dlmarion commented on issue #3211:
URL: https://github.com/apache/accumulo/issues/3211#issuecomment-1458089562

   > Is there a mechanism to alert on approaching resource limits or evaluate 
the cluster's current size/use before marking a tablet as onDemand?
   
   Given that we don't know which schedulers might be in use on a cluster, I 
think we just need to emit metrics that can be used for a scheduling system to 
make a determination that more tablet servers are needed. Accumulo doesn't do 
any alerting, that's typically set up by the users of the system - trigger an 
alert when some criteria is met.
   
   > Given the possibility of a limited resource footprint, what mechanism is 
going to be used for scheduling the tablet hosting?
   
   I don't think we should build or use a specific scheduler. 
[KEDA](https://keda.sh/) and 
[HPA](https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale/)
 exist already and I'm sure the different commercial cloud vendors may supply 
their own solutions as well (Azure has 
[VPA](https://learn.microsoft.com/en-us/azure/aks/vertical-pod-autoscaler) for 
example).
   
   > What determines hosting priority of onDemand tables? should this exist? 
should it be a per-client setting or perhaps per table?
   
   I don't think we have had any discussion of priority. Can you give an 
example of how priority would work?
   
   > Are "resources unavailable for hosting" an error condition that can be 
handled for onDemand tables?
   
   I'm not sure that we can do this reliably. If we were able to determine a 
cost (in terms of memory required to host a tablet), a TabletServer may or may 
not have enough memory to host another tablet. It may appear that the 
TabletServer doesn't have the memory, but after a GC it very well could. I'm 
not sure there is a good answer here.
   
   > Is there a possibility of scheduling deadlock situations?
   
   I don't think so, this is using the existing mechanism of assigning online 
unassigned tablets to tablet servers. When the client requests that an ondemand 
tablet is assigned, it just puts a marker entry in the tablet metadata for that 
tablet. The TabletGroupWatcher process in the Manager will notice this change 
and on it's next pass assign it to a TabletServer.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to