You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-user@hadoop.apache.org by Kishore <al...@gmail.com> on 2013/12/15 14:21:17 UTC

External db

Hi experts,

What exactly needed the nosql database like Cassandra, hbase and mongodb when we have hdfs, ofcourse    
It supports oltp, are we storing the results from hdfs and reanalysing in nosql?

Sent from my iPhone

Re: External db

Posted by Ted Yu <yu...@gmail.com>.
Kishore:
Some NoSQL from your initial post, such as mongodb, is not built on top of
hdfs.
See:
http://www.ikanow.com/blog/02/15/how-well-does-mongodb-integrate-with-hadoop/

Cheers


On Sun, Dec 15, 2013 at 5:42 AM, Peter Lin <wo...@gmail.com> wrote:

>
> you're question doesn't make any sense. Did you mean to ask
>
> What is the benefit of NoSql databases like Cassandra, HBase and MongoDB?
> Isn't that like asking "why do we need SqlServer and Oracle when we have
> NTFS?"
>
> The first thing is Nosql databases are generally grouped by 3 categories:
> document (aka couchdb and mongodb), column (cassandra and hbase) and graph
> (neo4j).
>
> There's a lot of talks and blogs out there, I would suggest spend a week
> and read. Google is your friend.
>
>
> On Sun, Dec 15, 2013 at 8:21 AM, Kishore <al...@gmail.com>wrote:
>
>> Hi experts,
>>
>> What exactly needed the nosql database like Cassandra, hbase and mongodb
>> when we have hdfs, ofcourse
>> It supports oltp, are we storing the results from hdfs and reanalysing in
>> nosql?
>>
>> Sent from my iPhone
>
>
>

Re: External db

Posted by Ted Yu <yu...@gmail.com>.
Kishore:
Some NoSQL from your initial post, such as mongodb, is not built on top of
hdfs.
See:
http://www.ikanow.com/blog/02/15/how-well-does-mongodb-integrate-with-hadoop/

Cheers


On Sun, Dec 15, 2013 at 5:42 AM, Peter Lin <wo...@gmail.com> wrote:

>
> you're question doesn't make any sense. Did you mean to ask
>
> What is the benefit of NoSql databases like Cassandra, HBase and MongoDB?
> Isn't that like asking "why do we need SqlServer and Oracle when we have
> NTFS?"
>
> The first thing is Nosql databases are generally grouped by 3 categories:
> document (aka couchdb and mongodb), column (cassandra and hbase) and graph
> (neo4j).
>
> There's a lot of talks and blogs out there, I would suggest spend a week
> and read. Google is your friend.
>
>
> On Sun, Dec 15, 2013 at 8:21 AM, Kishore <al...@gmail.com>wrote:
>
>> Hi experts,
>>
>> What exactly needed the nosql database like Cassandra, hbase and mongodb
>> when we have hdfs, ofcourse
>> It supports oltp, are we storing the results from hdfs and reanalysing in
>> nosql?
>>
>> Sent from my iPhone
>
>
>

Re: External db

Posted by Ted Yu <yu...@gmail.com>.
Kishore:
Some NoSQL from your initial post, such as mongodb, is not built on top of
hdfs.
See:
http://www.ikanow.com/blog/02/15/how-well-does-mongodb-integrate-with-hadoop/

Cheers


On Sun, Dec 15, 2013 at 5:42 AM, Peter Lin <wo...@gmail.com> wrote:

>
> you're question doesn't make any sense. Did you mean to ask
>
> What is the benefit of NoSql databases like Cassandra, HBase and MongoDB?
> Isn't that like asking "why do we need SqlServer and Oracle when we have
> NTFS?"
>
> The first thing is Nosql databases are generally grouped by 3 categories:
> document (aka couchdb and mongodb), column (cassandra and hbase) and graph
> (neo4j).
>
> There's a lot of talks and blogs out there, I would suggest spend a week
> and read. Google is your friend.
>
>
> On Sun, Dec 15, 2013 at 8:21 AM, Kishore <al...@gmail.com>wrote:
>
>> Hi experts,
>>
>> What exactly needed the nosql database like Cassandra, hbase and mongodb
>> when we have hdfs, ofcourse
>> It supports oltp, are we storing the results from hdfs and reanalysing in
>> nosql?
>>
>> Sent from my iPhone
>
>
>

Re: External db

Posted by Ted Yu <yu...@gmail.com>.
Kishore:
Some NoSQL from your initial post, such as mongodb, is not built on top of
hdfs.
See:
http://www.ikanow.com/blog/02/15/how-well-does-mongodb-integrate-with-hadoop/

