Red Hat Gluster Storage: Compatible Physical, Virtual Server and Client OS Platforms

Updated -

Red Hat Gluster Storage Compatibility

Red Hat has established the following generic and use-case specific criteria for Red Hat Gluster Storage compatibility of specific servers. Red Hat strongly recommends that all customers source their hardware platforms for running Red Hat Gluster Storage for On-premise based on the criteria specified below to ensure supportability of production deployments.

Generic Requirements for a server to be Red Hat Gluster Storage Compatible

  • Must be in the Red Hat Hardware Compatibility List for Storage for Red Hat Enterprise Linux 6.0 and newer.
  • 2-socket (with 4-core, 6-core, or 8-core) servers are recommended.
  • Minimum RAM requirements are use case specific. See below.
  • Reliable backplane RAID controller shipped by server vendors or from OEM manufacturers.
  • RAID 6 and RAID 1+0 Support in hardware RAID controller.
  • RAID controller card must be flash-backed or battery-backed.
  • 1X 50 GB GB SAS disks for RHSS installation if a separate partition is created for /var
    OR 200 GB if no separate partition is created for /var.
  • 1 X 10 GigE NIC for data traffic is recommended. It's advisable to use NIC bonding with 2 X 10 GigE for increasing throughput and resiliency. 1 GigE NIC may also be used.
  • Redundant power supply
  • Out of band management card to manage and monitor RHS nodes even when the server is down.(optional)

Note:

  • The same interface may be used for data and management traffic, but it's recommended to have separate networks for data and management.

High Performance Computing use-case

  • 2u/24 (JBODs may be attached)
  • 15000 RPM 900GB drives(2.5" inch SAS) OR Solid state disks.
  • Minimum RAM 48 GB

General Purpose File Serving use-case

  • 2u/12 (JBODs may be attached)
  • 7200 or 10000 RPM, up to 8 TB drives (3.5" SAS or SATA)
  • Minimum RAM 32 GB

Archival use-case

  • 4u/60 (JBODs may be attached)
  • 7200 or 10000 RPM, up to 8 TB drives (3.5" SAS or SATA)
  • Minimum RAM 16 GB

Note:

  • The number of disks supported per server in the JBOD configuration is limited to 36 with 3-way replication and erasure coding configuration.

Node Selection Guideline For Specific use-cases

Component HPC General Purpose Archival
Chassis (specifically applicable for Supermicro)

2u 24x2.5" Hotswap with redundant power

2u 12x3.5" Hotswap with redundant power

4u 60x3.5" Hotswap with redundant power

Processor Intel Xeon Nehalem-Ex and beyond, AMD equivalent Intel Xeon Nehalem-Ex and beyond, AMD equivalent Intel Xeon Nehalem-Ex and beyond, AMD equivalent
Disk 24x 2.5" 15000 RPM SAS OR SSD's along with 1.5K RPM SAS 12x 3.5" 7200 or 10000 RPM SAS/SATA OR SSD's along with 7.2K/10K RPM SAS/SATA 60x 3.5" 7200 or 10000 RPM SAS/SATA
minimum RAM 48GB 32GB 16GB
Networking 2x10GigE 2x10GigE (preferred) or 2x1GigE 2x10GigE (preferred) or 2x1GigE

Notes:

  • All data disks configured in groups of 12 drives in RAID6 configuration

Trusted Storage Pool Size Limits

Limit Minimum Maximum Notes
Trusted Server Pool Size 2 + 1 with arbitrated replicated volumes 128

Arbitrated replicated volumes with 2 storage nodes and 1 node hosting just the meta-data information in a local data-center using sync replication is supported. Please contact your Red Hat representative if you plan a deployment than needs a higher number of nodes in a storage pool.

Usable Capacity for most common configuration

No minimum

51.2 Petabytes with no replication. 25.6 Petabytes with 2-way replication in a single cluster of 128 nodes. Assuming a fully populated storage server with 60 drives per controller and 8 TB SATA drives. For distributed-dispersed and dispersed volumes 36 disks per controller is supported.

