You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@hbase.apache.org by Rajeshkumar J <ra...@gmail.com> on 2015/12/18 07:17:31 UTC
Type of Scan to be used for real time analysis
Hi,
My hbase table holds 10 million rows and I need to query it and I want
hbase to return the query within one or two seconds. Help me to choose
which type of scan do I have to use for this - range scan or rowfilter scan
Thanks
Re: Type of Scan to be used for real time analysis
Posted by Rajeshkumar J <ra...@gmail.com>.
Hi,
On Fri, Dec 18, 2015 at 7:30 PM, anil gupta <an...@gmail.com> wrote:
> If you know exact rowkey of row that you need to fetch then you just need
> to use GET. If you know just the prefix of rowkey, then you can use range
> scans in HBase. Does the above 2 scenario's cover your use case?
>
> On Fri, Dec 18, 2015 at 4:29 AM, Rajeshkumar J <
> rajeshkumarit8292@gmail.com>
> wrote:
>
> > Hi Anil,
> >
> > I have about 10 million rows with each rows having more than 10k
> > columns. I need to query this table based on row key and which will be
> the
> > apt query process for this
> >
> > Thanks
> >
> > On Fri, Dec 18, 2015 at 5:43 PM, anil gupta <an...@gmail.com>
> wrote:
> >
> > > Hi RajeshKumar,
> > >
> > > IMO, type of scan is not decided on the basis of response time. Its
> > decided
> > > on the basis of your query logic and data model.
> > > Also, Response time cannot be directly correlated to any filter or
> scan.
> > > Response time is more about how much data needs to read, cpu, network
> IO,
> > > etc to suffice requirement of your query.
> > > So, you will need to look at your data model and pick the best query.
> > >
> > > HTH,
> > > Anil
> > >
> > > On Thu, Dec 17, 2015 at 10:17 PM, Rajeshkumar J <
> > > rajeshkumarit8292@gmail.com
> > > > wrote:
> > >
> > > > Hi,
> > > >
> > > > My hbase table holds 10 million rows and I need to query it and I
> > want
> > > > hbase to return the query within one or two seconds. Help me to
> choose
> > > > which type of scan do I have to use for this - range scan or
> rowfilter
> > > scan
> > > >
> > > > Thanks
> > > >
> > >
> > >
> > >
> > > --
> > > Thanks & Regards,
> > > Anil Gupta
> > >
> >
>
>
>
> --
> Thanks & Regards,
> Anil Gupta
>
Re: Type of Scan to be used for real time analysis
Posted by Rajeshkumar J <ra...@gmail.com>.
Hi,
yes I am currently using range scan but it takes more than 10 seconds
based on the number of columns. But I want to bring this down to a second.
Is there any other which I can do to achieve this
Thanks
On Fri, Dec 18, 2015 at 7:30 PM, anil gupta <an...@gmail.com> wrote:
> If you know exact rowkey of row that you need to fetch then you just need
> to use GET. If you know just the prefix of rowkey, then you can use range
> scans in HBase. Does the above 2 scenario's cover your use case?
>
> On Fri, Dec 18, 2015 at 4:29 AM, Rajeshkumar J <
> rajeshkumarit8292@gmail.com>
> wrote:
>
> > Hi Anil,
> >
> > I have about 10 million rows with each rows having more than 10k
> > columns. I need to query this table based on row key and which will be
> the
> > apt query process for this
> >
> > Thanks
> >
> > On Fri, Dec 18, 2015 at 5:43 PM, anil gupta <an...@gmail.com>
> wrote:
> >
> > > Hi RajeshKumar,
> > >
> > > IMO, type of scan is not decided on the basis of response time. Its
> > decided
> > > on the basis of your query logic and data model.
> > > Also, Response time cannot be directly correlated to any filter or
> scan.
> > > Response time is more about how much data needs to read, cpu, network
> IO,
> > > etc to suffice requirement of your query.
> > > So, you will need to look at your data model and pick the best query.
