You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@cassandra.apache.org by Drew Kutcharian <dr...@venarc.com> on 2012/11/05 18:49:12 UTC

Single Node Cassandra Installation

Hey Guys,

What should I look out for when deploying a single node installation? We want to launch a product that uses Cassandra and since we are going to have very little load initially, we were thinking of just going live with one node and eventually add more nodes as the load (hopefully) grows. Is this practice recommended?

Thanks,

Drew

Re: Single Node Cassandra Installation

Posted by Rob Coli <rc...@palominodb.com>.
On Sat, Nov 10, 2012 at 6:16 PM, Drew Kutcharian <dr...@venarc.com> wrote:
> Thanks Rob, this makes sense. We only have one rack at this point, so I think it'd be better to start with PropertyFileSnitch to make Cassandra think that these nodes each are in a different rack without having to put them on different subnets. And I will have more flexibility (at the cost of keeping the property file in sync) when it comes to growth.

Many people run successfully with PFS, and as you say it provides
flexibility if you get a second rack. The overhead versus a
non-rack-aware snitch is not significant.

However if you are careful you should be able to switch to PFS or
another rack aware snitch with no problems when you actually need
it... :)

=Rob

-- 
=Robert Coli
AIM&GTALK - rcoli@palominodb.com
YAHOO - rcoli.palominob
SKYPE - rcoli_palominodb

Re: Single Node Cassandra Installation

Posted by Drew Kutcharian <dr...@venarc.com>.
Thanks Rob, this makes sense. We only have one rack at this point, so I think it'd be better to start with PropertyFileSnitch to make Cassandra think that these nodes each are in a different rack without having to put them on different subnets. And I will have more flexibility (at the cost of keeping the property file in sync) when it comes to growth. What do you think?

-- Drew


On Nov 5, 2012, at 7:50 PM, Rob Coli <rc...@palominodb.com> wrote:

> On Mon, Nov 5, 2012 at 12:23 PM, Drew Kutcharian <dr...@venarc.com> wrote:
>>> Switching from SimpleStrategy to RackAware can be a pain.
>> 
>> Can you elaborate a bit? What would be the pain point?
> 
> If you don't maintain the same replica placement vis a vis nodes on
> your cluster, you have to dump and reload.
> 
> Simple example, 6 node cluster RF=3 :
> 
> SimpleSnitch : A B C D E F
> 
> Data for natural range of A is also on B and C, the "next" nodes in the ring.
> 
> RackAwareSnitches : A B C D E F
> "racks" they are in  :  1 1  2  2 3 3
> 
> Data for natural range of A is also on C and E, because despite not
> being the next nodes in the RING, they are the first nodes in the next
> rack.
> 
> If however you go from simple to rack aware and put your nodes in racks like :
> 
> A B C D E F
> 1 2 3 1 2 3
> 
> Then you have the same replica placement that SimpleStrategy gives you
> and can safely switch strategies/snitches on an existing cluster. Data
> for A is on B and C, on the same hosts, but for different reasons. Use
> nodetool getendpoints to test.
> 
> =Rob
> 
> -- 
> =Robert Coli
> AIM&GTALK - rcoli@palominodb.com
> YAHOO - rcoli.palominob
> SKYPE - rcoli_palominodb


Re: Single Node Cassandra Installation

Posted by Rob Coli <rc...@palominodb.com>.
On Mon, Nov 5, 2012 at 12:23 PM, Drew Kutcharian <dr...@venarc.com> wrote:
>> Switching from SimpleStrategy to RackAware can be a pain.
>
> Can you elaborate a bit? What would be the pain point?

If you don't maintain the same replica placement vis a vis nodes on
your cluster, you have to dump and reload.

Simple example, 6 node cluster RF=3 :

SimpleSnitch : A B C D E F

Data for natural range of A is also on B and C, the "next" nodes in the ring.

RackAwareSnitches : A B C D E F
"racks" they are in  :  1 1  2  2 3 3

Data for natural range of A is also on C and E, because despite not
being the next nodes in the RING, they are the first nodes in the next
rack.

If however you go from simple to rack aware and put your nodes in racks like :

A B C D E F
1 2 3 1 2 3

Then you have the same replica placement that SimpleStrategy gives you
and can safely switch strategies/snitches on an existing cluster. Data
for A is on B and C, on the same hosts, but for different reasons. Use
nodetool getendpoints to test.

=Rob

-- 
=Robert Coli
AIM&GTALK - rcoli@palominodb.com
YAHOO - rcoli.palominob
SKYPE - rcoli_palominodb

Re: Single Node Cassandra Installation

Posted by Drew Kutcharian <dr...@venarc.com>.
I understand that with one node we will have no HA, but since we are just starting out we wanted to see what would be the bare minimum to go to production with and as we see traction we can add more nodes.

> Switching from SimpleStrategy to RackAware can be a pain.

