The following recommendations are for a shared services environment that provide multiple backup services using shared infrastructure on a new or existing Commcell. The backup services fall under the following categories:
-
Active Directory
-
VM and Kubernetes
-
Database
-
File and Object
-
Microsoft Office 365
-
Microsoft Dynamics 365
-
Salesforce
-
Cloud Storage
-
Endpoint
Recommendations for Initial Setup
CommServe
Small |
Medium |
Large |
Extra Large |
---|---|---|---|
Supports up to 25 servers, or 100 virtual machines, or 200 laptops in a single CommCell, including Hardware Specifications for Deduplication Mode. |
Supports up to 500 servers, or 1000 virtual machines, or 5000 laptops in a single CommCell environment. |
Supports up to 2500 servers, or 5000 virtual machines, or 10,000 laptops in a single CommCell environment. |
Supports up to 10,000 servers, or 20,000 virtual machines, or 50,000 laptops in a single CommCell environment. |
4 CPU cores |
8 CPU cores |
12 CPU cores |
16 CPU cores |
24 GB RAM |
32 GB RAM |
64 GB RAM |
128 GB RAM |
100 GB of disk space for CommServe database |
200 GB of disk space for CommServe database |
300 GB of SSD disk space for CommServe database |
500 GB of enterprise class SSD disk space for CommServe database |
The sizing above assumes the following Commvault packages are installed on the CommServe Server:
-
Server: CommServe, MediaAgent, Web Server, CommCell Console, Command Center, Workflow Engine, Metrics Server
-
Microsoft Windows: File System Core, File System, VSS Provider, VSS Hardware Provider
-
Virtualization: Virtual Server
-
Applications: Cloud Apps
-
Tools: High Availability Computing, Storage Accelerator, MongoDB, Message Queue
If the load on the Web Server increases beyond the recommended sizing guide limits by over 20%, sustained for a 7 day period, move the Web Server to a dedicated resource (physical computer or a virtual machine).
Web Server and Web Console
Initially, start with the Web Server and CommServe on the same computer. If the load on the Web Server increases beyond 20% of the recommended sizing guide limits, sustained for a 7 day period, install the Web Server and Web Console on a separate computer.
Small |
Medium |
Large |
---|---|---|
200 GB of disk space for cache |
500 GB of disk space for cache |
1 TB of disk space for cache |
8 CPU cores |
12 CPU cores |
16 CPU cores |
16 GB RAM |
32 GB RAM |
64 GB RAM |
2 million API requests in a 7 day period |
6 million API requests in a 7 day period |
12 million API requests in a 7 day period |
MediaAgent (Windows and Unix)
Small |
Medium |
Large |
Extra Large |
Notes |
---|---|---|---|---|
8 CPU cores |
16 CPU cores |
32 CPU cores |
48 CPU cores |
|
16 GB RAM |
32 GB RAM |
64 GB RAM |
128 GB RAM |
|
500 GB of DDB disk space |
1 TB of DDB disk space |
1.5 TB of DDB disk space |
3 TB of DDB disk space |
SSD mandated |
500 GB of index cache disk space |
1 TB of index cache disk space |
1.5 TB of index cache disk space |
2 TB of index cache disk space |
SSD recommended |
100 parallel streams |
250 parallel streams |
500 parallel streams |
1000 parallel streams |
Index Server (Windows and Unix)
Small |
Medium |
Large |
---|---|---|
8 CPU cores |
16 CPU cores |
32 CPU cores |
16 GB RAM |
32 GB RAM |
64 GB RAM |
1TB index disk space for the first 100M items |
3TB index disk space for the first 500M items |
6TB index disk space for up to 1 billion items |
Access Node (Windows)
Used for Microsoft Office 365, Microsoft Dynamics 365, and Active Directory applications.
Small |
Medium |
Large |
Extra Large |
Notes |
---|---|---|---|---|
8 CPU cores |
16 CPU cores |
32 CPU cores |
48 CPU cores |
|
16 GB RAM |
32 GB RAM |
64 GB RAM |
128 GB RAM |
|
1 TB disk space for storing job results for the first 1000 O365 users |
1 TB disk space for storing job results for the first 5000 O365 users |
2 TB disk space for storing job results for the first 10000 O365 users |
2 TB disk space for storing job results |
For shared job results, start with 1TB disk space and increase up to 3 TB. Add a new job result disk for every 100 client computers. |
Access Node (Linux)
Used for Azure Blob, Salesforce, VM and Kubernetes applications.
Small |
Medium |
Large |
Extra Large |
Notes |
---|---|---|---|---|
8 CPU cores |
16 CPU cores |
32 CPU cores |
48 CPU cores |
|
16 GB RAM |
32 GB RAM |
64 GB RAM |
128 GB RAM |
|
1 TB disk space for storing job results for the first 1000 VMs and Front-End Terabytes (FET) 25TB |
1 TB disk space for storing job results for the first 3000 VMs and Front-End Terabytes (FET) 50TB |
2 TB disk space for storing job results for up to 6000 VMs and Front-End Terabytes (FET) 100TB |
2 TB disk space for storing job results for up to 10000 VMs and Front-End Terabytes (FET) 120TB |
For shared job results, start with 1TB disk space and increase up to 3 TB. Add a new job result disk for every 100 client computers. |
Scaling Out Your Backup Infrastructure Horizontally or Vertically
When you add new regions to an existing configuration, scale up your backup infrastructure based on the increased workload on your backup environment. We ideally recommend to start with a small configuration, and then scale up to medium, large, and extra large configurations before you scale out and add new nodes. For hardware requirements for large and extra large environments, refer to the component sizing topics.
Note
If the new region already has an existing backup infrastructure running the Commvault software, evaluate if you can increase the hardware configuration on the existing computer and install a new instance of the software. For more information, see Multi-Instance Installations.