Hi,

If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
file, this file will ocupy 64MB in the HDFS?

Thanks,

But, how can I say that a 1KB file will only use 1KB of disc space, if
a block is configured has 64MB? In my view, if a 1KB use a block of
64MB, the file will occupy 64MB in the disc.

How can you disassociate a  64MB data block from HDFS of a disk block?

On Fri, Jun 10, 2011 at 5:01 PM, Marcos Ortiz <[EMAIL PROTECTED]> wrote:
> On 06/10/2011 10:35 AM, Pedro Costa wrote:
>
> Hi,
>
> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
> file, this file will ocupy 64MB in the HDFS?
>
> Thanks,
>
> HDFS is not very efficient storing small files, because each file is stored
> in a block (of 64 MB in your case), and the block metadata
> is held in memory by the NN. But you should know that this 1KB file only
> will use 1KB of disc space.
>
> For small files, you can use Hadoop archives.
> Regards
>
> --
> Marcos Luís Ortíz Valmaseda
>  Software Engineer (UCI)
>  http://marcosluis2186.posterous.com
>  http://twitter.com/marcosluis2186
>
>

On Fri, Jun 10, 2011 at 11:05 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:

> Hi,
>
> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
> file, this file will ocupy 64MB in the HDFS?
>

No, it will occupy something much closer to 1KB than 64MB. There is some
small overhead related to metadata about the block plus the space on the
underlying filesystem (ext2, ext3, etc)  required to store the 1KB file.
 You can just cd into the datanode data directories and look for the block
in question.

HDFS isn't really optimized for small files though.

 Sridhar

On Fri, Jun 10, 2011 at 8:42 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
> But, how can I say that a 1KB file will only use 1KB of disc space, if
> a block is configured has 64MB? In my view, if a 1KB use a block of
> 64MB, the file will occupy 64MB in the disc.

A block of HDFS is the unit of distribution and replication, not the
unit of storage.  HDFS uses the underlying file systems for physical
storage.

-- Philip

>
> How can you disassociate a  64MB data block from HDFS of a disk block?
>
> On Fri, Jun 10, 2011 at 5:01 PM, Marcos Ortiz <[EMAIL PROTECTED]> wrote:
>> On 06/10/2011 10:35 AM, Pedro Costa wrote:
>>
>> Hi,
>>
>> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
>> file, this file will ocupy 64MB in the HDFS?
>>
>> Thanks,
>>
>> HDFS is not very efficient storing small files, because each file is stored
>> in a block (of 64 MB in your case), and the block metadata
>> is held in memory by the NN. But you should know that this 1KB file only
>> will use 1KB of disc space.
>>
>> For small files, you can use Hadoop archives.
>> Regards
>>
>> --
>> Marcos Luís Ortíz Valmaseda
>>  Software Engineer (UCI)
>>  http://marcosluis2186.posterous.com
>>  http://twitter.com/marcosluis2186
>>
>>
>

On 06/10/2011 10:35 AM, Pedro Costa wrote:
> Hi,
>
> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
> file, this file will ocupy 64MB in the HDFS?
>
> Thanks,
HDFS is not very efficient storing small files, because each file is
stored in a block (of 64 MB in your case), and the block metadata
is held in memory by the NN. But you should know that this 1KB file only
will use 1KB of disc space.

For small files, you can use Hadoop archives.
Regards

--
Marcos Lu�s Ort�z Valmaseda
  Software Engineer (UCI)
  http://marcosluis2186.posterous.com
  http://twitter.com/marcosluis2186

This means that, when HDFS reads 1KB file from the disk, he will put
the data in blocks of 64MB?

On Fri, Jun 10, 2011 at 5:00 PM, Philip Zeyliger <[EMAIL PROTECTED]> wrote:
> On Fri, Jun 10, 2011 at 8:42 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
>> But, how can I say that a 1KB file will only use 1KB of disc space, if
>> a block is configured has 64MB? In my view, if a 1KB use a block of
>> 64MB, the file will occupy 64MB in the disc.
>
> A block of HDFS is the unit of distribution and replication, not the
> unit of storage.  HDFS uses the underlying file systems for physical
> storage.
>
> -- Philip
>
>>
>> How can you disassociate a  64MB data block from HDFS of a disk block?
>>
>> On Fri, Jun 10, 2011 at 5:01 PM, Marcos Ortiz <[EMAIL PROTECTED]> wrote:
>>> On 06/10/2011 10:35 AM, Pedro Costa wrote:
>>>
>>> Hi,
>>>
>>> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
>>> file, this file will ocupy 64MB in the HDFS?
>>>
>>> Thanks,
>>>
>>> HDFS is not very efficient storing small files, because each file is stored
>>> in a block (of 64 MB in your case), and the block metadata
>>> is held in memory by the NN. But you should know that this 1KB file only
>>> will use 1KB of disc space.
>>>
>>> For small files, you can use Hadoop archives.
>>> Regards
>>>
>>> --
>>> Marcos Luís Ortíz Valmaseda
>>>  Software Engineer (UCI)
>>>  http://marcosluis2186.posterous.com
>>>  http://twitter.com/marcosluis2186
>>>
>>>
>>
>

--
---------------------------
Pedro Sá da Costa

@: [EMAIL PROTECTED]
@: [EMAIL PROTECTED]

On Fri, Jun 10, 2011 at 9:08 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
> This means that, when HDFS reads 1KB file from the disk, he will put
> the data in blocks of 64MB?