Please contact your Red Hat representative if you plan a deployment of higher scale than the maximum described here

Supported Virtual Platforms

Red Hat Gluster Storage nodes can be VMs running on listed virtual environments

Virtual Platforms Red Hat Gluster Storage supported releases
Red Hat Virtualization 4.2 Red Hat Gluster Storage 3.3.x
Red Hat Virtualization 4.1 Red Hat Gluster Storage 3.2, 3.3.x
Red Hat Virtualization 4.0 Red Hat Gluster Storage 3.1
Red Hat Enterprise Virtualization 3.6 Red Hat Gluster Storage 3.1
Red Hat Enterprise Virtualization 3.5 Red Hat Gluster Storage 3.1
Red Hat Enterprise Virtualization 3.4 Red Hat Gluster Storage, Red Hat Storage 2.1/U1/U2(Compatibility Mode)
VMWare ESXi 5.x, 6.x Red Hat Storage Server 3.x, Red Hat Gluster Storage (Minimum 4 vCPUs, 16 GB RAM), Use VMDKs or RDMs

NOTE: Please refer to the Red Hat Enterprise Virtualization Life Cycle page for full Manager and Hypervisor compatibility details.

Supported Client and Client OS Platforms

If you are using NFS/CIFS the following OSs are supported

  • All Fedora and Debian based distributions.
  • UNIX (Solaris 10+), AIX, HP-UX
  • Microsoft Windows Server 2008, Windows 7, Windows 8, Windows Server 2012, Windows Server 2016.

If you are planning to use the GlusterFS FUSE client, support is limited to:

  • Red Hat Enterprise Linux 5.8 and newer.
  • Red Hat Enterprise Linux 6.0 and newer.
  • Red Hat Enterprise Linux 7.0 and newer.

For GlusterFS FUSE client support from other Linux based client OS platforms/distributions see this:

Red Hat does not test native glusterfs clients which it does not ship. For these clients, the scope of support is limited to the server side. If such clients are based on the same upstream release of Gluster, Red Hat will support and offer to diagnose issues around Red Hat Gluster Storage Server related to their use. If the issue is determined to be specific to a client then you may be asked to reproduce on a RHEL system or, using NFS to confirm the same. In such cases, the customer would be advised to seek support from the provider of the glusterfs client and not Red Hat.

Other Related KBases

Table of Contents

No

19 Comments

Has supermicro server support been removed?  It's mentioned, but there's no table of configurations in the article.

Kevin - Super Micro testing is in progress. Stay tuned! It will be posted here once all testing has completed.

Supermico configs have been decided, will be doing some in-house testing at RHT and will announce SM configs shortly.

Could the max.diskcapacity on the public HCL please be raised to 4 Tb instead of 2-3 Tb ?
I've discussed this already with Sayan and Scott Clinton, this is needed for a customer usecase in the Benelux region.
They confirmed to me that this would be allowed. Thanks in advance

Would it be possible to add a change log to this doc, so that's it's easier to spot what has been changed in a new version?

Why Windows 8 is not in Support OS list of CIFS?

I would like some performance estimates HPC, General purpose and Archival use cases to help chose hardware for the load that is expected.

What about support for RHSS 3.x on VMware ESX 5.x?

Is 4 TB hard drives supported yet?

Hi,
Any plan to support Oracle VM x86 as virtualisation plaftorm for RHSS 3 ?

Any update about this ?

How about RHELOSP 6? and RHEV3.5?

Why Windows 8.1 isn't in Support OS list of CIFS?

trying to install a vsphere 5.5 rhs 3.0.4 but do not have the boot loader proceed to the next screen. What is the reason?

2Dae Sub Lee
Press Tab and Enter. It seems that buttons are hidden :(

I had the same issue and was able to get the install to work by removing the "quiet" option from the boot string.

Hello,

could be updated/clarify the disk (JBOD vs RAID) recommendations according to the documentation?

link: https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3/html/Administration_Guide/chap-Configuring_Red_Hat_Storage_for_Enhancing_Performance.html

(added 24 dic 2015) and http://www.redhat.com/en/about/blog/red-hat-announces-new-capabilities-red-hat-gluster-storage (added 24 dic 2015)

Thanks. JLBG

I notice that Windows 10 is not listed in the NFS/CIFS listing. It may also be helpful to break out NFS and CIFS into a matrix to see what supports what.

How much RAM is required for Gluster in a RHHI (hyperconverged VMs+Gluster) deployment? Section 2.2 of the Deployment Guide [1] suggests small=64G, medium=128G, large=256G but that's for everything (Gluster, KVM overhead, and VM use).

[1] - https://access.redhat.com/documentation/en-us/red_hat_hyperconverged_infrastructure_for_virtualization/1.5/html-single/deploying_red_hat_hyperconverged_infrastructure_for_virtualization/index#rhhi-req-pm