USE CASE question. analytics, grouping, summing, counting

classic Classic list List threaded Threaded
2 messages Options
Sanjeev Vihar Sanjeev Vihar
Reply | Threaded
Open this post in threaded view
|

USE CASE question. analytics, grouping, summing, counting

This post has NOT been accepted by the mailing list yet.
I am currently using an open source columnar database (infyniDB) which works well for us because it uses a familiar mySQL coding structure.

However we are taking no advantage of caching (particularly query-result caching).  Also none of our data tables are held in memory.  I am clear that we are missing an opportunity by not using Memory.

Our schema is simple with a main Fact table (approx 1gb) then 2 dimension tables.

Our data is updated daily, but users do update the dimension tables more regularly.  This concerns me because I am concerned that the dimension entries will become stale after a user make an update.

I'm not asking for a pathway to adoption, but at this stage just a simple "Yes try it" or "Try it but it's not ideal", or "This is not a great use case for Ignite" would be so useful.



vkulichenko vkulichenko
Reply | Threaded
Open this post in threaded view
|

Re: USE CASE question. analytics, grouping, summing, counting

Sanjeev Vihar wrote
I am currently using an open source columnar database (infyniDB) which works well for us because it uses a familiar mySQL coding structure.

However we are taking no advantage of caching (particularly query-result caching).  Also none of our data tables are held in memory.  I am clear that we are missing an opportunity by not using Memory.

Our schema is simple with a main Fact table (approx 1gb) then 2 dimension tables.

Our data is updated daily, but users do update the dimension tables more regularly.  This concerns me because I am concerned that the dimension entries will become stale after a user make an update.

I'm not asking for a pathway to adoption, but at this stage just a simple "Yes try it" or "Try it but it's not ideal", or "This is not a great use case for Ignite" would be so useful.
Hi Sanjeev,

You use case is definitely supported by Ignite. I think the best way to implement it is to do both updates and queries using Ignite API, utilizing write-through to update database. This way you will never have any stale entries. I would recommend to take a look at documentation page about the persistent store, I beleive it can be helpful for you: https://apacheignite.readme.io/docs/persistent-store

Let us know if you have more questions.

-Val