an even more molecular level (assuming you have a very large world
+with huge database demands) a database service that is strictly dedicated to
+handling certain types of objects (like dynamic world objects such as rocks
+and such, as opposed to character objects like their posessions in their
+backpacks for example).
+
+You would have other servers running the service managers for the AI of the
+monsters, etc, etc, etc...
+
+So instead of dividing up the world into geographical sectors and having a
+single system handle each sector... you would divide the world up into
+different processes, and have each system handle the specific process for
+the entire world.
+
+Let me know if I'm getting pretty close to what you're trying to say here.
+;) My only question on the whole thing would be... what happens if you run
+into a problem where activity on a process gets to be too much for a
+server... would you be forced to upgrade that server? Or could you
+theoretically split the process between two systems? (running the same
+serice twice in the same cluster, on different systems).
+
+----- Original Message -----
+From: &quot;Vincent Archer&quot; &lt;<A HREF="mailto:archer@nevrax.com">archer@nevrax.com</A>&gt;
+To: &lt;<A HREF="mailto:nel@nevrax.org">nel@nevrax.org</A>&gt;
+Sent: Monday, February 19, 2001 10:33 AM
+Subject: Re: [Nel] Dynamic load balancing?
+
+
+&gt;<i> According to Glaze:
+</I>&gt;<i> &gt; Hello,
+</I>&gt;<i> &gt;
+</I>&gt;<i> &gt; I have one &quot;small&quot; question and I hope this is the right place to ask
+</I>it. I
+&gt;<i> &gt; was just wonder how (or if) you will solve the question of how to
+</I>balance
+&gt;<i> &gt; the workload on the game servers, will you use some kind of dynamic load
+</I>&gt;<i> &gt; balancing or have you opted for another approach.
+</I>&gt;<i>
+</I>&gt;<i> Hmm, I realise I've explained what is dynamic load balancing, but I
+</I>&gt;<i> haven't answered this question.
+</I>&gt;<i>
+</I>&gt;<i> Let's hope we have a good discussion on this :)
+</I>&gt;<i>
+</I>&gt;<i> If you look at some of the basics in the... ok in the future design
+</I>&gt;<i> documents, you see we're working with a kind of ORB approach, with
+</I>&gt;<i> a naming service, which lets you discuss with specific services.
+</I>&gt;<i>
+</I>&gt;<i> We're aiming for a functional approach, not the classic EQ/UO/AC
+</I>&gt;<i> geographical approach. In that classic model, each processus is a
+</I>&gt;<i> complete copy of the code. It contains everything, from spell effects
+</I>&gt;<i> to pathfinding to combat code. Each process serves a specific &quot;area&quot;,
+</I>&gt;<i> which is defined by an X/Y/Z box (or, for EQ, a zone ID number, since
+</I>&gt;<i> each area is strictly separate).
+</I>&gt;<i>
+</I>&gt;<i> We're not going to use this approach. Chiefly because static allocation
+</I>&gt;<i> fails when you put too many objects in the same area (be it because a
+</I>&gt;<i> guild has decided to hold a meeting of all its 350 members there, or
+</I>&gt;<i> because we want to have an event with an army of 100 mobs spawning).
+</I>&gt;<i> And, as I explained somewhere else, dynamic allocation means:
+</I>&gt;<i> 1) A complete lack of control on what process is doing what, and
+</I>&gt;<i> 2) A need for a very fast way of finding who is near you and under
+</I>&gt;<i> which process control
+</I>&gt;<i>
+</I>&gt;<i> What we're aiming for is a functional approach. That is, each service
+</I>&gt;<i> provides specific functions. For example, all items in the game are
+</I>&gt;<i> handled by a single service. Combat runs on a separate service.
+</I>&gt;<i> And so on.
+</I>&gt;<i>
+</I>&gt;<i> I'll let you imagine how this works, and then will further explain as
+</I>&gt;<i> you try to shoot holes into that design. :)
+</I>&gt;<i>
+</I>&gt;<i> --
+</I>&gt;<i> Vincent Archer Email:
+</I><A HREF="mailto:archer@nevrax.com">archer@nevrax.com</A>
+&gt;<i>
+</I>&gt;<i> Nevrax France. Off on the yellow brick road
+</I>we go!
+&gt;<i> _______________________________________________
+</I>&gt;<i> Nel mailing list
+</I>&gt;<i> <A HREF="mailto:Nel@nevrax.org">Nel@nevrax.org</A>
+</I>&gt;<i> <A HREF="http://www.nevrax.org/mailman/listinfo.cgi/nel">http://www.nevrax.org/mailman/listinfo.cgi/nel</A>
+</I>&gt;<i>
+</I>
+
+</pre>
+
+
+
+
+
+
+<!--endarticle-->
+ <HR>
+ <P><UL>
+ <!--threads-->
+ <LI> Previous message: <A HREF="000232.html">[Nel] Dynamic load balancing?</A></li>
+ <LI> Next message: <A HREF="000229.html">[Nel] Dynamic load balancing?</A></li>
+ <LI> <B>Messages sorted by:</B>
+ <a href="date.html#217">[ date ]</a>
+ <a href="thread.html#217">[ thread ]</a>
+ <a href="subject.html#217">[ subject ]</a>
+ <a href="author.html#217">[ author ]</a>
+ </LI>
+ </UL>
+</body></html>