8 Replies Latest reply on Feb 5, 2016 6:21 AM by Toby Erkson

Tableau server 9.2 processes scalability

Su ve Feb 2, 2016 11:42 AM

Hi guys,

We have a tableau server with 8cores, 32GB RAM on single server environment.

We have at max of 200 concurrent users.

Recently we have upgraded the machine hardware and also tableau server from 8.0 to 9.2.

We have

Vizql-2

application-1

data engine- 1

data server- 2

Background- 1

Cache - 2

And we are planning to keep the cache refresh time as: High Availability (or 30 minutes).

But our database is not that fast. So, how this is going to impact the performance?

How can I distribute the processes on the server for maximum performance?

We have good amount of extracts.

Thanks,

Suve

1. Re: Tableau server 9.2 processes scalability

Andrew Macey Feb 2, 2016 5:43 PM (in response to Su ve)

http://onlinehelp.tableau.com/current/server/en-us/help.htm#perf_extracts_view.htm
2 of 2 people found this helpful
Like Show 1 Likes(1)

Actions
2. Re: Tableau server 9.2 processes scalability

Nathan Panuco Feb 3, 2016 10:23 AM (in response to Su ve)
Hi Su!

If you are doing a lot of extracts, you may want to look into upping the number of backgrounder processes you have running. Check out this article on environments where lots of extracts are performed: http://onlinehelp.tableau.com/current/server/en-us/help.htm#perf_extracts_view.htm#idf65a85c6-9053-4c4b-8bf2-866998cb7c7…

One-Machine example: Optimized for heavy extract usage
This example shows a 64-bit, 8+ core, 16+ GB system configured for heavy extract usage.
This configuration would look like the following Process Status table on the Server Status page.
Configuration Notes:
The primary server runs 2 VizQL server processes, 2 cache server processes, and 2 data server processes. These are the recommended defaults from installation.
As a general rule, run a cache server process for every VizQL server process on the node.
Calculate the least number of background processes to run by taking the machine’s total number of cores and divide it by 4. To determine the maximum number, divide by 2.
Both the background and data engine processes are CPU-intensive and the configuration shown above balances them.
Schedule extract refreshes for off-peak times to help the VizQL server, application server, data engine, and background processes to not compete with one another for system resources.
1 of 1 people found this helpful
Like Show 0 Likes(0)

Actions
3. Re: Tableau server 9.2 processes scalability

John Kuo Feb 3, 2016 10:33 AM (in response to Su ve)

With your current hardware specs, you can add one more backgrounder process to help with scheduled tasks.

By the way, are you sure you have 200 concurrent users? That seems high as typical concurrency estimation is 10% of total users. If you really have 200 concurrent users you need to add more CPU cores and RAM for your server. Do you know if you have server core-based or user-based license? If it's user-based, Tableau does not how many server cores you are running on Tableau Server so work with IT and see if you can up the specs to 16 cores and 64GB RAM. After that, I'd increase VizQL and Cache Server to 4.

Best,

John
2 of 2 people found this helpful
Like Show 2 Likes(2)

Actions
4. Re: Tableau server 9.2 processes scalability

Su ve Feb 3, 2016 6:43 PM (in response to Nathan Panuco)

Thanks Nathan!
Like Show 0 Likes(0)

Actions
5. Re: Tableau server 9.2 processes scalability

Su ve Feb 3, 2016 7:07 PM (in response to John Kuo)

John,
I will add one more Backgrounder.

We have 250 licensed users. However, at the peak hours (i.e., on Monday Mornings) Access Views are hitting as high as 180. We have core licensing with Guest Login enabled. We are maximum no of cores as per our License. We cannot increase the cores.
At the same peak hours many Extract Refresh Tasks are running too.
So, can I add more Vizql and application processes to withstand the concurrency?
Or do I need to increase RAM for that?

And also I have found this article saying, "Should have same no of Application and Vizql processes". Do I need to take care of that too?
Optimizing Tableau Server Performance | Tableau Software

Thanks,
Suve
Like Show 0 Likes(0)

Actions
6. Re: Tableau server 9.2 processes scalability

Toby Erkson Feb 4, 2016 6:12 AM (in response to Su ve)

Also...
Su ve wrote:

...
But our database is not that fast. So, how this is going to impact the performance?
...
Are you saying the data source is not fast? Please provide more detail on this.
Like Show 0 Likes(0)

Actions
7. Re: Tableau server 9.2 processes scalability

Su ve Feb 4, 2016 5:17 PM (in response to Toby Erkson)

Toby,

Our database mostly contain nested views. And also the reports are using complex nested queries.
Running those queries on the database takes more time. So, when I change the Cache refresh time to 10 minutes or so, does it impact the database performance?

Thanks,
Suve
Like Show 0 Likes(0)

Actions
8. Re: Tableau server 9.2 processes scalability

Toby Erkson Feb 5, 2016 6:21 AM (in response to Su ve)

Su ve wrote:
...And also the reports are using complex nested queries.
Running those queries on the database takes more time. So, when I change the Cache refresh time to 10 minutes or so, does it impact the database performance?
Sounds like you are using Custom SQL for pulling data, which is the less preferred method and can impact performance at least for anything live. I really don't know how Custom SQL affects extracts.

Caching would lower the impact to a database assuming subsequent visits by others will use this data as well. Caching is for reports that have been called up by a user.
Like Show 0 Likes(0)

Actions