No.

>
> On Fri, Jun 10, 2011 at 5:00 PM, Philip Zeyliger <[EMAIL PROTECTED]> wrote:
>> On Fri, Jun 10, 2011 at 8:42 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
>>> But, how can I say that a 1KB file will only use 1KB of disc space, if
>>> a block is configured has 64MB? In my view, if a 1KB use a block of
>>> 64MB, the file will occupy 64MB in the disc.
>>
>> A block of HDFS is the unit of distribution and replication, not the
>> unit of storage.  HDFS uses the underlying file systems for physical
>> storage.
>>
>> -- Philip
>>
>>>
>>> How can you disassociate a  64MB data block from HDFS of a disk block?
>>>
>>> On Fri, Jun 10, 2011 at 5:01 PM, Marcos Ortiz <[EMAIL PROTECTED]> wrote:
>>>> On 06/10/2011 10:35 AM, Pedro Costa wrote:
>>>>
>>>> Hi,
>>>>
>>>> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
>>>> file, this file will ocupy 64MB in the HDFS?
>>>>
>>>> Thanks,
>>>>
>>>> HDFS is not very efficient storing small files, because each file is stored
>>>> in a block (of 64 MB in your case), and the block metadata
>>>> is held in memory by the NN. But you should know that this 1KB file only
>>>> will use 1KB of disc space.
>>>>
>>>> For small files, you can use Hadoop archives.
>>>> Regards
>>>>
>>>> --
>>>> Marcos Luís Ortíz Valmaseda
>>>>  Software Engineer (UCI)
>>>>  http://marcosluis2186.posterous.com
>>>>  http://twitter.com/marcosluis2186
>>>>
>>>>
>>>
>>
>
>
>
> --
> ---------------------------
> Pedro Sá da Costa
>
> @: [EMAIL PROTECTED]
> @: [EMAIL PROTECTED]
>

So, I'm not getting how a 1KB file can cost a block of 64MB. Can
anyone explain me?

On Fri, Jun 10, 2011 at 5:13 PM, Philip Zeyliger <[EMAIL PROTECTED]> wrote:
> On Fri, Jun 10, 2011 at 9:08 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
>> This means that, when HDFS reads 1KB file from the disk, he will put
>> the data in blocks of 64MB?
>
> No.
>
>>
>> On Fri, Jun 10, 2011 at 5:00 PM, Philip Zeyliger <[EMAIL PROTECTED]> wrote:
>>> On Fri, Jun 10, 2011 at 8:42 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
>>>> But, how can I say that a 1KB file will only use 1KB of disc space, if
>>>> a block is configured has 64MB? In my view, if a 1KB use a block of
>>>> 64MB, the file will occupy 64MB in the disc.
>>>
>>> A block of HDFS is the unit of distribution and replication, not the
>>> unit of storage.  HDFS uses the underlying file systems for physical
>>> storage.
>>>
>>> -- Philip
>>>
>>>>
>>>> How can you disassociate a  64MB data block from HDFS of a disk block?
>>>>
>>>> On Fri, Jun 10, 2011 at 5:01 PM, Marcos Ortiz <[EMAIL PROTECTED]> wrote:
>>>>> On 06/10/2011 10:35 AM, Pedro Costa wrote:
>>>>>
>>>>> Hi,
>>>>>
>>>>> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
>>>>> file, this file will ocupy 64MB in the HDFS?
>>>>>
>>>>> Thanks,
>>>>>
>>>>> HDFS is not very efficient storing small files, because each file is stored
>>>>> in a block (of 64 MB in your case), and the block metadata
>>>>> is held in memory by the NN. But you should know that this 1KB file only
>>>>> will use 1KB of disc space.
>>>>>
>>>>> For small files, you can use Hadoop archives.
>>>>> Regards
>>>>>
>>>>> --
>>>>> Marcos Luís Ortíz Valmaseda
>>>>>  Software Engineer (UCI)
>>>>>  http://marcosluis2186.posterous.com
>>>>>  http://twitter.com/marcosluis2186
>>>>>
>>>>>
>>>>
>>>
>>

I am also relatively new to hadoop, so others may feel free to correct me if
am wrong.

