Re: NoSE: Automated schema design for Cassandra

2017-05-11 Thread Michael Mior
Thanks for the feedback! I did change column families to tables. I agree
the documentation could use some work. If you're interested in seeing what
the input and output look like, here's a sample:

https://michael.mior.ca/projects/nose/rubis

So far we haven't had any schemas used directly for production although it
has provided some advice on design alternatives. NoSE actually already
contains a mechanism to execute different schema alternatives which is what
we used during our evaluation. However, it does not currently directly
provide mechanism for synthetic data generation. It would definitely be
possible however to add automated generation of test data in the future.

Cheers,
--
Michael Mior
mm...@uwaterloo.ca

2017-05-10 3:55 GMT-04:00 Jacques-Henri Berthemet <
jacques-henri.berthe...@genesys.com>:

> Hi,
>
>
>
> This is interesting, I’d just advise to put full examples and more
> documentation on how to use it (the articles are a bit too detailed).
>
> Also, you should not mention “column families” but just tables.
>
>
>
> Was this used to generate a schema used for production?
>
> Do you think it’s possible to generate test code to validate the workload?
>
>
>
> *--*
>
> *Jacques-Henri Berthemet*
>
>
>
> *From:* michael.m...@gmail.com [mailto:michael.m...@gmail.com] *On Behalf
> Of *Michael Mior
> *Sent:* mardi 9 mai 2017 17:30
> *To:* user <user@cassandra.apache.org>
> *Subject:* NoSE: Automated schema design for Cassandra
>
>
>
> Hi all,
>
>
>
> I wanted to share a tool I've been working on that tries to help automate
> the schema design process for Cassandra. The short description is that you
> provide information on the kind of data you want to store and the queries
> and updates you want to issue, and NoSE will perform a cost-based analysis
> to suggest an optimal schema.
>
>
>
> There's lots of room for improvement and many Cassandra features which are
> not currently supported, but hopefully some in the community may still find
> it useful as a starting point.
>
>
>
> Link to more details and the source code below:
>
>
>
> https://michael.mior.ca/projects/nose/
>
>
>
> If you're interested in trying it out, don't hesitate to reach out and I'm
> happy to help!
>
>
>
> Cheers,
>
> --
>
> Michael Mior
>
> mm...@uwaterloo.ca
>


RE: NoSE: Automated schema design for Cassandra

2017-05-10 Thread Jacques-Henri Berthemet
Hi,

This is interesting, I’d just advise to put full examples and more 
documentation on how to use it (the articles are a bit too detailed).
Also, you should not mention “column families” but just tables.

Was this used to generate a schema used for production?
Do you think it’s possible to generate test code to validate the workload?

--
Jacques-Henri Berthemet

From: michael.m...@gmail.com [mailto:michael.m...@gmail.com] On Behalf Of 
Michael Mior
Sent: mardi 9 mai 2017 17:30
To: user <user@cassandra.apache.org>
Subject: NoSE: Automated schema design for Cassandra

Hi all,

I wanted to share a tool I've been working on that tries to help automate the 
schema design process for Cassandra. The short description is that you provide 
information on the kind of data you want to store and the queries and updates 
you want to issue, and NoSE will perform a cost-based analysis to suggest an 
optimal schema.

There's lots of room for improvement and many Cassandra features which are not 
currently supported, but hopefully some in the community may still find it 
useful as a starting point.

Link to more details and the source code below:

https://michael.mior.ca/projects/nose/<https://michael.mior.ca/projects/nose/>

If you're interested in trying it out, don't hesitate to reach out and I'm 
happy to help!

Cheers,
--
Michael Mior
mm...@uwaterloo.ca<mailto:mm...@uwaterloo.ca>


NoSE: Automated schema design for Cassandra

2017-05-09 Thread Michael Mior
Hi all,

I wanted to share a tool I've been working on that tries to help automate
the schema design process for Cassandra. The short description is that you
provide information on the kind of data you want to store and the queries
and updates you want to issue, and NoSE will perform a cost-based analysis
to suggest an optimal schema.

There's lots of room for improvement and many Cassandra features which are
not currently supported, but hopefully some in the community may still find
it useful as a starting point.

Link to more details and the source code below:

https://michael.mior.ca/projects/nose/

If you're interested in trying it out, don't hesitate to reach out and I'm
happy to help!

Cheers,
--
Michael Mior
mm...@uwaterloo.ca