System requirements
Server
Exivity can be installed on any Microsoft Windows 2016 or higher server in your on-premises data center or in the cloud. Depending on the amount of data, Exivity recommends the following system configuration:
| Size | CUPR | Web & Backend | RabbitMQ | PSQL | CPU | RAM | Storage |
|---|---|---|---|---|---|---|---|
| Small | < 5000 | 1 – 2 node(s) | 0 – 1 node | 0 – 1 node | 4 cores | 12 GB | 200 GB |
| Medium | < 10000 | 1 – 2 node(s) | 0 – 1 node | 0 – 1 node | 6 cores | 16 GB | 400 GB |
| Large | < 15000 | 2 – 4 nodes | 1 instance* | 1 instance* | 8 cores | 24 GB | 600 GB |
| X-Large | > 15000 | 4 or more nodes** | 1 instance* | 1 instance* | 8 cores | 32 GB | TBD** |
* a managed high available and high-performance cluster would be recommended
** consult with Exivity for advisory on appropriate sizing of extra-large environments
Each node can be either a physical or virtual machine. The CPU and RAM requirements provided in the table above are defined per node whereas the Storage requirements are to be considered for the complete UMB deployment as a whole and should be hosted on shared storage (i.e. SMB/NFS). An operating system will need to be pre-installed on every node. Web and Backend nodes require a version of Windows 2016 Standard or higher to be installed. In scenarios where the RabbitMQ and/or PostgreSQL nodes are deployed on dedicated or clustered environments, the operating system guidelines specific to those applications are to be considered.
Client
The Exivity front-end supports the following desktop browsers:
- Google Chrome v59+
- Microsoft Edge v41+ (EdgeHTML 16+ / Blink 80+)
- Opera v46+
- Mozilla Firefox v65+ (support added in Exivity v2.10.0)
- Apple Safari v10.1+ (support added in Exivity v3.2.7)
We aim to provide the fastest metering and billing solution available today, and this means we have to rely on modern (web) technologies. Part of our speed comes from pre-processing the raw data, and part comes from having almost all processed data available right in the browser, and streaming the missing pieces on request.
To efficiently and reliably achieve this we use some very specific technologies not yet available in all browsers. When they do catch up, we'll fully support those browsers.