NN keeps track of a file by "inode" and the blocks related to that inode. In
your case, since your file size is smaller than the block size, NN will have
only ONE block associated with this inode (assuming only one replication).
To track the 1KB file, the approximate memory that it will cost NN is the
memory to store the INODE-to-BLOCK relation.

As far as the DN is concerned, all it knows is that there is data that
resides on its disk with a specific name (that associates with the Block
name that NN knows). It does not know (or care) about what the block size is
or where else it is replicated etc! Hence for the DN it is just another file
and it consumes how much ever space is required to store this file (1KB in
your case).

So, it does not cost either the NN or DN 64MB to store a 1KB file.

Regards,
John George

On 6/10/11 11:47 AM, "Pedro Costa" <[EMAIL PROTECTED]> wrote:

> So, I'm not getting how a 1KB file can cost a block of 64MB. Can
> anyone explain me?
>
> On Fri, Jun 10, 2011 at 5:13 PM, Philip Zeyliger <[EMAIL PROTECTED]> wrote:
>> On Fri, Jun 10, 2011 at 9:08 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
>>> This means that, when HDFS reads 1KB file from the disk, he will put
>>> the data in blocks of 64MB?
>>
>> No.
>>
>>>
>>> On Fri, Jun 10, 2011 at 5:00 PM, Philip Zeyliger <[EMAIL PROTECTED]>
>>> wrote:
>>>> On Fri, Jun 10, 2011 at 8:42 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
>>>>> But, how can I say that a 1KB file will only use 1KB of disc space, if
>>>>> a block is configured has 64MB? In my view, if a 1KB use a block of
>>>>> 64MB, the file will occupy 64MB in the disc.
>>>>
>>>> A block of HDFS is the unit of distribution and replication, not the
>>>> unit of storage.  HDFS uses the underlying file systems for physical
>>>> storage.
>>>>
>>>> -- Philip
>>>>
>>>>>
>>>>> How can you disassociate a  64MB data block from HDFS of a disk block?
>>>>>
>>>>> On Fri, Jun 10, 2011 at 5:01 PM, Marcos Ortiz <[EMAIL PROTECTED]> wrote:
>>>>>> On 06/10/2011 10:35 AM, Pedro Costa wrote:
>>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
>>>>>> file, this file will ocupy 64MB in the HDFS?
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> HDFS is not very efficient storing small files, because each file is
>>>>>> stored
>>>>>> in a block (of 64 MB in your case), and the block metadata
>>>>>> is held in memory by the NN. But you should know that this 1KB file only
>>>>>> will use 1KB of disc space.
>>>>>>
>>>>>> For small files, you can use Hadoop archives.
>>>>>> Regards
>>>>>>
>>>>>> --
>>>>>> Marcos Luís Ortíz Valmaseda
>>>>>>  Software Engineer (UCI)
>>>>>>  http://marcosluis2186.posterous.com
>>>>>>  http://twitter.com/marcosluis2186
>>>>>>
>>>>>>
>>>>>
>>>>
>>>

Pedro,
You need to distinguish between "HDFS" files and blocks, and "Low-level Disk" files
and blocks.

Large HDFS files are broken into HDFS blocks and stored in multiple Datanodes.
On the Datanodes, each HDFS block is stored as a Low-level Disk file.
So if you have the block size set to 64MB, then a 70MB HDFS file would be split into
a 64MB block and a 6MB final block.  Each of those blocks will become a disk file
on one or more datanodes (depending on replication settings), and will take up
however much disk storage is needed for that disk file.  Of course we could have
just said "save a 64MB chunk for each block", but that would have been wasteful.  
Instead, it just uses as much disk as is actually needed for the amount of data in
that block.

Obviously, only the last block in an HDFS file can be smaller than the block size.

It's worth mentioning that Low-level Disk "blocks" are different.  Because of the way
hard drive hardware works, disk blocks are fixed size, typically either 4KB or 8KB.
It is impossible to allocate less than a full disk block of low-level disk storage.
But this constraint does not apply to HDFS blocks, which are higher-level constructs.

--Matt
On Jun 10, 2011, at 9:13 AM, Philip Zeyliger wrote:

On Fri, Jun 10, 2011 at 9:08 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
> This means that, when HDFS reads 1KB file from the disk, he will put
> the data in blocks of 64MB?

No.

