Add xray chart examples for keda hpa

examples/xray/xray-with-keda-hpa/README.md

@@ -0,0 +1,58 @@
+# Xray with KEDA Autoscaling
+
+This example shows how to deploy Xray with `fullSplit` (each microservice as its own Deployment) and override the sizing file's autoscaling to use [KEDA](https://keda.sh/) instead of the regular CPU/memory-based HPA.
+
+See the [values-xray-with-keda-hpa.yaml](values-xray-with-keda-hpa.yaml) for the configuration example.
+
+## Why KEDA fits Xray
+
+* Xray scales horizontally (more pods), not vertically (bigger pods).
+* Each Xray worker consumes a largely fixed/static number of messages at a time, so adding pods increases throughput roughly linearly with queue backlog.
+* Queue-based KEDA autoscaling matches how Xray actually works: spin up more replicas when RabbitMQ queues grow, and scale back down when the work drains.
+
+## How KEDA scales Xray pods
+
+* `splitXraytoSeparateDeployments.fullSplit: true` deploys each microservice as its own Deployment (`server`, `analysis`, `indexer`, …).
+* For each worker service with `autoscaling.keda.enabled: true`, the chart creates a KEDA `ScaledObject` that watches that service's RabbitMQ queues (`QueueLength` trigger; queue names/thresholds default from the chart's `values.yaml`).
+* When messages in a watched queue exceed the queue threshold, KEDA adds replicas up to `maxReplicas`; when queues drain, it scales back down toward `minReplicas`.
+* `server` is different: it has no RabbitMQ queue triggers, so KEDA scales it on CPU/memory instead.
+
+## Notes
+
+* KEDA must already be installed on the target cluster. See the [KEDA deployment docs](https://keda.sh/docs/2.10/deploy/).
+* This file only sets `autoscaling`/`keda` blocks; it does not set each microservice's `enabled` flag. Whether a given microservice (e.g. `sbom`, `aiscanner`, `jascontextual`, `jasexposures`) is actually deployed is a separate decision — enable it in your base/sizing values as needed. The KEDA config here applies once that service is enabled.
+* This file only overrides autoscaling; it does not set resource requests/limits. Apply it on top of a sizing file so containers still get correct resources.
+
+## Prod sizing reference (min/max replicas)
+
+The values below are JFrog's published prod sizing `minReplicas`/`maxReplicas` per tier, per microservice. [values-xray-with-keda-hpa.yaml](values-xray-with-keda-hpa.yaml) uses the `xlarge` row; pick a different tier's values here if that better matches your sizing file.
+
+| Service | xsmall | small | medium | large | xlarge | 2xlarge |
+|---|---|---|---|---|---|---|
+| server | 2–3 | 2–3 | 2–4 | 2–4 | 2–7 | 2–7 |
+| analysis | 1–3 | 1–3 | 1–4 | 1–4 | 1–7 | 1–7 |
+| indexer | 1–3 | 1–3 | 1–4 | 1–4 | 1–7 | 1–7 |
+| persist | 1–3 | 1–3 | 1–4 | 1–4 | 1–7 | 1–7 |
+| policyenforcer | 1–3 | 1–3 | 1–4 | 1–4 | 1–7 | 1–7 |
+| sbom | 1–3 | 1–3 | 1–4 | 1–4 | 1–7 | 1–7 |
+| aiscanner | 1–3 | 1–3 | 1–3 | 1–3 | 1–3 | 1–3 |
+| reporting | 1–3 | 1–3 | 1–4 | 1–4 | 1–7 | 1–7 |
+| jascontextual | 1–10 | 1–10 | 1–10 | 1–10 | 1–10 | 1–10 |
+| jasexposures | 1–10 | 1–10 | 1–10 | 1–10 | 1–10 | 1–10 |
+
+> Autoscaling is disabled by default on `xsmall` (except `server`); the ranges above are what applies if you turn it on.
+
+## Deploy
+
+Always apply a sizing file to provide the correct resources to the containers:
+
+```shell
+curl -fsSL https://raw.githubusercontent.com/jfrog/charts/master/stable/xray/sizing/xray-large.yaml \
+  -o xray-large.yaml
+```
+
+Install Xray with the following command:
+
+```shell
+helm upgrade --install xray jfrog/xray -f <your existing custom values> -f values-xray-with-keda-hpa.yaml -f xray-large.yaml
+```

examples/xray/xray-with-keda-hpa/values-xray-with-keda-hpa.yaml

@@ -0,0 +1,82 @@
+splitXraytoSeparateDeployments:
+  fullSplit: true
+
+server:
+  autoscaling:
+    enabled: true
+    minReplicas: 2
+    maxReplicas: 7
+    keda:
+      enabled: true
+
+analysis:
+  autoscaling:
+    enabled: true
+    minReplicas: 1
+    maxReplicas: 7
+    keda:
+      enabled: true
+
+indexer:
+  autoscaling:
+    enabled: true
+    minReplicas: 1
+    maxReplicas: 7
+    keda:
+      enabled: true
+
+persist:
+  autoscaling:
+    enabled: true
+    minReplicas: 1
+    maxReplicas: 7
+    keda:
+      enabled: true
+
+policyenforcer:
+  autoscaling:
+    enabled: true
+    minReplicas: 1
+    maxReplicas: 7
+    keda:
+      enabled: true
+
+sbom:
+  autoscaling:
+    enabled: true
+    minReplicas: 1
+    maxReplicas: 7
+    keda:
+      enabled: true
+
+aiscanner:
+  autoscaling:
+    enabled: true
+    minReplicas: 1
+    maxReplicas: 3
+    keda:
+      enabled: true
+
+reporting:
+  autoscaling:
+    enabled: true
+    minReplicas: 1
+    maxReplicas: 7
+    keda:
+      enabled: true
+
+jascontextual:
+  autoscaling:
+    enabled: true
+    minReplicas: 1
+    maxReplicas: 10
+    keda:
+      enabled: true
+
+jasexposures:
+  autoscaling:
+    enabled: true
+    minReplicas: 1
+    maxReplicas: 10
+    keda:
+      enabled: true