Yes you could store multiple delimiter separated rowkeys. Another  
option would be to use the column qualifiers to store the rowkeys for  
the EmployeeTable and just store a dummy value. For example, the  
EmployeeName table could contain the column family  
"EmployeeTableKey" (in this example I am assuming the rowkey for the  
EmployeeTable is an integer):

Rowkey           ColumnFamily:Qualifier           Value

Sachin             EmployeeTableKey:240               0
Sachin             EmployeeTableKey:153               0
Sachin             EmployeeTableKey:194               0
Rahul               EmployeeTableKey:250               0
Rahul               EmployeeTableKey:1                   0


-Sanjit

On Jun 3, 2009, at 11:27 PM, Sachin wrote:

>
> Thank you very much Sanjit.
>
> I got the answer of my first point. But the answer of the 2nd point is
> still unclear to me.
> Here I had asked to search on employeename instead of ID.
>
> Name can repeat in many rows. So how can we create the name as rowkey?
> Do we need to take the distinct of the employee name and insert in
> this EmployeeName table? In the EmployeeName table, the value should
> be the rowkey of employee table (means EmployeeId). There can be many
> employees with different IDs and with same names. So in the value, do
> we need to store the EmployeeId in comma seperated or what else?
>
> Regards,
> Sachin
>
>
>
> On Jun 3, 9:50 pm, Sanjit Jhala <[email protected]> wrote:
>> Responses inline below.
>>
>> -Sanjit
>> On Jun 3, 2009, at 6:10 AM, Sachin wrote:
>>
>>
>>
>>
>>
>>
>>
>>> Hi,
>>
>>> One of my team mate Sanyama raised few queries in this group. I  
>>> wanted
>>> to raise few more queries regarding this.
>>
>>> First of all, I am MS SQL Server developer and unfortunately think
>>> according to that. We are using C#.net as a development language.
>>
>>> Searching through net, we are highly impressed with Hybertable   
>>> and so
>>> want to use this (instead of SQL Server) in our new project in which
>>> massive search operations are there.
>>
>>> Before starting the project we are studying hypertable and would  
>>> like
>>> to know how the following things can be achieved:
>>
>>> 1. Joining the tables - In hypertable joining of the tables is not
>>> there. So do we require to get the data from more than one table and
>>> do the joining in frond end (C#) code?
>>
>> Doing the join on the front end might not be very efficient depending
>> on the volume of data returned by your queries. The general idea is  
>> to
>> design around the lack of "join" functionality by denormalizing your
>> data. Another option could be using a map-reduce system to do joins
>> over large datasets although this is an offline solution.
>>
>>
>>
>>> 2. Searching on non-key fields - Suppose I have employee table which
>>> have 2 columns: EmployeeId and EmployeeName, and want to search  
>>> based
>>> on employee name such as "Show me all employees whose name is Sachin
>>> or all employees in which Sachin exists in EmployeeName". How can we
>>> achieve this?
>>
>> You could create a separate table with the EmployeeName in the rowkey
>> and the value would be the rowkey(s) of the Employee table. This  
>> table
>> is basically an index you create. However, Hypertable does not  
>> provide
>> transactions across tables, so the application code needs to be
>> designed to handle potential inconsistency.
>>
>>
>>
>>> Eager to get response on the above queries.
>>
>>> Regards,
>>> Sachin- Hide quoted text -
>>
>> - Show quoted text -- Hide quoted text -
>>
>> - Show quoted text -
> >


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Hypertable Development" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/hypertable-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to