Hi, I found : For select queries , memory allocation is related to the amount of columns on the select list (its around 2M per column) , in many cases it also related to many other factors like if your query include aggregation and sorting and it also related to your cluster setup (how many nodes ) ...

Thank you, Jim for your time! Let's continue! There are 2 parameters: RUNTIMECAP - Sets the maximum amount of time any query on the pool has to execute. QUEUETIMEOUT - Integer value, in seconds, indicating the maximum time the request can wait for resources to become available before rejection. Both...

documentation:
https://www.youtube.com/watch?v=-ClP29- ... AyPvCIWVgb
https://my.vertica.com/docs/7.1.x/HTML/ ... _Pools.htm
https://my.vertica.com/wbt/11_UserDefin ... /index.htm

Hi, I need the confirmation that I understood correctly how a query is executing, passing from a resource pool to another. I imagine the query starts in 1_pool. After runtimecap corresponding to 1_pool it falls in 2_pool if runtimecap 2_pool > runtimecap 1_pool . 1_pool |---------runtimecap---------...

Thank you, again.

Hi, Thank you for your answers. You said: "Vertica does not perform well with wide tables." My question: "wide" is referring to number of rows? number of columns ? From my 1000 columns , 900 are number(16,5). You said: "The memory allocated to these instances is quite low." My question is: How to ca...

Hi, Volumes with the same configuration - (SSD/gp2), size, IOPS, EBS - on the same instance had different I/O performances ( read/write per second, latency). So, we tried to put some volumes from AWS in RAID configuration but didn't obtain an increase of performance. I think due to network involved....