Hi,

On Thu, Nov 27, 2014 at 6:36 PM, Inosh Goonewardena <[email protected]> wrote:

> Hi,
>
> On Thu, Nov 27, 2014 at 4:34 PM, Maninda Edirisooriya <[email protected]>
> wrote:
>
>> This seems nice. But I am not sure how the precomputation works and how
>> the intermediate indexes are stored in Service Layer.
>>
>
> Basically the idea should be to store outputs from both batch and speed
> layers in the serving layer so that service layer can respond to ad-hoc
> queries using already processed data. According to the given architecture,
> outputs from both BAM and CEP is going to be saved in CEP Event table which
> can be a RDBMS table or in memory grid.
>
>

For this use case RDBMS tables seem to be a better option over the
in-memory tables.

In addition to the RDBMS, Cassandra event tables can also be used through
an existing extension. (however for the current version, this involves
applying a patch to the siddhi core).


Thanks
Rajeev



> And also does the user have to write both hive scripts and Siddhi queries
>> for the same scenario?
>>
>> Thanks.
>>
>>
>> *Maninda Edirisooriya*
>> Senior Software Engineer
>>
>> *WSO2, Inc.*lean.enterprise.middleware.
>>
>> *Blog* : http://maninda.blogspot.com/
>> *E-mail* : [email protected]
>> *Skype* : @manindae
>> *Twitter* : @maninda
>>
>> On Thu, Nov 27, 2014 at 2:25 PM, Dinithi De Silva <[email protected]>
>> wrote:
>>
>>> Hi all,
>>>
>>> Myself with Zamly ([email protected]) and Inshaf ([email protected]) are
>>> going to develop $subject as described below.
>>>
>>> *Introduction to Lambda architecture*
>>> The lambda architecture solves the problem of computing arbitrary
>>> functions on arbitrary data in real time by decomposing the problem into
>>> three layers: the batch layer, the serving layer, and the speed layer.[1]
>>>
>>> *Batch layer*
>>> The batch layer stores the master copy of the dataset and precomputes
>>> batch views on that master dataset.
>>>
>>> *Speed layer*
>>> The speed layer is needed to compensate for the high latency of the
>>> batch layer and it does this by computing real time views in Storm.
>>>
>>> *Serving layer*
>>> The serving layer is responsible for indexing and exposing the batch
>>> view and real time view, so that they can be queried for updated results.
>>>
>>> *Approach*
>>> We are planing to implement Lambda architecture using CEP and BAM.
>>> Please find the diagram for the architecture below.
>>>
>>> ​Feed backs and suggestions for improvements would be appreciated.
>>>
>>> [1] -
>>> http://jameskinley.tumblr.com/post/37398560534/the-lambda-architecture-principles-for
>>>
>>> --
>>> *Dinithi De Silva*
>>> Associate Software Engineer, WSO2 Inc.
>>> m:+94716667655 | e:[email protected] | w: www.wso2.com
>>> | a: #20, Palm Grove, Colombo 03
>>>
>>> _______________________________________________
>>> Architecture mailing list
>>> [email protected]
>>> https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture
>>>
>>>
>>
>> _______________________________________________
>> Architecture mailing list
>> [email protected]
>> https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture
>>
>>
>
>
> --
> Regards,
>
> Inosh Goonewardena
> Associate Technical Lead- WSO2 Inc.
> Mobile: +94779966317
>
> _______________________________________________
> Architecture mailing list
> [email protected]
> https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture
>
>


-- 
Rajeev Sampath
Senior Software Engineer
WSO2, Inc.; http://www.wso2.com.

Mobile:
* +94716265766*
_______________________________________________
Architecture mailing list
[email protected]
https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture

Reply via email to