You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@hbase.apache.org by Matt Mankins <ma...@gmail.com> on 2008/02/09 21:23:18 UTC
S3 / HDFS
Hi.
Is it possible to use HBase on a Hadoop system using Amazon's S3
instead of HDFS?
If I'm at EC2, would this be significantly slower than HDFS? Any
ideas what the best practices are?
Thanks,
Matt
Re: S3 / HDFS
Posted by Joost Ouwerkerk <jo...@openplaces.com>.
These guys have a mechanism (using http://fuse.sourceforge.net/) that
allows a local EC2 drive to be backed by S3 storage using a RAID array
-- worth a try if direct S3 is too slow: http://
www.elasticdrive.com/. Haven't tried it yet.
Joost.
On 9-Feb-08, at 4:23 PM, Matt Mankins wrote:
> Excellent. Thanks for the clarification! I'll post any findings.
>
> Matt
>
>
> On Feb 9, 2008, at 4:11 PM, Jim Kellerman wrote:
>
>> I believe that S3 is one of several "file systems" that Hadoop
>> supports. HBase only uses the Hadoop
>> file system abstraction, so it *should* work on any file system
>> that Hadoop can.
>>
>> ---
>> Jim Kellerman, Senior Engineer; Powerset
>>
Re: S3 / HDFS
Posted by Matt Mankins <ma...@gmail.com>.
Excellent. Thanks for the clarification! I'll post any findings.
Matt
On Feb 9, 2008, at 4:11 PM, Jim Kellerman wrote:
> I believe that S3 is one of several "file systems" that Hadoop
> supports. HBase only uses the Hadoop
> file system abstraction, so it *should* work on any file system
> that Hadoop can.
>
> ---
> Jim Kellerman, Senior Engineer; Powerset
>
RE: S3 / HDFS
Posted by Jim Kellerman <ji...@powerset.com>.
I believe that S3 is one of several "file systems" that Hadoop supports. HBase only uses the Hadoop
file system abstraction, so it *should* work on any file system that Hadoop can.
---
Jim Kellerman, Senior Engineer; Powerset
> -----Original Message-----
> From: Matt Mankins [mailto:mankins@gmail.com]
> Sent: Saturday, February 09, 2008 1:05 PM
> To: hbase-user@hadoop.apache.org
> Subject: Re: S3 / HDFS
>
> http://wiki.apache.org/hadoop/AmazonS3 seems to imply that
> Hadoop can use HDFS *or* S3.
>
> If that's the case, would HBase need to be modified to use S3?
>
> Perhaps I'm just confused as to which way the layers stack:
>
> HBase
> -----------
> Hadoop
> -----------
> S3 OR HDFS OR Local File System (?)
>
>
> Or is HDFS in the middle:
>
>
> HBase
> ---------
> HDFS
> --------
> S3 / Local Disk
>
>
> Or perhaps something else... ?
>
> If the tablet sizes are large, I could see how each S3 fetch
> would slow things down.
>
> I'll have to just try it and let you know.
>
> Matt
>
> On Feb 9, 2008, at 3:47 PM, Jim Kellerman wrote:
>
> > It should work because HBase only relies on HDFS, not the
> underlying
> > storage media.
> > As to speed, I can't say. But if S3 is x% slower than HDFS on local
> > disk, I would expect that HBase would be at least that much slower.
> >
> > It would be an interesting experiment. If someone does try
> it, please
> > post your results.
> >
> > ---
> > Jim Kellerman, Senior Engineer; Powerset
> >
> >
> >> -----Original Message-----
> >> From: Joost Ouwerkerk [mailto:joosto@openplaces.com]
> >> Sent: Saturday, February 09, 2008 12:42 PM
> >> To: hbase-user@hadoop.apache.org
> >> Subject: Re: S3 / HDFS
> >>
> >> Great question that I'm also wondering about. How to handle
> >> persistence when using Hbase on EC2, since EC2 instance
> disks are not
> >> reliable. I'm guessing that a straight S3 file system
> would be way
> >> too slow unless it has some kind of RAID-like mechanism
> whereby local
> >> disk access is backed by S3 persistence.
> >>
> >> On 9-Feb-08, at 3:23 PM, Matt Mankins wrote:
> >>
> >>> Hi.
> >>>
> >>> Is it possible to use HBase on a Hadoop system using Amazon's S3
> >>> instead of HDFS?
> >>>
> >>> If I'm at EC2, would this be significantly slower than HDFS? Any
> >>> ideas what the best practices are?
