Thanks for your interest in our work.  In your test case, it is clear that 
MySQL is faster.  If this test case is representative of your workload, and you 
are happy to load the data into MySQL, then you should stay with MySQL.

By the way, in your test case, the bulk of the execution time of ibis should be 
spent in printing the values.  The other thing to note might be that the time 
reported by MySQL is likely not to include the time to print the values.  
Anyway, just something to think about.

-- John --

> On Mar 27, 2014, at 10:17 AM, liuyingbo <[email protected]> wrote:
> 
> Dr.Wu:
>  
> I am very interested in your bitmap index library- Fastbit, I am a student 
> and my major is astronomical techniques and methods.
> Right now, I am trying to use this wonderful tool to do something about FITS 
> query, because our data is increasing every day.
>  
> Can I say MySQL be better than Fastbit? Because: (0.02 sec) < 1.02006 ,or 
> (0.02 sec) <1.02489.
>  
> ==============================
> Mysql:
>  
> Here is the performance of query  “select * from t_solar where 
> obs_date>1395904432 and obs_date<1395904483;”
>  
> ------------------------------------------
> 1500 rows in set (0.02 sec)
>  
> =========================
>  
> Fastbit:
>  
> Here is the performance of query  “/home/lyb/20140327/fastbit# 
> /home/lyb/softwares/fastbit-ibis1.3.5/examples/ibis -d 
> /home/lyb/20140327/fastbit/ -v -q "select observotary,obs_date,observer where 
> obs_date>1395904432 and obs_date<1395904483"”
>  
> -------------------------------------------
> total CPU time 1.02006 s, total elapsed time 1.02489 s
>  
>  
_______________________________________________
FastBit-users mailing list
[email protected]
https://hpcrdm.lbl.gov/cgi-bin/mailman/listinfo/fastbit-users

Reply via email to