feat: add call for machines

Signed-off-by: Sunyanan Choochotkaew <[email protected]>

docs/sponsor/call-for-machines.md

@@ -0,0 +1,42 @@
+# Call for Machines
+
+As mentioned in several research works, the power consumption behavior can be varied by several factors.
+Using the right power model to predict power consumption on the machine that has no power meter is critical the precision of the reported values.
+
+We are seeking machines with the following characteristics, commonly used in cloud computing environments but not limited to:
+
+## Processor
+- Intel: Cascade Lake, Skylake (or newer)
+- Ampere: Altra, AmpereOne (or newer)
+- AMD: Opteron, EPYC (or newer)
+- Apple: M1, M2 (or newer)
+
+## Accelerator Chips
+- GPU: NVIDIA A100, V100, or other high-performance models for AI/ML workloads
+- TPU: Google Cloud TPU, suitable for large-scale AI training and inference
+- Custom Accelerators: AWS Inferentia, or other custom-built accelerators for optimized workloads
+
+## Power meters
+- intel_rapl
+- intel_powerclamp
+- amd-pstate
+- acpi
+- hmc
+- redfish
+- nvidia-smi
+- tpu-monitoring
+- amdgpu
+
+## Available machines on CI
+
+List of currently available bare metal machines for power model training and validation CI pipeline in [kepler-metal-ci](https://github.com/sustainable-computing-io/kepler-metal-ci/tree/main).
+
+|processor|cores|chips|memory|power meters|source|sponsor
+|---|---|---|----|---|---|---|
+|intel_xeon_e_2278g| 16 | 1 | - | RAPL (package, core, dram)|equinix c3.small.x86|CNCF
+|intel_xeon_platinum_8259cl| 96 | 2 | 377| RAPL (package, dram)|ec2 m5.metal|RedHat
+|intel_xeon_e5_2686v4| 72 | 2 | 503| RAPL (package, dram)|ec2 i3.metal|RedHat
+|intel_xeon_platinum_8275cl| 96 | 2 | 188| RAPL (package, dram)|ec2 c5.metal|RedHat
+|intel_xeon_platinum_8259cl| 96 | 2 | 755| RAPL (package, dram)|ec2 r5.metal|RedHat
+|intel_xeon_platinum_8252c| 48 | 2 | 188 | RAPL (package, dram)|ec2 m5zn.metal|RedHat
+|intel_xeon_platinum_8488c| 96 | 1 | 377 | RAPL (package, dram)|ec2 m7i.metal-24xl|RedHat