>
> On Fri, Jun 10, 2011 at 5:00 PM, Philip Zeyliger <[EMAIL PROTECTED]> wrote:
>> On Fri, Jun 10, 2011 at 8:42 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
>>> But, how can I say that a 1KB file will only use 1KB of disc space, if
>>> a block is configured has 64MB? In my view, if a 1KB use a block of
>>> 64MB, the file will occupy 64MB in the disc.
>>
>> A block of HDFS is the unit of distribution and replication, not the
>> unit of storage.  HDFS uses the underlying file systems for physical
>> storage.
>>
>> -- Philip
>>
>>>
>>> How can you disassociate a  64MB data block from HDFS of a disk block?
>>>
>>> On Fri, Jun 10, 2011 at 5:01 PM, Marcos Ortiz <[EMAIL PROTECTED]> wrote:
>>>> On 06/10/2011 10:35 AM, Pedro Costa wrote:
>>>>
>>>> Hi,
>>>>
>>>> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
>>>> file, this file will ocupy 64MB in the HDFS?
>>>>
>>>> Thanks,
>>>>
>>>> HDFS is not very efficient storing small files, because each file is stored
>>>> in a block (of 64 MB in your case), and the block metadata
>>>> is held in memory by the NN. But you should know that this 1KB file only
>>>> will use 1KB of disc space.
>>>>
>>>> For small files, you can use Hadoop archives.
>>>> Regards
>>>>
>>>> --
>>>> Marcos Luís Ortíz Valmaseda
>>>>  Software Engineer (UCI)
>>>>  http://marcosluis2186.posterous.com
>>>>  http://twitter.com/marcosluis2186
>>>>
>>>>
>>>
>>
>
>
>
> --
> ---------------------------
> Pedro Sá da Costa
>
> @: [EMAIL PROTECTED]
> @: [EMAIL PROTECTED]
>

It will only take up ~1KB of local datanode disk space (+ metadata
space such as the CRC32 of every 512 bytes, along with replication @
1KB per replicated block, in this case 2KB) but the real cost is a
block entry in the Namenode --- all block data at the namenode lives
in memory, which is a much more scare resource for the cluster in a
relative sense.

On Fri, Jun 10, 2011 at 11:47 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
> So, I'm not getting how a 1KB file can cost a block of 64MB. Can
> anyone explain me?
>
> On Fri, Jun 10, 2011 at 5:13 PM, Philip Zeyliger <[EMAIL PROTECTED]> wrote:
>> On Fri, Jun 10, 2011 at 9:08 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
>>> This means that, when HDFS reads 1KB file from the disk, he will put
>>> the data in blocks of 64MB?
>>
>> No.
>>
>>>
>>> On Fri, Jun 10, 2011 at 5:00 PM, Philip Zeyliger <[EMAIL PROTECTED]> wrote:
>>>> On Fri, Jun 10, 2011 at 8:42 AM, Pedro Costa <[EMAIL PROTECTED]> wrote:
>>>>> But, how can I say that a 1KB file will only use 1KB of disc space, if
>>>>> a block is configured has 64MB? In my view, if a 1KB use a block of
>>>>> 64MB, the file will occupy 64MB in the disc.
>>>>
>>>> A block of HDFS is the unit of distribution and replication, not the
>>>> unit of storage.  HDFS uses the underlying file systems for physical
>>>> storage.
>>>>
>>>> -- Philip
>>>>
>>>>>
>>>>> How can you disassociate a  64MB data block from HDFS of a disk block?
>>>>>
>>>>> On Fri, Jun 10, 2011 at 5:01 PM, Marcos Ortiz <[EMAIL PROTECTED]> wrote:
>>>>>> On 06/10/2011 10:35 AM, Pedro Costa wrote:
>>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> If I define HDFS to use blocks of 64 MB, and I store in HDFS a 1KB
>>>>>> file, this file will ocupy 64MB in the HDFS?
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> HDFS is not very efficient storing small files, because each file is stored
>>>>>> in a block (of 64 MB in your case), and the block metadata
>>>>>> is held in memory by the NN. But you should know that this 1KB file only
>>>>>> will use 1KB of disc space.
>>>>>>
>>>>>> For small files, you can use Hadoop archives.
>>>>>> Regards
>>>>>>
>>>>>> --
>>>>>> Marcos Luís Ortíz Valmaseda
>>>>>>  Software Engineer (UCI)
>>>>>>  http://marcosluis2186.posterous.com
>>>>>>  http://twitter.com/marcosluis2186
>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>

--
Twitter: @jpatanooga
Solution Architect @ Cloudera
hadoop: http://www.cloudera.com
blog: http://jpatterson.floe.tv

FYI, I've added this to the FAQ since it comes up every so often.

Where do I see the FAQ ?

-----Original Message-----
From: Allen Wittenauer [mailto:[EMAIL PROTECTED]]
Sent: Monday, June 13, 2011 10:06 AM
To: [EMAIL PROTECTED]
Subject: Re: Block size in HDFS
FYI, I've added this to the FAQ since it comes up every so often.

On Jun 13, 2011, at 1:37 PM, Jain, Prem wrote:

> Where do I see the FAQ ?
http://wiki.apache.org/hadoop/FAQ