added readme to CI inputs

Author: liorsve

.github/workflows/benchmark_configs/README.md

@@ -0,0 +1,81 @@
+# Benchmark Configuration Guide
+
+This directory contains configuration files for the benchmark CI workflow.
+
+## CI Workflow Inputs
+
+### Predefined Configs
+
+| Input              | Description                          | Options                                                                                   |
+| ------------------ | ------------------------------------ | ----------------------------------------------------------------------------------------- |
+| `primary_driver`   | Client library to benchmark          | `spring-data-valkey`, `spring-data-redis`, `valkey-glide`, `jedis`, `lettuce`, `redisson` |
+| `secondary_driver` | Underlying driver for spring-data-\* | `valkey-glide`, `jedis`, `lettuce`, `none`                                                |
+| `topology`         | Server topology                      | `standalone`, `cluster`                                                                   |
+| `workload`         | Test scenario                        | `example-workload`, `example-workload-single-client`                                      |
+
+### Custom Configs
+
+For custom use cases, you can provide custom JSON configs directly:
+
+| Input                    | Description                                              |
+| ------------------------ | -------------------------------------------------------- |
+| `custom_driver_config`   | Custom driver JSON (overrides driver/topology selection) |
+| `custom_workload_config` | Custom workload JSON (overrides workload selection)      |
+
+### Version Inputs
+
+| Input               | Description                                                                                                 |
+| ------------------- | ----------------------------------------------------------------------------------------------------------- |
+| `primary_version`   | Version of primary driver. Leave empty for `spring-data-valkey` (uses branch HEAD)                          |
+| `secondary_version` | Version or commit ID of secondary driver. Commit IDs (7-40 hex chars) trigger source build for valkey-glide |
+| `job_id_prefix`     | Optional prefix for job ID (e.g., `nightly`, `pr-123`)                                                      |
+
+## How Config Selection Works
+
+The CI workflow maps your input selections to JSON config files using a naming convention.
+
+### Driver Config Resolution
+
+The workflow constructs the driver config filename from your inputs:
+
+| Primary Driver       | Secondary Driver | Topology     | Resolved File                                      |
+| -------------------- | ---------------- | ------------ | -------------------------------------------------- |
+| `valkey-glide`       | (ignored)        | `standalone` | `example-valkey-glide-standalone.json`             |
+| `valkey-glide`       | (ignored)        | `cluster`    | `example-valkey-glide-cluster.json`                |
+| `spring-data-valkey` | `valkey-glide`   | `standalone` | `example-spring-data-valkey-glide-standalone.json` |
+| `spring-data-valkey` | `valkey-glide`   | `cluster`    | `example-spring-data-valkey-glide-cluster.json`    |
+| `spring-data-valkey` | `jedis`          | `cluster`    | `example-spring-data-valkey-jedis-cluster.json`    |
+
+**Pattern:**
+
+- Standalone drivers: `example-{primary_driver}-{topology}.json`
+- Spring-data drivers: `example-{primary_driver}-{secondary_driver}-{topology}.json`
+  - Note: `valkey-glide` is shortened to `glide` in filenames
+
+### Workload Config Resolution
+
+The `workload` input maps directly to a file:
+
+- `example-workload` → `workloads/example-workload.json`
+- `example-workload-single-client` → `workloads/example-workload-single-client.json`
+
+### Custom Configs
+
+When you provide `custom_driver_config` or `custom_workload_config`, the workflow:
+
+1. Writes your JSON to a temp file
+2. Validates it with `jq`
+3. Uses it instead of the predefined config
+
+This allows testing configurations not in the predefined set.
+
+For the full specification of config formats to use when constructing custom configs, see the JSON schema files:
+
+- **Driver config schema:** [schemas/driver-config.schema.json](schemas/driver-config.schema.json)
+- **Workload config schema:** [schemas/workload-config.schema.json](schemas/workload-config.schema.json)
+
+### Key config rules:
+
+- `spring-data-valkey` and `spring-data-redis` require `secondary_driver_id`
+- `spring-data-redis` does NOT support `valkey-glide` as secondary driver
+- Standalone drivers (`jedis`, `lettuce`, `valkey-glide`, `redisson`) ignore `secondary_driver_id`