leetbit
Bench Clusteractive

A three-node cluster kept deliberately boring — identical machines on an isolated 10 GbE segment — used only for benchmarks that need real hardware and repeatable numbers rather than a shared workstation.

Since
2025-06
Location
Home lab · Rack B
Components
4
Last verified
Jul 10, 2026
Nodes
3 (identical)
Node OS
Debian 13 (trixie)
HardwareThree identical nodes; uniformity is the point.1
node-a / node-b / node-cCompute nodeactive×3
Identical benchmark nodes, reimaged between campaigns.
CPUIntel Xeon E-2388G (8C / 16T)
Memory64 GB DDR4-3200 ECC
Storage1 TB NVMe (single, no RAID)
NIC10 GbE (Intel X710)
Since2025-06
Network1
switch-01Switchactive
Single 10 GbE switch; the cluster's only fabric.
Ports8× 10 GbE SFP+
VLAN40 (isolated)
Software2
Benchmark harnessToolingactive
The same versioned harness used on the workbench, driven cluster-wide.
ProvisioningAutomationactive
PXE reimage to a known image before each campaign.

Some measurements do not belong on the Kernel Workbench: they need more than one machine, or they need a host with nothing else running on it. The Bench Cluster is that environment — three identical nodes that exist to produce numbers nobody has to caveat.

Topology#

Three nodes on a single 10 GbE switch, on their own VLAN. There is intentionally nothing clever here: no shared storage, no orchestration layer, no noisy neighbours. Uniformity is the feature.

Three identical bench nodes connected to a single 10 GbE switch on VLAN 40.
Physical topology — three identical nodes, one switch, one isolated segment.

How it is used#

A benchmark is staged from the workbench, pushed to the nodes, and run with the cluster otherwise idle. Results are pulled back and archived. Because the nodes are identical and reimaged between campaigns, a number measured on node-a should reproduce on node-b and node-c — and when it does not, that divergence is itself the finding.

Relations

where else this lives in the system
status activeupdated Jul 11, 2026verified Jul 10, 2026