Hello, everyone,
I wanted to share the sizing of an accelerated WAN infra and the differences between WAN Low bandwidth mode vs WAN High bandwidth mode.
The DR consists of 6 replica VMs
- Sizing for Low bandwidth under 100 MB
Using RVTool extract the list of VMs with the total
- Original VM on Source Vmware Tenant side:
- 1649 GB = 1.649 TB
· The connection between the client and the DR is a 100 Mb.
Server1 1024 GB Digest 2% 20.48 Microsoft Windows Server 2012 (64-bit)
Server2 200 GB Digest 2% 4 Microsoft Windows Server 2012 (64-bit)
Server3 150 GB Digest 2% 3 Microsoft Windows Server 2016 (64-bit)
Server4 200 GB Digest 2% 4 Microsoft Windows Server 2016 (64-bit)
Server5 125 GB Digest 2% 2,5 Microsoft Windows Server 2012 (64-bit)
Server6 300 GB Digest 2% 6 Microsoft Windows Server 2012 (64-bit)
Digest Size = 2% of Provisioned VM Size 1649 GB Digest 2% 32,98 GB / (33 GB)
Many-to-One WAN Acceleration Scenario
Actual situation:
Total Global Cache Size = (# of Source WAN Accelerators) * (Size of Global Cache Configured in Target WAN Accelerator Properties) 3 wan side Tenant x 100 GB ( actual Wan cache SP “to much allocate space”) =300
Total Free Disk Space to Provide = Total Global Cache Size + Digest Size 300 + 33,58= 333,58
- The WAN WAN Low bandwidth mode requires much more disk space for the WAN cache.If you have an equal or higher bandwidth, you can switch to WAN high bandwidth mode, which saves a lot of space. Support advised me to switch to WAN HIG mode.
Sizing for High bandwidth mode = OR > 100 MB
I would advise to consider switching of WAN accelerators to High bandwidth mode.
- The main difference between the two modes is the use of the “global cache”. High bandwidth mode does not use it. Instead it uses advanced compression and deduplication mechanisms, including digest files, which are used also by low bandwidth mode. The disk space and compute requirements are lower for high-bandwidth mode in many cases.
In my case:
Total Global Cache Size = (3 of Source WAN Accelerators) * (Size of Global Cache Configured in Target WAN Accelerator Properties) 3 wan side Tenant x 50 GB ( actual Wan cache SP “to much allocate space”) =150
1679 GB 1%=16,79
Total Global Cache: Size 3 wan side Tenant x 50 GB = 150 GB
Total fee disk space to Provide: 150+16,79= 166,79
action plan:
- Disable Replication Jobs.
- Switch to Hig bandwith mode.
- VCC Clear WAN Cache .
- Removed xxx GB SP WAN cache disk.
- Added 170 GB SP WAN cache disk.
- Rescan Infra on both sites.
- Rescan SP tenant side.
- Enable job replication again.
- Start the first replication of the job
SP Proxy\WAN Cache used in High
In summary, if you have a 100Mb line, switch immediately to WAN Hig bandwit mode and you will save a lot of space.
