Skip to main content
NetApp Knowledge Base

What are some considerations of Cloud Volumes ONTAP AWS Performance?

  1. Last updated
  2. Save as PDF
Views:
3,045
Visibility:
Public
Votes:
1
Category:
cloud-volumes-ontap-cvo
Specialty:
cloud
Last Updated:

Applies to

Cloud Volumes ONTAP (CVO) on AWS

Answer

The AWS backend has many options, and there are choices within CVO that can cause issues if not considered:

Disk
Instance Type

  • AWS instance types can be described as the hardware dedicated to the Cloud Volumes ONTAP install, including CPU, RAM, and network bandwidth

    • General Purpose - Balanced resources
      • T2 - low cost, burst performance oriented
      • M4 - balanced resources for many workloads
    • Compute optimized - better for workloads that need more CPU resources (deduplication, compression, compaction)
      • C4 - Latest generation Intel Xeon processors
    • Memory optimized - better for workloads with large working sets (high file count, complex directory structure, database workloads)
      • X1 - Optimized for large-scale, enterprise-class, in-memory applications
      • R4 - Optimized for memory-intensive applications
Other Considerations
  • Cloud Volumes ONTAP HA functionality
    • Cloud Volumes ONTAP does support HA configurations, though there may be performance issues for some workloads. Improvements to this feature are constantly being made.
  • NVLOG performance
    • As NVRAM is not a physical RAM DIMM but SSD backed, this can become a bottleneck
    • For workloads with higher write speed needs, a larger instance size is needed, and faster drives to prevent bottlenecks upstream
    • Write speed (HA Supported as of ONTAP 9.8, only single node in 9.7)
      • Normal - Data is written to NVRAM prior to being committed to disk - this is the safest option and should be used in most cases
      • High - Data is simply left in active memory buffers and committed to disk - since data is NOT written to NVRAM, should an unplanned shutdown occur, data could be lost, not recommended and should only be used for workloads with transient data that can safely be lost
  • Usage profile options that get setup when deploying Cloud Volumes ONTAP
    • Highest Performance - Recommended for applications requiring lowest latency
    • Performance with Efficiency - Good performance with ONTAP storage efficiency (deduplication)
  • Shared Tenancy vs. Dedicated Hardware
    • Shared tenancy is the most commonly selected option, but you could experience "noisy neighbor" symptoms from other AWS workloads, dedicated hardware avoids this but is more costly
  • Node IOPs limit
    • Each backend disk can handle

Additional Information

 

 

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.