Skip to main content
NetApp Knowledge Base

High Read or Write Latency due to disk bottleneck from user workload

  1. Last updated
  2. Save as PDF
Views:
3,400
Visibility:
Public
Votes:
1
Category:
ontap-9
Specialty:
perf
Last Updated:
 

Applies to

  • ONTAP 9
  • FAS systems with HDD disks

Issue

  • High latency for READ and WRITE operations is seen in volumes located on SATA or SAS disk aggregate which are highly utilized (100%)
  • Slow backup is observed due to high disk utilization
  • Slow retrieval of data from capacity tier to performance tier can be faced
  • Slow copies are seen from users on one cluster compared to another using the cp command from Linux
  • Use statit to identify if high disk utilization or latency in the aggregate containing the volume 
    disk             ut%  xfers  ureads--chain-usecs writes--chain-usecs cpreads-chain-usecs greads--chain-usecs gwrites-chain-usecs
/data_aggr1/plex0/rg0:                                                                  
0a.11.1           10  23.42    0.00   ....     .  11.21  61.56   133  12.21  57.15    50   0.00   ....     .   0.00   ....     .  
0a.10.1           11  23.80    0.00   ....     .  11.66  59.25   147  12.14  57.37    55   0.00   ....     .   0.00   ....     .  
1a.12.1          100 158.69  120.29   5.43  2299  10.26  34.39   553  28.15   9.93   839   0.00   ....     .   0.00   ....     .  
0a.11.2          100 156.09  118.85   5.39  2253   9.96  34.97   553  27.28  10.05   794   0.00   ....     .   0.00   ....     .  
0a.10.2          100 158.41  121.44   5.35  2337   9.95  35.11   542  27.01   9.93   835   0.00   ....     .   0.00   ....     .  
0a.12.2          100 153.05  115.90   5.59  2158   9.88  35.00   524  27.27   9.97   803   0.00   ....     .   0.00   ....     .  
0a.11.3          100 162.10  124.59   5.45  2262  10.05  35.10   528  27.45  10.10   809   0.00   ....     .   0.00   ....     .  
0a.10.3          100 158.72  121.88   5.39  2298   9.82  34.82   573  27.02  10.05   838   0.00   ....     .   0.00   ....     .  
1a.12.3          100 157.60  120.63   5.49  2274   9.85  34.90   571  27.11   9.82   875   0.00   ....     .   0.00   ....     .  
0a.11.4          100 157.67  120.74   5.39  2328   9.96  35.29   568  26.97  10.10   857   0.00   ....     .   0.00   ....     .  
0a.10.4          100 156.72  119.29   5.64  2180   9.96  34.65   556  27.47   9.79   825   0.00   ....     .   0.00   ....     .
  • Performance degradation is experienced on the client
  • Active IQ Unified Manager (AIQUM) monitoring the cluster raises an alert of higher latency
    • Example: Latency value of XX.X ms/op on volname has triggered a WARNING event based on threshold setting of 20.0 ms/op.
  • The read latency is 25~40ms
  • Disk utilization is high according to sysstat -x (columns removed for readability): 
    •  CPU  ...    HDD   
      ...   util  
 10%  ...    90%  
  9%  ...   100%

 

 

 

Sign in to view the entire content of this KB article.

New to NetApp?

Learn more about our award-winning Support

NetApp provides no representations or warranties regarding the accuracy or reliability or serviceability of any information or recommendations provided in this publication or with respect to any results that may be obtained by the use of the information or observance of any recommendations provided herein. The information in this document is distributed AS IS and the use of this information or the implementation of any recommendations or techniques herein is a customer's responsibility and depends on the customer's ability to evaluate and integrate them into the customer's operational environment. This document and the information contained herein may be used solely in connection with the NetApp products discussed in this document.