> > >
> > > HTH,
> > > Anil
> > >
> > > On Thu, Dec 17, 2015 at 10:17 PM, Rajeshkumar J <
> > > rajeshkumarit8292@gmail.com
> > > > wrote:
> > >
> > > > Hi,
> > > >
> > > > My hbase table holds 10 million rows and I need to query it and I
> > want
> > > > hbase to return the query within one or two seconds. Help me to
> choose
> > > > which type of scan do I have to use for this - range scan or
> rowfilter
> > > scan
> > > >
> > > > Thanks
> > > >
> > >
> > >
> > >
> > > --
> > > Thanks & Regards,
> > > Anil Gupta
> > >
> >
>
>
>
> --
> Thanks & Regards,
> Anil Gupta
>
Re: Type of Scan to be used for real time analysis
Posted by anil gupta <an...@gmail.com>.
If you know exact rowkey of row that you need to fetch then you just need
to use GET. If you know just the prefix of rowkey, then you can use range
scans in HBase. Does the above 2 scenario's cover your use case?
On Fri, Dec 18, 2015 at 4:29 AM, Rajeshkumar J <ra...@gmail.com>
wrote:
> Hi Anil,
>
> I have about 10 million rows with each rows having more than 10k
> columns. I need to query this table based on row key and which will be the
> apt query process for this
>
> Thanks
>
> On Fri, Dec 18, 2015 at 5:43 PM, anil gupta <an...@gmail.com> wrote:
>
> > Hi RajeshKumar,
> >
> > IMO, type of scan is not decided on the basis of response time. Its
> decided
> > on the basis of your query logic and data model.
> > Also, Response time cannot be directly correlated to any filter or scan.
> > Response time is more about how much data needs to read, cpu, network IO,
> > etc to suffice requirement of your query.
> > So, you will need to look at your data model and pick the best query.
> >
> > HTH,
> > Anil
> >
> > On Thu, Dec 17, 2015 at 10:17 PM, Rajeshkumar J <
> > rajeshkumarit8292@gmail.com
> > > wrote:
> >
> > > Hi,
> > >
> > > My hbase table holds 10 million rows and I need to query it and I
> want
> > > hbase to return the query within one or two seconds. Help me to choose
> > > which type of scan do I have to use for this - range scan or rowfilter
> > scan
> > >
> > > Thanks
> > >
> >
> >
> >
> > --
> > Thanks & Regards,
> > Anil Gupta
> >
>
--
Thanks & Regards,
Anil Gupta
Re: Type of Scan to be used for real time analysis
Posted by Rajeshkumar J <ra...@gmail.com>.
Hi Anil,
I have about 10 million rows with each rows having more than 10k
columns. I need to query this table based on row key and which will be the
apt query process for this
Thanks
On Fri, Dec 18, 2015 at 5:43 PM, anil gupta <an...@gmail.com> wrote:
> Hi RajeshKumar,
>
> IMO, type of scan is not decided on the basis of response time. Its decided
> on the basis of your query logic and data model.
> Also, Response time cannot be directly correlated to any filter or scan.
> Response time is more about how much data needs to read, cpu, network IO,
> etc to suffice requirement of your query.
> So, you will need to look at your data model and pick the best query.
>
> HTH,
> Anil
>
> On Thu, Dec 17, 2015 at 10:17 PM, Rajeshkumar J <
> rajeshkumarit8292@gmail.com
> > wrote:
>
> > Hi,
> >
> > My hbase table holds 10 million rows and I need to query it and I want
> > hbase to return the query within one or two seconds. Help me to choose
> > which type of scan do I have to use for this - range scan or rowfilter
> scan
> >
> > Thanks
> >
>
>
>
> --
> Thanks & Regards,
> Anil Gupta
>
Re: Type of Scan to be used for real time analysis
Posted by anil gupta <an...@gmail.com>.
Hi RajeshKumar,
IMO, type of scan is not decided on the basis of response time. Its decided
on the basis of your query logic and data model.
Also, Response time cannot be directly correlated to any filter or scan.
Response time is more about how much data needs to read, cpu, network IO,
etc to suffice requirement of your query.
So, you will need to look at your data model and pick the best query.
HTH,
Anil
On Thu, Dec 17, 2015 at 10:17 PM, Rajeshkumar J <rajeshkumarit8292@gmail.com
> wrote:
> Hi,
>
> My hbase table holds 10 million rows and I need to query it and I want
> hbase to return the query within one or two seconds. Help me to choose
> which type of scan do I have to use for this - range scan or rowfilter scan
>
> Thanks
>
--
Thanks & Regards,
Anil Gupta