Cluster Setup
For the Best Performance
In the cluster deployment mode, various modules are encapsulated as services and deployed across multiple machines. Each machine has different focuses based on its role. Machines responsible for running the execution service require more computational resources, including RAM, compared to those with less demanding tasks.
Cluster mode is generally more resource-intensive than the standalone deployment for the same task, due to the extra overhead associated with inter-process communication. However, A single machine can easily reach its resource limit, but a cluster rarely does.
Please consult Arcology team for further details on setting up a cluster.
Hardware Requirement
Minimum Requirement
The Minimum cluster deployment will require at least two machines. One for storage and RPC , one for all others.
RPC & Storage Service:
16GB RAM
8 CPU cores
NVMe SSD with 1T free space
10G Bandwidth
Other Services:
64GB RAM or more or
16 CPU cores
NVMe SSD with 500GB free space
10G Bandwidth
Optimized Configuration
For larger and more demanding deployments, where performance and scalability are critical, it’s important to note that if your deployment exceeds 32 servers, additional tuning and resource management may be necessary to maintain optimal performance.
Last updated