> >>>
> >>> Thanks,
> >>>
> >>> Matt
> >>
>
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.516 / Virus Database: 269.20.0/1268 - Release
> Date: 2/9/2008 11:54 AM
>
>
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.516 / Virus Database: 269.20.0/1268 - Release Date: 2/9/2008 11:54 AM
Re: S3 / HDFS
Posted by Matt Mankins <ma...@gmail.com>.
http://wiki.apache.org/hadoop/AmazonS3 seems to imply that Hadoop can
use HDFS *or* S3.
If that's the case, would HBase need to be modified to use S3?
Perhaps I'm just confused as to which way the layers stack:
HBase
-----------
Hadoop
-----------
S3 OR HDFS OR Local File System (?)
Or is HDFS in the middle:
HBase
---------
HDFS
--------
S3 / Local Disk
Or perhaps something else... ?
If the tablet sizes are large, I could see how each S3 fetch would
slow things down.
I'll have to just try it and let you know.
Matt
On Feb 9, 2008, at 3:47 PM, Jim Kellerman wrote:
> It should work because HBase only relies on HDFS, not the
> underlying storage media.
> As to speed, I can't say. But if S3 is x% slower than HDFS on local
> disk, I would
> expect that HBase would be at least that much slower.
>
> It would be an interesting experiment. If someone does try it,
> please post your
> results.
>
> ---
> Jim Kellerman, Senior Engineer; Powerset
>
>
>> -----Original Message-----
>> From: Joost Ouwerkerk [mailto:joosto@openplaces.com]
>> Sent: Saturday, February 09, 2008 12:42 PM
>> To: hbase-user@hadoop.apache.org
>> Subject: Re: S3 / HDFS
>>
>> Great question that I'm also wondering about. How to handle
>> persistence when using Hbase on EC2, since EC2 instance disks
>> are not reliable. I'm guessing that a straight S3 file
>> system would be way too slow unless it has some kind of
>> RAID-like mechanism whereby local disk access is backed by S3
>> persistence.
>>
>> On 9-Feb-08, at 3:23 PM, Matt Mankins wrote:
>>
>>> Hi.
>>>
>>> Is it possible to use HBase on a Hadoop system using Amazon's S3
>>> instead of HDFS?
>>>
>>> If I'm at EC2, would this be significantly slower than HDFS? Any
>>> ideas what the best practices are?
>>>
>>> Thanks,
>>>
>>> Matt
>>
RE: S3 / HDFS
Posted by Jim Kellerman <ji...@powerset.com>.
It should work because HBase only relies on HDFS, not the underlying storage media.
As to speed, I can't say. But if S3 is x% slower than HDFS on local disk, I would
expect that HBase would be at least that much slower.
It would be an interesting experiment. If someone does try it, please post your
results.
---
Jim Kellerman, Senior Engineer; Powerset
> -----Original Message-----
> From: Joost Ouwerkerk [mailto:joosto@openplaces.com]
> Sent: Saturday, February 09, 2008 12:42 PM
> To: hbase-user@hadoop.apache.org
> Subject: Re: S3 / HDFS
>
> Great question that I'm also wondering about. How to handle
> persistence when using Hbase on EC2, since EC2 instance disks
> are not reliable. I'm guessing that a straight S3 file
> system would be way too slow unless it has some kind of
> RAID-like mechanism whereby local disk access is backed by S3
> persistence.
>
> On 9-Feb-08, at 3:23 PM, Matt Mankins wrote:
>
> > Hi.
> >
> > Is it possible to use HBase on a Hadoop system using Amazon's S3
> > instead of HDFS?
> >
> > If I'm at EC2, would this be significantly slower than HDFS? Any
> > ideas what the best practices are?
> >
> > Thanks,
> >
> > Matt
>
>
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.516 / Virus Database: 269.20.0/1268 - Release
> Date: 2/9/2008 11:54 AM
>
>
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.516 / Virus Database: 269.20.0/1268 - Release Date: 2/9/2008 11:54 AM
Re: S3 / HDFS
Posted by Joost Ouwerkerk <jo...@openplaces.com>.
Great question that I'm also wondering about. How to handle
persistence when using Hbase on EC2, since EC2 instance disks are not
reliable. I'm guessing that a straight S3 file system would be way
too slow unless it has some kind of RAID-like mechanism whereby local
disk access is backed by S3 persistence.
On 9-Feb-08, at 3:23 PM, Matt Mankins wrote:
> Hi.
>
> Is it possible to use HBase on a Hadoop system using Amazon's S3
> instead of HDFS?
>
> If I'm at EC2, would this be significantly slower than HDFS? Any
> ideas what the best practices are?
>
> Thanks,
>
> Matt