Multimangos

[zhenya`]

17:47 <@Zor> No, not CORBA.

17:47 <@Zor> But ACE+TAO? Yes.

17:47 <@Zor> (TAO is a ACE-based CORBA framework)

[epozdnyakoff]

Maps use lot memory and then triggring data swaping or not allow use disabled map unloading and use aditional time for map loading.

What are performance bottlenecks in Mangos anyway? Suppose we put _lots_ of memory into server, what will be the next candidate for optimization?

[Lord X-Press]

What are performance bottlenecks in Mangos anyway? Suppose we put _lots_ of memory into server, what will be the next candidate for optimization?

Youre DB, check it with Jet Profiler and find spikes and there reasons

[epozdnyakoff]

Youre DB, check it with Jet Profiler and find spikes and there reasons

But it doesn't have anything to do with multiprocessing. It's more of database server's load balancing capabilities and it's deployment on a separate machine. I couldn't think of anything more than vmaps.

[Lord X-Press]

But it doesn't have anything to do with multiprocessing. It's more of database server's load balancing capabilities and it's deployment on a separate machine. I couldn't think of anything more than vmaps.

Thats correct, but you asked for bottelnecks, and the biggest bottleneck is a unoptimized DB, when youre mysql process stucks mangos can't do anything against this...multiprocessing might be better for multirealm system and there performance, but has no effect on the mysql service

[Patman128]

Yes, MySQL can be slow, but most database stuff is loaded into memory anyway, so the only time it calls MySQL is for character data, which is fairly non-frequent (on loading a character and saving them, which is every few minutes) Optimizing mangos would make a big difference on large scale performance.

[spermik]

Hi, I will add my little opinion here. I generally agree with Alex`s idea. So the really quick description of multimangos architecture:

1) realmd remains unchanged

2) all clients connects to the supervisior server (which would be the hardest part), announced as realm in realmlist

3) "behind" the supervisior there are several game(or map) servers (simillar to curent worldd)

4) supervisior handles the easy and fast packets (like chat)

5) more complicated opcodes are routed to game servers, where they are executed

6) the game server manages most of the game logic (ie AIs, spells, colision calculations)

7) 1 game server can run more maps (ie all instances, battleground server,...) 1 map = 1 thread

8) access to the world database perform all game server on their own

9) access to the character database is made via the supervisior, which solves the race conditions in creating new items (mails, characters...)

10) communication between supervisior and the map servers is made by TCP/IP sockets, probably 1 connection per map.

I hope this will help you deciding how to write it

Cheers Tabak

[zhenya`]

8) access to the world database perform all game server on their own

may fail. because how to synchronisation between game servers? :-D

[DasBlub]

8) access to the world database perform all game server on their own

may fail. because how to synchronisation between game servers? :-D

the worlddatabase is normaly not changed, only when you change something with a gm-command (like spawning creatures), and for this case you could implement a ZMSG_WORLD_DATABASE_CHANGED which will be send to the single map-servers.

4) supervisior handles the easy and fast packets (like chat)

i would create an own process for the chat too

[zhenya`]

chat gmtickets social auction mail

[alexluana]

any news?

Cheers

[Derex]

The project is in progress, we have implemented realm, proxy and database servers. AKA the cluster works upto level characters screen.

At this point we start implementing Maps. If somebody is interested in the project:

#multimangos @ irc.freenode.net

http://www.assembla.com/spaces/dashboard/index/multimangos

[Ebrithil]

Great work Derex, as usual.

Btw, can you pls shortly explain me what do you mean with "proxy" for MaNGOS?

Ty

[FeRkEl]

Great work Derex, as usual.

Btw, can you pls shortly explain me what do you mean with "proxy" for MaNGOS?

Ty

don't know if i am right, but think it is the part redirect the packets send from client to the server cluster parts like maps or so

[XTZGZoReX]

the worlddatabase is normaly not changed, only when you change something with a gm-command (like spawning creatures), and for this case you could implement a ZMSG_WORLD_DATABASE_CHANGED which will be send to the single map-servers.

It's not like we're using some binary protocol to send opcodes, so...

You people should jump on the IRC if you want to follow development more.

[nanounico]

derex, really god work, but, i ask, why dont put in a git?

[DasBlub]

it is in git... assembla uses git...

[nanounico]

yes, but in this, is necessary configure for work fine, if you use, git pull git://git.assembla.com/multimangos.git dont work.!

[DasBlub]

yes, but in this, is necessary configure for work fine, if you use, git pull git://git.assembla.com/multimangos.git dont work.!

it is 'git clone' and not 'git pull' to get the repo the first time

[nanounico]

ok, and later git pull from locale?

EDIT: Any test this? is stable?

[Derex]

It is not stabile and will not be stabile soon, atm I am just adding map servers to the cluster, aka hacking mangos to work as a map server.

[voidx]

any news about this?

[jdguzman]

if you want news about this I would recommend following the commits on the projects repo. If you look on there you will see that they are far from having anything ready and have set themselves a goal of June 2010 for the first release.

Cheers.

[xeross155]

Hmm, would be awesome if this would eventually become the mangos main branch.

[waza123]

Any news about Multi-processing ?