Best Practices for running VMware vSphere on NetApp ONTAP

Best Practices for running VMware vSphere on NetApp ONTAP

Running VMware vSphere on NetApp ONTAP is as good a combination as peanut butter and jelly. And it’s not just me who is saying that NetApp and VMware have more than 50000 joint customers that are eating that sandwich. In this blog post, I have listed most of the best practices that you should consider when you are running VMware vSphere on top of NetApp ONTAP. Before we list all the best practices, let’s try and look at the various advantages that NetApp brings to the table:

  1. Unified Storage: You can run any type of workload (vSphere Infrastructure[VSI] or virtual desktop infrastructure[VDI]) on any kind of disks (SSD or SATA) using any protocols (FC, iSCSI, NFS etc.) anywhere (On-Prem or the Cloud) and you will get all the same set of features.
  2. NetApp Snapshot: Snapshots are a space efficient data protection mechanism. ONTAP based snapshots are more efficient as compared to the VMware-based snapshots as you don’t have any performance impact on the ESXi host.
  3. Storage Efficiency: With its latest release, ONTAP offers storage efficiency features like inline deduplication, compression, and compaction. These offer a high amount of savings in terms of the storage required to run all different VMware workloads.
  4. NFS: Deploying vSphere with ONTAP NFS datastores results in a tenfold increase in datastore density with a correlating reduction in the number of datastores. But, you should also consider the performance benefits that you get when using multiple datastores running on a single NetApp controller. Having different datastores also allows you to implement different recovery policies for your VMs.
  5. VMware vStorage APIs for Array Integration (VAAI): VAAI enables copy offload to ONTAP for VM cloning operations, space reservation for thick virtual disk files and ONTAP Snapshot copies for vSphere linked clones.
  6. Multipathing: vSphere includes built-in support for multiple paths to storage devices, referred to as native multipathing (NMP). Both NMP and NetApp ONTAP support the Asymmetric Logical Unit Access (ALUA) protocol to negotiate optimized and nonoptimized paths. In ONTAP, an ALUA-optimized path follows a direct data path, using a target port on the node that hosts the LUN being accessed. ALUA is turned on by default in both vSphere and ONTAP. The NMP recognizes the NetApp cluster as ALUA, and it uses the ALUA storage array type plug-in (VMW_SATP_ALUA) and selects the round robin path selection plug-in (VMW_PSP_RR).
  7. Virtual Storage Console (VSC): Using VSC to configure LUNs and NFS datastores ensures that these resources are configured according to NetApp best practices.

Next, let’s look at the best practices. Please keep in mind that this list isn’t an exhaustive one, you can always refer to TR-4597 to get everything you need for running vSphere on NetApp.

  • To get the largest storage savings using features like deduplication, compression and compaction, group similar operating systems and similar applications onto one datastore.
  • For high-I/O applications such as databases, avoid datastores and use guest owned file systems such as NFS or iSCSI file systems managed by the guest or with RDMs.
  • Always use NetApp FlexVol volumes for running NFS datastores.
  • Use VLANs to separate the storage network traffic from everything else. If you share physical paths, then use QoS or add additional ports to ensure sufficient bandwidth.
  • When using Jumbo frames, configure it end to end to avoid any performance or connection problems.
  • When using Ethernet storage networks, use Rapid Spanning Tree Protocol. If using Cisco switches then enable the spanning tree portfast feature.
  • Always use multichassis link aggregation group when configuring link aggregation, this avoids any downtime during a single physical switch failure.
  • When using LACP, use dynamic multimode interface groups with IP hash and use IP hash teaming policy on ESXi hosts.
  • Create a logical interface (LIF) for each SVM on each node in the ONTAP cluster for maximum availability and mobility.
  • For iSCSI networks, use multiple physical NICs connected to multiple physical switches to provide HA and increased throughput.
  • If Challenge-Handshake Authentication Protocol (CHAP) is used in ESXi for target authentication, it must also be configured in ONTAP using CLI or OnCommand System Manager.
  • There is no automatic datastore conversion between NFSv3 and NFSv4.1, create a new NFSv4.1 datastore and use Storage vMotion to migrate VMs to the new datastore.
  • NFS export policies are used to control access to vSphere hosts. Use one export policy with multiple volumes (datastores) on the NetApp cluster.
  • Use ‘nfs’ as the protocol in the export policy when using the NetApp NFS Plug-In for VMware VAAI. This ensures that VAAI copy offload works with both NFSv3 and NFSv4 versions.
  • When creating datastores for VMware clusters with VSC, select the cluster rather than a single ESXi server. This will trigger VSC to automatically mount the datastore to all hosts in the cluster.
  • Use the VSC mount function to apply existing datastores to new servers.
  • Mount the junction path for volumes for vSphere at the root volume of the SVM on NetApp.
  • When using block protocols, use thin provisioned LUNs running on thin provisioned FlexVol volumes with a capacity that is two times the size of the LUN to get the best deduplication savings.
  • When using NFS, use thin provisioned NFS FlexVol volumes as datastores.
  • Snapshot copies might lock blocks that could be reduced by compression or deduplication. When using scheduled post process efficiency or using one-time scanners, make sure that they run and complete before the next Snapshot copy is taken.
  • When using Storage DRS(SDRS) with VSC, first create a datastore with VSC, then use vCenter to create the datastore cluster and add the datastore to it. After the datastore cluster is created, additional datastores can be added to the cluster directly from the VSC provisioning wizard.
  • All datastores in the cluster should use the same type of storage (such as SAS, SATA, SSD), be either all VMFS or NFS datastores, and have the same replication and protection settings.
  • After SDRS moves VMDKs, recreate your Snapshot copies at the source datastore because they are not moved with the VM.
  • All VVol datastores must be a part of a single ONTAP SVM and accessed using a single protocol.
  • As with traditional datastores, you should group similar guest operating systems and applications into the same datastore to maximize deduplication savings.
  • Do not install or migrate VASA Provider, vCenter Server (appliance or Windows based), or VSC onto a VVol datastore.

And on top of all these best practices, you should always use NetApp VSC to provision datastores (VMFS or NFS and traditional or VVols).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s