Cheers


On Sun, Dec 15, 2013 at 5:42 AM, Peter Lin <wo...@gmail.com> wrote:

>
> you're question doesn't make any sense. Did you mean to ask
>
> What is the benefit of NoSql databases like Cassandra, HBase and MongoDB?
> Isn't that like asking "why do we need SqlServer and Oracle when we have
> NTFS?"
>
> The first thing is Nosql databases are generally grouped by 3 categories:
> document (aka couchdb and mongodb), column (cassandra and hbase) and graph
> (neo4j).
>
> There's a lot of talks and blogs out there, I would suggest spend a week
> and read. Google is your friend.
>
>
> On Sun, Dec 15, 2013 at 8:21 AM, Kishore <al...@gmail.com>wrote:
>
>> Hi experts,
>>
>> What exactly needed the nosql database like Cassandra, hbase and mongodb
>> when we have hdfs, ofcourse
>> It supports oltp, are we storing the results from hdfs and reanalysing in
>> nosql?
>>
>> Sent from my iPhone
>
>
>

Re: External db

Posted by Peter Lin <wo...@gmail.com>.
you're question doesn't make any sense. Did you mean to ask

What is the benefit of NoSql databases like Cassandra, HBase and MongoDB?
Isn't that like asking "why do we need SqlServer and Oracle when we have
NTFS?"

The first thing is Nosql databases are generally grouped by 3 categories:
document (aka couchdb and mongodb), column (cassandra and hbase) and graph
(neo4j).

There's a lot of talks and blogs out there, I would suggest spend a week
and read. Google is your friend.


On Sun, Dec 15, 2013 at 8:21 AM, Kishore <al...@gmail.com> wrote:

> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb
> when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in
> nosql?
>
> Sent from my iPhone

Re: External db

Posted by Shekhar Sharma <sh...@gmail.com>.
HDFS is batch mode suitable for OLAP not for OLTP..

OLTP requires the data need to be updated, deleted , basically CRUD
operation..But HDFS doesnt support random writes..so not suitable for
OLTP...And thats where the nosql databases comes in
Regards,
Som Shekhar Sharma
+91-8197243810


On Sun, Dec 15, 2013 at 6:51 PM, Kishore <al...@gmail.com> wrote:
> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in nosql?
>
> Sent from my iPhone

Re: External db

Posted by Chris Mawata <ch...@gmail.com>.
Hi Kishore,
   you are comparing apples and oranges. HDFS is a file system (you read 
and write files to it). The NoSQL datastores are more like
a database. You can query and depending on the type of NoSQL database 
the querying can be more or less sophisticated e.g. "Give
me a document that contains a sub-document xyz".  You could write a 
MapReduce program to do the querying (that is how Hive works
for example) but HDFS itself just stores files. Some NoSQL datastores 
like HBase and Accumulo use HDFS as the underlying
file system just as a MySQL database might be using a WIndows NTFS file 
system or Linux ext4 system to store the data.
Chris


On 12/15/2013 8:21 AM, Kishore wrote:
> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in nosql?
>
> Sent from my iPhone.
>


Re: External db

Posted by Chris Mawata <ch...@gmail.com>.
Hi Kishore,
   you are comparing apples and oranges. HDFS is a file system (you read 
and write files to it). The NoSQL datastores are more like
a database. You can query and depending on the type of NoSQL database 
the querying can be more or less sophisticated e.g. "Give
me a document that contains a sub-document xyz".  You could write a 
MapReduce program to do the querying (that is how Hive works
for example) but HDFS itself just stores files. Some NoSQL datastores 
like HBase and Accumulo use HDFS as the underlying
file system just as a MySQL database might be using a WIndows NTFS file 
system or Linux ext4 system to store the data.
Chris


On 12/15/2013 8:21 AM, Kishore wrote:
> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in nosql?
>
> Sent from my iPhone.
>


Re: External db

Posted by Shekhar Sharma <sh...@gmail.com>.
HDFS is batch mode suitable for OLAP not for OLTP..

OLTP requires the data need to be updated, deleted , basically CRUD
operation..But HDFS doesnt support random writes..so not suitable for
OLTP...And thats where the nosql databases comes in
Regards,
Som Shekhar Sharma
+91-8197243810


On Sun, Dec 15, 2013 at 6:51 PM, Kishore <al...@gmail.com> wrote:
> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in nosql?
>
> Sent from my iPhone

Re: External db

Posted by Peter Lin <wo...@gmail.com>.
you're question doesn't make any sense. Did you mean to ask

What is the benefit of NoSql databases like Cassandra, HBase and MongoDB?
Isn't that like asking "why do we need SqlServer and Oracle when we have
NTFS?"

