Google has released their specifications on their
protocol buffers.
in the blog, someone asks
"Marko Loparic said...
Interesting. I wonder how it compares to HDF5?
Hierarchical Data Format - Wikipedia
Is it simpler? Different goals?"
Anybody know the answer?
Matthew Dougherty
713-433-3849
National Center for Macromolecular Imaging
Baylor College of Medicine/Houston Texas USA
A Wednesday 09 July 2008, Dougherty, Matthew T. escrigué:
Google has released their specifications on their
protocol buffers.http://google-opensource.blogspot.com/2008/07/protocol-buffers-google
s-data.html
Interesting...
in the blog, someone asks
"Marko Loparic said...
Interesting. I wonder how it compares to HDF5?
http://en.wikipedia.org/wiki/Hierarchical_Data_Format
Is it simpler? Different goals?"
Anybody know the answer?
Well, after having a quick look at the Protocol Buffer documents, I'd
say that they are more a way to serialize data for RPC purposes than
anything else. In that sense, it is more comparable to ASN.1 DER or
OPeNDAP (which should be more familiar to the readers of this list)
than to HDF5 itself.
Incidentally, a technology from Google which is somewhat comparable to
HDF5 is Bigtable (see a description in [1]). However, Bigtable seems
to me much more tailored to the particular needs of saving web contents
(no wonder here: it has come out from Google) than HDF5, which is more
general IMO.
[1] http://labs.google.com/papers/bigtable.html
Cheers,
--
Francesc Alted
Freelance developer
Tel +34-964-282-249
----------------------------------------------------------------------
This mailing list is for HDF software users discussion.
To subscribe to this list, send a message to hdf-forum-subscribe@hdfgroup.org.
To unsubscribe, send a message to hdf-forum-unsubscribe@hdfgroup.org.
The HDF Group has posted an initial comparison of Google Protocol Buffers and HDF5 on our wiki at http://wiki.hdfgroup.org/Google+Protocol+Buffers+and+HDF5
If you would like to add comments, the invite key for write/comment privileges is hdf5wiki.
On Jul 9, 2008, at 12:13 AM, Dougherty, Matthew T. wrote:
Google has released their specifications on their
protocol buffers.http://google-opensource.blogspot.com/2008/07/protocol-buffers-googles-data.html
in the blog, someone asks
"Marko Loparic said...
Interesting. I wonder how it compares to HDF5?
http://en.wikipedia.org/wiki/Hierarchical_Data_Format
Is it simpler? Different goals?"
Anybody know the answer?
Matthew Dougherty
713-433-3849
National Center for Macromolecular Imaging
Baylor College of Medicine/Houston Texas USA=========================================================================
------------------------------------------------------------
Ruth Aydt
The HDF Group
1901 South First Street, Suite C-2
Champaign, IL 61820
aydt@hdfgroup.org
(217)265-7837 (office) (217)333-9049 (fax)
------------------------------------------------------------
I agree it is more serial message passing using a quasi XML approach, than moving significantly large datasets.
It might be good for somebody to respond authoritatively into that blog.
Because it is google, if nobody responds to the question, a large body of people may think HDF is irrelevant or obsolete.
It would be a good opportunity to talk up HDF, explain the appropriate use of HDF and how the two methods could be used in tandem.
Matthew Dougherty
713-433-3849
National Center for Macromolecular Imaging
Baylor College of Medicine/Houston Texas USA
-----Original Message-----
From: Francesc Alted [mailto:faltet@pytables.com]
Sent: Wed 7/9/2008 2:29 AM
To: hdf-forum@hdfgroup.org
Subject: Re: [hdf-forum] GOOGLE/scalable data exchange
A Wednesday 09 July 2008, Dougherty, Matthew T. escrigué:
Google has released their specifications on their
protocol buffers.http://google-opensource.blogspot.com/2008/07/protocol-buffers-google
s-data.html
Interesting...
in the blog, someone asks
"Marko Loparic said...
Interesting. I wonder how it compares to HDF5?
http://en.wikipedia.org/wiki/Hierarchical_Data_Format
Is it simpler? Different goals?"
Anybody know the answer?
Well, after having a quick look at the Protocol Buffer documents, I'd
say that they are more a way to serialize data for RPC purposes than
anything else. In that sense, it is more comparable to ASN.1 DER or
OPeNDAP (which should be more familiar to the readers of this list)
than to HDF5 itself.
Incidentally, a technology from Google which is somewhat comparable to
HDF5 is Bigtable (see a description in [1]). However, Bigtable seems
to me much more tailored to the particular needs of saving web contents
(no wonder here: it has come out from Google) than HDF5, which is more
general IMO.
[1] http://labs.google.com/papers/bigtable.html
Cheers,
--
Francesc Alted
Freelance developer
Tel +34-964-282-249
----------------------------------------------------------------------
This mailing list is for HDF software users discussion.
To subscribe to this list, send a message to hdf-forum-subscribe@hdfgroup.org.
To unsubscribe, send a message to hdf-forum-unsubscribe@hdfgroup.org.
Matthew,
Thank you for bringing this to our attention. We (THG) will respond.
Elena
On Jul 9, 2008, at 11:04 AM, Dougherty, Matthew T. wrote:
I agree it is more serial message passing using a quasi XML approach, than moving significantly large datasets.
It might be good for somebody to respond authoritatively into that blog.
Because it is google, if nobody responds to the question, a large body of people may think HDF is irrelevant or obsolete.
It would be a good opportunity to talk up HDF, explain the appropriate use of HDF and how the two methods could be used in tandem.
Matthew Dougherty
713-433-3849
National Center for Macromolecular Imaging
Baylor College of Medicine/Houston Texas USA=========================================================================
-----Original Message-----
From: Francesc Alted [mailto:faltet@pytables.com]
Sent: Wed 7/9/2008 2:29 AM
To: hdf-forum@hdfgroup.org
Subject: Re: [hdf-forum] GOOGLE/scalable data exchangeA Wednesday 09 July 2008, Dougherty, Matthew T. escrigué:
> Google has released their specifications on their
> protocol buffers.
>
> http://google-opensource.blogspot.com/2008/07/protocol-buffers-google
>s-data.htmlInteresting...
>
> in the blog, someone asks
>
> "Marko Loparic said...
>
> Interesting. I wonder how it compares to HDF5?
>
> http://en.wikipedia.org/wiki/Hierarchical_Data_Format
>
> Is it simpler? Different goals?"
>
> Anybody know the answer?Well, after having a quick look at the Protocol Buffer documents, I'd
say that they are more a way to serialize data for RPC purposes than
anything else. In that sense, it is more comparable to ASN.1 DER or
OPeNDAP (which should be more familiar to the readers of this list)
than to HDF5 itself.Incidentally, a technology from Google which is somewhat comparable to
HDF5 is Bigtable (see a description in [1]). However, Bigtable seems
to me much more tailored to the particular needs of saving web contents
(no wonder here: it has come out from Google) than HDF5, which is more
general IMO.[1] http://labs.google.com/papers/bigtable.html
Cheers,
--
Francesc Alted
Freelance developer
Tel +34-964-282-249----------------------------------------------------------------------
This mailing list is for HDF software users discussion.
To subscribe to this list, send a message to hdf-forum-subscribe@hdfgroup.org.
To unsubscribe, send a message to hdf-forum-unsubscribe@hdfgroup.org.
