mgsql and a support optimization query speed by index global

[Josef Zvonicek]

Hi,

I am trying to work with mgsql on big global data and I have a problem with speed. 

Is there any index optimization when mgsql engine generate a temporary query routine? 

It seems that no.

Does anybody have experience with mgsql and big data?

Thanks Josef.

[chris....@gmail.com]

Query optimisation is always a difficult process to get right.  The ‘gold standard’ is that the SQL engine will automatically chose the most optimum path through the tables and choose the best indices to use.

 

However, in cases where the system is unable to best optimise a query, manual intervention is needed.  Most SQL engines will allow for this (at least to some degree) and is something that I have on the TODO list for mgsql.

 

Given that mgsql is Open Source I don’t know of anyone who might be using it with large data sets.  However the SQL engine I created before mgsql is certainly being successfully used for large databases – this being a proprietary product.  That said, my intention is that mgsql should be an improvement over what I’ve created previously!

 

Chris.

--
You received this message because you are subscribed to the Google Groups "Enterprise Web Developer Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to enterprise-web-develope...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/enterprise-web-developer-community/2dc15e82-c5c7-4a35-aa9f-dc8f07040e7cn%40googlegroups.com.

[K.S. Bhaskar]

Chris –

You may find AIM (Application Independent Metadata) helpful. See https://opensourcevista.net/NancysVistAServer/39thVCMVideosAndSlides/211020-1YDBAIMIntro.pdf for slides and https://opensourcevista.net/NancysVistAServer/39thVCMVideosAndSlides/39thVCM20211023YottaDB-ApplicationIndependentMetadata-Bhaskar.mp4 for the talk (there will soon be links from yottadb.com as well).

AIM is at https://gitlab.com/YottaDB/Util/YDBAIM and pull requests are always welcome!

Regards

– Bhaskar

[chris....@gmail.com]

Further to my previous post, MGSQL v1.4.22 is now released and should hopefully address problems with SQL optimisation.

 

The SQL optimiser has been improved and there is now the facility to provide ‘hints’ to the optimiser in the form of explicit index specification in the query text.  It is also possible to steer the optimiser towards parsing tables specified in a FROM statement in a particular order.  Full documentation is in the distribution.

 

https://github.com/chrisemunt/mgsql

 

Chris.