hope this fix your issue I have found the answer for the problem, when I stored the data in s3, the data inside the file was with one more column that was not defined in the hive table metastore. So when Presto tried to query the data, it found that there are varchar instead of integer.
PRESTO + dose presto cordinator can manage 160 workers node?
I wish this helpful for you TL;DR in general, coordinator can easily handle much more worker nodes than 160, but your milage may vary. Longer version Presto coordinator can manage 1000 workers. However, you're asking for a coordinator with particular memory and CPU resources. The answer is -- it depends.
SQL (Presto) - 'compress' rows when date ranges are sequential
will help you This is a gaps-and-islands problem. The key idea is to use lag() to find where the value changes and then do a cumulative sum to assign groups. However, I have come to prefer a lag on the date column rather than the value column. It turns out, this is much more convenient when you have multiple values that might change.
select event_id, min(period_start), max(period_end), rating
from (select t.*,
sum(case when prev_period_end = period_end then 0 else 1 end) over (partition by event_id order by period_start) as grp
from (select t.*,
lag(period_end) over (partition by event_id, rating order by period_start) as prev_period_end
group by event_id, rating, grp;
Presto DB + what we can do in case of some presto workers are down + auto start
wish helps you This is something we (Presto community) are currently working on, please follow: High Availability. We are also invite you to put your comments there, so we could verify if your use case will be covered. Also, if you are using AWS for Presto you might to consider checking StarburstData offering for Presto Coordinator High Availability
after modified the config files in /etc/presto/conf, how to restart presto-server
will be helpful for those in need In aws emr, after modified the config file in /etc/presto/conf, how can we restart presto-server? Just on master node or on all nodes? , On EMR you can restart Presto with