5 Essential Tips for Maximizing Your Experience at Nutanix .NEXT for Bloggers
Thanks all for the fast response. The more I know the more I Like the Nutanix software....
Jacky , giving the size of your DB (80 GB) the I/O in your benchmark is all on the SSD layer. The I/O latencies reported from the DB side are higher than what is expected from SSD with Nutanix. This could be due to Async I/O where many I/Os are issued and then the DB wait for the completion. OR this could be due to a saturation of a component in the I/O chain. It could be useful to have a look at the IO stats from Prism during the benchmark and to the CPU utilization of the CVM . You did not specify what TPS rate did you reach with your current traditional system are they higher or lower than Nutanix now?
Hi, you did not specify which processor are you using on the Nutanix 6035-G4. Usually all the modernn DBMS perform a lot of Async I/O trying to feed data to the CPUs as fast as possible. Hence the CPU speed of each core become a relevant factor. The E5-2690 v2 is a very fast CPU clocked at 3.00 Ghz and likely the 6035-G4 has a v3 CPU with a similar number of cores but with a slower clock. This "could" impact the overall benchmark results. You should have a look at the total CPU utilization during the benchmark to have a hint about it. Regards, Stefano
Hi Since the POC is based on DELL appliances I have asked yesterday DELL to open a support Ticket. There can be a lot to be improved in the customer environment but this does not explain the write latencies that should depend on the Nutanix implementation of the storage. Additionally the read could go on the SSD or on the HDD tier but ALL the write in this environmente should go to the SSD tier which should have latencies around 1 ms...
Yes, it is a single image. The desktops are non persistent and are assigned to the users at the logon. They are using roaming profiles and folder redirection on an external storage. This download of the user profile triggers a high number of writes at logon time. Today with an increased load of around 18-20 active desktop/node the average write latency is up to 7.5 ms and the average read latency is up to 2.3 m. What is worse is the peak latency observed. We monitored a specific Virtual Desktop that was "very active" . It has from 40 to 80 write/sec and the write latency was up to over 20 msec. I think there is surely something wrong because the total load is very low an average of 240 IO/sec. But every peak of requests produces a surge in the latency. One last word: the minimum write latency never goes below 6 ms.... We have not the data from the storage tab of Prism because we can get the data only asking the customer to get them for us and he does
Already have an account? Login
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.
Sorry, we're still checking this file's contents to make sure it's safe to download. Please try again in a few minutes.
Sorry, our virus scanner detected that this file isn't safe to download.