Can you elaborate a bit? What would be the pain point?




On Nov 5, 2012, at 10:12 AM, Michael Kjellman <mk...@barracuda.com> wrote:

> Should be fine if one node can deal with your read and write load.
> Switching from SimpleStrategy to RackAware can be a pain. That¹s a
> potential growth point way down the line (if you ever have your nodes on
> different switches). You might want to just setup your keyspace as
> RackAware if you intend to keep this keyspace in production in the future.
> 
> Redistributing load could take a while if your one node gets really
> loaded, but obviously there are well documented ways to fix this with
> nodetool.
> 
> I'd make sure you start off your single node with a token of 0 so it's
> easy to add more nodes later.
> 
> Also, with a single node you will have no replication so I assume this is
> okay in your production environment.
> 
> On 11/5/12 9:59 AM, "zGreenfelder" <zg...@gmail.com> wrote:
> 
>> On Mon, Nov 5, 2012 at 12:49 PM, Drew Kutcharian <dr...@venarc.com> wrote:
>>> Hey Guys,
>>> 
>>> What should I look out for when deploying a single node installation?
>>> We want to launch a product that uses Cassandra and since we are going
>>> to have very little load initially, we were thinking of just going live
>>> with one node and eventually add more nodes as the load (hopefully)
>>> grows. Is this practice recommended?
>>> 
>> 
>> I'm far from an expert and single node should be fine for development,
>> but for your production site, I'd suggest going with a minimal 'real'
>> cluster (2-3 nodes).   in my limited experience with hadoop, single
>> node is significantly different from multinode, and I wouldn't want to
>> start from a position where config  is certain to have to
>> fundamentally change as opposed to just grow.   But that's just my
>> first guess, you could try a dev move from a single node to a dual
>> node to find out how difficult it might be.
>> 
>> 
>> -- 
>> Even the Magic 8 ball has an opinion on email clients: Outlook not so
>> good.
> 
> 
> 'Like' us on Facebook for exclusive content and other resources on all Barracuda Networks solutions.
> Visit http://barracudanetworks.com/facebook
> 
> 


Re: Single Node Cassandra Installation

Posted by Michael Kjellman <mk...@barracuda.com>.
Should be fine if one node can deal with your read and write load.
Switching from SimpleStrategy to RackAware can be a pain. That¹s a
potential growth point way down the line (if you ever have your nodes on
different switches). You might want to just setup your keyspace as
RackAware if you intend to keep this keyspace in production in the future.

Redistributing load could take a while if your one node gets really
loaded, but obviously there are well documented ways to fix this with
nodetool.

I'd make sure you start off your single node with a token of 0 so it's
easy to add more nodes later.

Also, with a single node you will have no replication so I assume this is
okay in your production environment.

On 11/5/12 9:59 AM, "zGreenfelder" <zg...@gmail.com> wrote:

>On Mon, Nov 5, 2012 at 12:49 PM, Drew Kutcharian <dr...@venarc.com> wrote:
>> Hey Guys,
>>
>> What should I look out for when deploying a single node installation?
>>We want to launch a product that uses Cassandra and since we are going
>>to have very little load initially, we were thinking of just going live
>>with one node and eventually add more nodes as the load (hopefully)
>>grows. Is this practice recommended?
>>
>
>I'm far from an expert and single node should be fine for development,
>but for your production site, I'd suggest going with a minimal 'real'
>cluster (2-3 nodes).   in my limited experience with hadoop, single
>node is significantly different from multinode, and I wouldn't want to
>start from a position where config  is certain to have to
>fundamentally change as opposed to just grow.   But that's just my
>first guess, you could try a dev move from a single node to a dual
>node to find out how difficult it might be.
>
>
>-- 
>Even the Magic 8 ball has an opinion on email clients: Outlook not so
>good.


'Like' us on Facebook for exclusive content and other resources on all Barracuda Networks solutions.
Visit http://barracudanetworks.com/facebook



Re: Single Node Cassandra Installation

Posted by zGreenfelder <zg...@gmail.com>.
On Mon, Nov 5, 2012 at 12:49 PM, Drew Kutcharian <dr...@venarc.com> wrote:
> Hey Guys,
>
> What should I look out for when deploying a single node installation? We want to launch a product that uses Cassandra and since we are going to have very little load initially, we were thinking of just going live with one node and eventually add more nodes as the load (hopefully) grows. Is this practice recommended?
>

I'm far from an expert and single node should be fine for development,
but for your production site, I'd suggest going with a minimal 'real'
cluster (2-3 nodes).   in my limited experience with hadoop, single
node is significantly different from multinode, and I wouldn't want to
start from a position where config  is certain to have to
fundamentally change as opposed to just grow.   But that's just my
first guess, you could try a dev move from a single node to a dual
node to find out how difficult it might be.


-- 
Even the Magic 8 ball has an opinion on email clients: Outlook not so good.