The first thing is Nosql databases are generally grouped by 3 categories:
document (aka couchdb and mongodb), column (cassandra and hbase) and graph
(neo4j).

There's a lot of talks and blogs out there, I would suggest spend a week
and read. Google is your friend.


On Sun, Dec 15, 2013 at 8:21 AM, Kishore <al...@gmail.com> wrote:

> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb
> when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in
> nosql?
>
> Sent from my iPhone

Re: External db

Posted by Peter Lin <wo...@gmail.com>.
you're question doesn't make any sense. Did you mean to ask

What is the benefit of NoSql databases like Cassandra, HBase and MongoDB?
Isn't that like asking "why do we need SqlServer and Oracle when we have
NTFS?"

The first thing is Nosql databases are generally grouped by 3 categories:
document (aka couchdb and mongodb), column (cassandra and hbase) and graph
(neo4j).

There's a lot of talks and blogs out there, I would suggest spend a week
and read. Google is your friend.


On Sun, Dec 15, 2013 at 8:21 AM, Kishore <al...@gmail.com> wrote:

> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb
> when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in
> nosql?
>
> Sent from my iPhone

Re: External db

Posted by Shekhar Sharma <sh...@gmail.com>.
HDFS is batch mode suitable for OLAP not for OLTP..

OLTP requires the data need to be updated, deleted , basically CRUD
operation..But HDFS doesnt support random writes..so not suitable for
OLTP...And thats where the nosql databases comes in
Regards,
Som Shekhar Sharma
+91-8197243810


On Sun, Dec 15, 2013 at 6:51 PM, Kishore <al...@gmail.com> wrote:
> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in nosql?
>
> Sent from my iPhone

Re: External db

Posted by Shekhar Sharma <sh...@gmail.com>.
HDFS is batch mode suitable for OLAP not for OLTP..

OLTP requires the data need to be updated, deleted , basically CRUD
operation..But HDFS doesnt support random writes..so not suitable for
OLTP...And thats where the nosql databases comes in
Regards,
Som Shekhar Sharma
+91-8197243810


On Sun, Dec 15, 2013 at 6:51 PM, Kishore <al...@gmail.com> wrote:
> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in nosql?
>
> Sent from my iPhone

Re: External db

Posted by Peter Lin <wo...@gmail.com>.
you're question doesn't make any sense. Did you mean to ask

What is the benefit of NoSql databases like Cassandra, HBase and MongoDB?
Isn't that like asking "why do we need SqlServer and Oracle when we have
NTFS?"

The first thing is Nosql databases are generally grouped by 3 categories:
document (aka couchdb and mongodb), column (cassandra and hbase) and graph
(neo4j).

There's a lot of talks and blogs out there, I would suggest spend a week
and read. Google is your friend.


On Sun, Dec 15, 2013 at 8:21 AM, Kishore <al...@gmail.com> wrote:

> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb
> when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in
> nosql?
>
> Sent from my iPhone

Re: External db

Posted by Chris Mawata <ch...@gmail.com>.
Hi Kishore,
   you are comparing apples and oranges. HDFS is a file system (you read 
and write files to it). The NoSQL datastores are more like
a database. You can query and depending on the type of NoSQL database 
the querying can be more or less sophisticated e.g. "Give
me a document that contains a sub-document xyz".  You could write a 
MapReduce program to do the querying (that is how Hive works
for example) but HDFS itself just stores files. Some NoSQL datastores 
like HBase and Accumulo use HDFS as the underlying
file system just as a MySQL database might be using a WIndows NTFS file 
system or Linux ext4 system to store the data.
Chris


On 12/15/2013 8:21 AM, Kishore wrote:
> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in nosql?
>
> Sent from my iPhone.
>


Re: External db

Posted by Chris Mawata <ch...@gmail.com>.
Hi Kishore,
   you are comparing apples and oranges. HDFS is a file system (you read 
and write files to it). The NoSQL datastores are more like
a database. You can query and depending on the type of NoSQL database 
the querying can be more or less sophisticated e.g. "Give
me a document that contains a sub-document xyz".  You could write a 
MapReduce program to do the querying (that is how Hive works
for example) but HDFS itself just stores files. Some NoSQL datastores 
like HBase and Accumulo use HDFS as the underlying
file system just as a MySQL database might be using a WIndows NTFS file 
system or Linux ext4 system to store the data.
Chris


On 12/15/2013 8:21 AM, Kishore wrote:
> Hi experts,
>
> What exactly needed the nosql database like Cassandra, hbase and mongodb when we have hdfs, ofcourse
> It supports oltp, are we storing the results from hdfs and reanalysing in nosql?
>
> Sent from my iPhone.
>