Where do you find 40gb disks now a days?

Normally your performance is going to be better with more space but your network may be your limiting factor for some computations.  That could give you some paradoxical scaling.  Hbase will rarely show this behavior.

Keep in mind you also want to allow for an os partition. Current standard practice is to reserve as much as 100 GB for that partition but in your case 10gb better:-)

Note that if you account for this, the node counts don't scale as simply.  The overhead of these os partitions goes up with number of nodes.  

On Jan 18, 2013, at 8:55 AM, Panshul Whisper <[EMAIL PROTECTED]> wrote:

> If we look at it with performance in mind,
> is it better to have 20 Nodes with 40 GB HDD
> or is it better to have 10 Nodes with 80 GB HDD?
>
> they are connected on a gigabit LAN
>
> Thnx
>
>
> On Fri, Jan 18, 2013 at 2:26 PM, Jean-Marc Spaggiari <[EMAIL PROTECTED]> wrote:
> 20 nodes with 40 GB will do the work.
>
> After that you will have to consider performances based on your access
> pattern. But that's another story.
>
> JM
>
> 2013/1/18, Panshul Whisper <[EMAIL PROTECTED]>:
> > Thank you for the replies,
> >
> > So I take it that I should have atleast 800 GB on total free space on
> > HDFS.. (combined free space of all the nodes connected to the cluster). So
> > I can connect 20 nodes having 40 GB of hdd on each node to my cluster. Will
> > this be enough for the storage?
> > Please confirm.
> >
> > Thanking You,
> > Regards,
> > Panshul.
> >
> >
> > On Fri, Jan 18, 2013 at 1:36 PM, Jean-Marc Spaggiari <
> > [EMAIL PROTECTED]> wrote:
> >
> >> Hi Panshul,
> >>
> >> If you have 20 GB with a replication factor set to 3, you have only
> >> 6.6GB available, not 11GB. You need to divide the total space by the
> >> replication factor.
> >>
> >> Also, if you store your JSon into HBase, you need to add the key size
> >> to it. If you key is 4 bytes, or 1024 bytes, it makes a difference.
> >>
> >> So roughly, 24 000 000 * 5 * 1024 = 114GB. You don't have the space to
> >> store it. Without including the key size. Even with a replication
> >> factor set to 5 you don't have the space.
> >>
> >> Now, you can add some compression, but even with a lucky factor of 50%
> >> you still don't have the space. You will need something like 90%
> >> compression factor to be able to store this data in your cluster.
> >>
> >> A 1T drive is now less than $100... So you might think about replacing
> >> you 20 GB drives by something bigger.
> >> to reply to your last question, for your data here, you will need AT
> >> LEAST 350GB overall storage. But that's a bare minimum. Don't go under
> >> 500GB.
> >>
> >> IMHO
> >>
> >> JM
> >>
> >> 2013/1/18, Panshul Whisper <[EMAIL PROTECTED]>:
> >> > Hello,
> >> >
> >> > I was estimating how much disk space do I need for my cluster.
> >> >
> >> > I have 24 million JSON documents approx. 5kb each
> >> > the Json is to be stored into HBASE with some identifying data in
> >> coloumns
> >> > and I also want to store the Json for later retrieval based on the Id
> >> data
> >> > as keys in Hbase.
> >> > I have my HDFS replication set to 3
> >> > each node has Hadoop and hbase and Ubuntu installed on it.. so approx
> >> > 11
> >> GB
> >> > is available for use on my 20 GB node.
> >> >
> >> > I have no idea, if I have not enabled Hbase replication, is the HDFS
> >> > replication enough to keep the data safe and redundant.
> >> > How much total disk space I will need for the storage of the data.
> >> >
> >> > Please help me estimate this.
> >> >
> >> > Thank you so much.
> >> >
> >> > --
> >> > Regards,
> >> > Ouch Whisper
> >> > 010101010101
> >> >
> >>
> >
> >
> >
> > --
> > Regards,
> > Ouch Whisper
> > 010101010101
> >
>
>
>
> --
> Regards,
> Ouch Whisper
> 010101010101

I disagree. There are some significant advantages to using "many small nodes" instead of "few big nodes". As Ted points out, there are some disadvantages as well, so you have to look at the trade-offs. But consider:

- NUMA: If your hadoop nodes span physical NUMA nodes, then performance will suffer from remote memory accesses. The Linux scheduler tries to minimize this, but I've found that about 1/3 of memory accesses are remote on a 2-socket machine. This effect will be more severe on bigger machines. Hadoop nodes that fit on a NUMA node will have not access remote memory at all (at least on vSphere).

- Disk partitioning: Smaller nodes with fewer disks each can significantly increase average disk utilization, not decrease it. Having many threads operating against many disks in the "big node" case tends to leave some disks idle while others are over-subscribed. Partitioning disks among nodes decreases this effect. The extreme case is one disk per node, where no disks will be idle as long as there is work to do.

- Management: Not a performance effect, but smaller nodes enable easier multi-tenancy, multiple virtual Hadoop clusters, sharing physical hardware with other workloads, etc.

