A FlexGroup sees outliers over 100ms into seconds causing client timeouts

• ONTAP 9
• FlexGroups
• CIFS
• NFS

• A particular FlexGroup has good average latency, but looking at histograms shows high outliers.
• A client application is very latency sensitive
• Example: High outliers may be seen in various outliers such as the NFSv3 lookup latency histogram.

Cluster::> set adv

Warning: These advanced commands are potentially dangerous; use them only when directed to do so by NetApp personnel.
Do you want to continue? {y|n}: y

Cluster::*> statistics
    statistics    statistics-v1
Cluster::*> statistics start -object nfsv3 -counter lookup_latency_hist
Statistics collection is being started for sample-id: sample_381

Cluster::*> statistics
    statistics    statistics-v1
Cluster::*> statistics show

Object: nfsv3
Instance: svm2
Start-time: 12/19/2024 11:30:20
End-time: 12/19/2024 11:37:20
Scope: node1
Number of Constituents: 1 (complete_aggregation)
    Counter                                                     Value
    -------------------------------- --------------------------------
    lookup_latency_hist                                             -
...
                              <100ms                               12
                              <200ms                               16
                              <400ms                                9
                                 <2s                               15
                                 <6s                                3
Cluster::*>