Setting up my Riak architecture

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Setting up my Riak architecture

prog112
Hello.
I'm currently developing my database relationships with my other server applications and I'm having a hard time figuring out how to approach it. Here's what I'm wondering about:

1) How does accessing Riak really work? Currently I'm having one machine with all my tools (Riak, master server, dynamic game server instnaces) on it. My database is set up on 127.0.0.1 in /etc/riak/app.config does it mean it can only be accessed from localhost?
2) Say later on I'm going to have more seperate machines to run my game server nodes. My game servers do require to get some info about players from the database. Should I connect to database directly from those game servers instances or should I do it via master server? Won't master server get too much traffic then though? How can I connect to database from my game servers anyway if they're at separate machines?

What do you guys think? All feedback will be very appreciated.
Reply | Threaded
Open this post in threaded view
|

Re: Setting up my Riak architecture

Guido Medina-2
Riak is a multi-master database, your client will create a connection pool to all nodes in the cluster.
For production it is recommended at least 5 nodes in the cluster, to get started, I guess you should first read some documentation, here is some that can help you:

http://docs.basho.com/riak/latest/

Start with the section "Riak for Operators" and once you are comfortable, jump to "Riak for Developers"

HTH,

Guido.

On 24/07/13 13:51, prog112 wrote:
Hello.
I'm currently developing my database relationships with my other server
applications and I'm having a hard time figuring out how to approach it.
Here's what I'm wondering about:

1) How does accessing Riak really work? Currently I'm having one machine
with all my tools (Riak, master server, dynamic game server instnaces) on
it. My database is set up on 127.0.0.1 in /etc/riak/app.config does it mean
it can only be accessed from localhost?
2) Say later on I'm going to have more seperate machines to run my game
server nodes. My game servers do require to get some info about players from
the database. Should I connect to database directly from those game servers
instances or should I do it via master server? Won't master server get too
much traffic then though? How can I connect to database from my game servers
anyway if they're at separate machines?

What do you guys think? All feedback will be very appreciated. 



--
View this message in context: http://riak-users.197444.n3.nabble.com/Setting-up-my-Riak-architecture-tp4028520.html
Sent from the Riak Users mailing list archive at Nabble.com.

_______________________________________________
riak-users mailing list
[hidden email]
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com


_______________________________________________
riak-users mailing list
[hidden email]
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
Reply | Threaded
Open this post in threaded view
|

Re: Setting up my Riak architecture

Hector Castro-2
In reply to this post by prog112
Hi Maksymilian,

My response is inline below.

--
Hector


On Wed, Jul 24, 2013 at 8:51 AM, prog112 <[hidden email]> wrote:

>
> Hello.
> I'm currently developing my database relationships with my other server
> applications and I'm having a hard time figuring out how to approach it.
> Here's what I'm wondering about:
>
> 1) How does accessing Riak really work? Currently I'm having one machine
> with all my tools (Riak, master server, dynamic game server instnaces) on
> it. My database is set up on 127.0.0.1 in /etc/riak/app.config does it mean
> it can only be accessed from localhost?

We recommend that production Riak deployments run with at least 5
nodes. There are several reasons why we recommend 5, and they're
outlined pretty well in this post. [0]

If you want Riak to only be accessible via the local host, then
binding to 127.0.0.1 works. If you want to communicate with Riak from
other nodes, bind to an address that is accessible to the other nodes,
or 0.0.0.0.

> 2) Say later on I'm going to have more seperate machines to run my game
> server nodes. My game servers do require to get some info about players from
> the database. Should I connect to database directly from those game servers
> instances or should I do it via master server? Won't master server get too
> much traffic then though? How can I connect to database from my game servers
> anyway if they're at separate machines?

Any node in a Riak cluster can coordinate requests for any other Riak
node that is responsible for the data you're reading/writing. That
said, it's common for users to deploy Riak clusters behind a load
balancer that evenly distributes requests to each node in the cluster.
[1]

Hope that helps.

> What do you guys think? All feedback will be very appreciated.
>

[0] http://basho.com/why-your-riak-cluster-should-have-at-least-five-nodes/
[1] http://docs.basho.com/riak/latest/ops/advanced/configs/load-balancing-proxy/

>
> --
> View this message in context: http://riak-users.197444.n3.nabble.com/Setting-up-my-Riak-architecture-tp4028520.html
> Sent from the Riak Users mailing list archive at Nabble.com.
>
> _______________________________________________
> riak-users mailing list
> [hidden email]
> http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com

_______________________________________________
riak-users mailing list
[hidden email]
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com