Jeff

----- Original Message -----

From: "Ted Dunning" <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Sent: Friday, January 18, 2013 3:36:30 PM
Subject: Re: Estimating disk space requirements

If you make 20 individual small servers, that isn't much different from 20 from one server. The only difference would be if the neighbors of the separate VMs use less resource.
On Fri, Jan 18, 2013 at 3:34 PM, Panshul Whisper < [EMAIL PROTECTED] > wrote:

ah now i understand what you mean.
I will be creating 20 individual servers on the cloud, and not create one big server and make several virtual nodes inside it.
I will be paying for 20 different nodes.. all configured with hadoop and connected to the cluster.
Thanx for the intel :)

On Fri, Jan 18, 2013 at 11:59 PM, Ted Dunning < [EMAIL PROTECTED] > wrote:

<blockquote>
It is usually better to not subdivide nodes into virtual nodes. You will generally get better performance form the original node because you only pay for the OS once and because your disk I/O will be scheduled better.
If you look at EC2 pricing, however, the spot market often has arbitrage opportunities where one size node is absurdly cheap relative to others. In that case, it pays to scale the individual nodes up or down.
The only reasonable reason to split nodes to very small levels is for testing and training.
On Fri, Jan 18, 2013 at 2:30 PM, Panshul Whisper < [EMAIL PROTECTED] > wrote:

<blockquote>

Thnx for the reply Ted,
You can find 40 GB disks when u make virtual nodes on a cloud like Rackspace ;-)
About the os partitions I did not exactly understand what you meant.
I have made a server on the cloud.. And I just installed and configured hadoop and hbase in the /use/local folder.
And I am pretty sure it does not have a separate partition for root.
Please help me explain what u meant and what else precautions should I take.
Thanks,
Regards,
Ouch Whisper
01010101010
On Jan 18, 2013 11:11 PM, "Ted Dunning" < [EMAIL PROTECTED] > wrote:

<blockquote>
Where do you find 40gb disks now a days?
Normally your performance is going to be better with more space but your network may be your limiting factor for some computations. That could give you some paradoxical scaling. Hbase will rarely show this behavior.
Keep in mind you also want to allow for an os partition. Current standard practice is to reserve as much as 100 GB for that partition but in your case 10gb better:-)
Note that if you account for this, the node counts don't scale as simply. The overhead of these os partitions goes up with number of nodes.

On Jan 18, 2013, at 8:55 AM, Panshul Whisper < [EMAIL PROTECTED] > wrote:

<blockquote>
If we look at it with performance in mind,
is it better to have 20 Nodes with 40 GB HDD
or is it better to have 10 Nodes with 80 GB HDD?
they are connected on a gigabit LAN
Thnx

On Fri, Jan 18, 2013 at 2:26 PM, Jean-Marc Spaggiari < [EMAIL PROTECTED] > wrote:

<blockquote>
20 nodes with 40 GB will do the work.

After that you will have to consider performances based on your access
pattern. But that's another story.

JM

2013/1/18, Panshul Whisper < [EMAIL PROTECTED] >:
Regards, Ouch Whisper
010101010101
</blockquote>

</blockquote>

</blockquote>
</blockquote>
Regards, Ouch Whisper
010101010101
</blockquote>

Jeff makes some good points here.

On Fri, Jan 18, 2013 at 5:01 PM, Jeffrey Buell <[EMAIL PROTECTED]> wrote:

> I disagree.  There are some significant advantages to using "many small
> nodes" instead of "few big nodes".  As Ted points out, there are some
> disadvantages as well, so you have to look at the trade-offs.  But consider:
>
> - NUMA:  If your hadoop nodes span physical NUMA nodes, then performance
> will suffer from remote memory accesses.  The Linux scheduler tries to
> minimize this, but I've found that about 1/3 of memory accesses are remote
> on a 2-socket machine.  This effect will be more severe on bigger
> machines.  Hadoop nodes that fit on a NUMA node will have not access remote
> memory at all (at least on vSphere).
>

This is definitely a good point with respect to untainted Hadoop, but with
a system like MapR, there is a significant amount of core locality that
goes on to minimize NUMA-remote fetches.  This can have significant impact,
of course.

- Disk partitioning:  Smaller nodes with fewer disks each can significantly
> increase average disk utilization, not decrease it.  Having many threads
> operating against many disks in the "big node" case tends to leave some
> disks idle while others are over-subscribed.
>

Again, this is an implementation side-effect.  Good I/O scheduling and
proper striping can mitigate this substantially.

Going the other way, splitting disks between different VM's can be
disastrous.
>  Partitioning disks among nodes decreases this effect.  The extreme case
> is one disk per node, where no disks will be idle as long as there is work
> to do.
>

Yes.  Even deficient implementations should succeed in this case.

You do lose the ability to allow big-memory jobs that would otherwise span
multiple slots.
> - Management: Not a performance effect, but smaller nodes enable easier
> multi-tenancy, multiple virtual Hadoop clusters, sharing physical hardware
> with other workloads, etc.
>

Definitely true.