chore: delete obsolete top-level k8s/ and config/k8s/ manifests
Superseded by infra/k8s/ which has the current, actively maintained manifests (11 services, databases, GPU taints, tailscale, training). -1,899 lines (10 files) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
jgrusewski
@@ -1,464 +0,0 @@
|
||||
# Kubernetes Horizontal Pod Autoscaler (HPA) Manifests
|
||||
# Version: 1.0
|
||||
# Last Updated: 2025-10-07
|
||||
#
|
||||
# These HPA configurations provide:
|
||||
# - CPU-based autoscaling for all services
|
||||
# - Custom metrics for advanced scaling (request rate, queue depth)
|
||||
# - Configurable scale-up/scale-down behavior
|
||||
# - Min/max replica counts for cost optimization
|
||||
|
||||
---
|
||||
# ============================================================================
|
||||
# API GATEWAY HPA (Latency-Sensitive, Aggressive Scaling)
|
||||
# ============================================================================
|
||||
|
||||
apiVersion: autoscaling/v2
|
||||
kind: HorizontalPodAutoscaler
|
||||
metadata:
|
||||
name: api-gateway-hpa
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: api-gateway
|
||||
component: gateway
|
||||
spec:
|
||||
scaleTargetRef:
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
name: api-gateway
|
||||
|
||||
# Replica limits
|
||||
minReplicas: 3 # Always maintain 3 for HA (survive 1 node failure + 1 rolling update)
|
||||
maxReplicas: 20 # Max capacity: 20 instances × 2K req/s = 40K req/s
|
||||
|
||||
# Scaling metrics
|
||||
metrics:
|
||||
# CPU-based scaling (primary metric)
|
||||
- type: Resource
|
||||
resource:
|
||||
name: cpu
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 70 # Scale when CPU > 70%
|
||||
|
||||
# Memory-based scaling (secondary metric)
|
||||
- type: Resource
|
||||
resource:
|
||||
name: memory
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 80 # Scale when memory > 80%
|
||||
|
||||
# Custom metric: Request rate (requires Prometheus adapter)
|
||||
- type: Pods
|
||||
pods:
|
||||
metric:
|
||||
name: grpc_requests_per_second
|
||||
target:
|
||||
type: AverageValue
|
||||
averageValue: "2000" # Scale when > 2000 req/s per pod
|
||||
|
||||
# Scaling behavior (aggressive scale-up, slow scale-down)
|
||||
behavior:
|
||||
scaleUp:
|
||||
stabilizationWindowSeconds: 30 # Wait 30s before scaling up
|
||||
policies:
|
||||
- type: Percent
|
||||
value: 100 # Double replicas
|
||||
periodSeconds: 15
|
||||
- type: Pods
|
||||
value: 4 # Or add 4 pods
|
||||
periodSeconds: 15
|
||||
selectPolicy: Max # Use whichever scales faster
|
||||
|
||||
scaleDown:
|
||||
stabilizationWindowSeconds: 300 # Wait 5 minutes before scaling down
|
||||
policies:
|
||||
- type: Percent
|
||||
value: 25 # Remove 25% of replicas
|
||||
periodSeconds: 60
|
||||
- type: Pods
|
||||
value: 1 # Or remove 1 pod
|
||||
periodSeconds: 60
|
||||
selectPolicy: Min # Use whichever scales slower
|
||||
|
||||
---
|
||||
# ============================================================================
|
||||
# TRADING SERVICE HPA (Latency-Sensitive, Aggressive Scaling)
|
||||
# ============================================================================
|
||||
|
||||
apiVersion: autoscaling/v2
|
||||
kind: HorizontalPodAutoscaler
|
||||
metadata:
|
||||
name: trading-service-hpa
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: trading-service
|
||||
component: backend
|
||||
spec:
|
||||
scaleTargetRef:
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
name: trading-service
|
||||
|
||||
# Replica limits
|
||||
minReplicas: 3 # HA requirement
|
||||
maxReplicas: 20 # Based on performance testing: 50K orders/sec ÷ 2.5K orders/sec per pod
|
||||
|
||||
# Scaling metrics
|
||||
metrics:
|
||||
# CPU-based scaling
|
||||
- type: Resource
|
||||
resource:
|
||||
name: cpu
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 70
|
||||
|
||||
# Memory-based scaling
|
||||
- type: Resource
|
||||
resource:
|
||||
name: memory
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 80
|
||||
|
||||
# Custom metric: Order submission rate
|
||||
- type: Pods
|
||||
pods:
|
||||
metric:
|
||||
name: order_submissions_per_second
|
||||
target:
|
||||
type: AverageValue
|
||||
averageValue: "2500" # Scale when > 2500 orders/s per pod
|
||||
|
||||
# Scaling behavior (aggressive scale-up)
|
||||
behavior:
|
||||
scaleUp:
|
||||
stabilizationWindowSeconds: 30
|
||||
policies:
|
||||
- type: Percent
|
||||
value: 100
|
||||
periodSeconds: 15
|
||||
- type: Pods
|
||||
value: 4
|
||||
periodSeconds: 15
|
||||
selectPolicy: Max
|
||||
|
||||
scaleDown:
|
||||
stabilizationWindowSeconds: 300
|
||||
policies:
|
||||
- type: Percent
|
||||
value: 25
|
||||
periodSeconds: 60
|
||||
- type: Pods
|
||||
value: 1
|
||||
periodSeconds: 60
|
||||
selectPolicy: Min
|
||||
|
||||
---
|
||||
# ============================================================================
|
||||
# BACKTESTING SERVICE HPA (Job Queue Pattern, Conservative Scaling)
|
||||
# ============================================================================
|
||||
|
||||
apiVersion: autoscaling/v2
|
||||
kind: HorizontalPodAutoscaler
|
||||
metadata:
|
||||
name: backtesting-service-hpa
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: backtesting-service
|
||||
component: backend
|
||||
spec:
|
||||
scaleTargetRef:
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
name: backtesting-service
|
||||
|
||||
# Replica limits
|
||||
minReplicas: 2 # Minimum workers for job queue
|
||||
maxReplicas: 10 # Rarely need more than 10 concurrent backtests
|
||||
|
||||
# Scaling metrics
|
||||
metrics:
|
||||
# CPU-based scaling (primary for CPU-intensive backtests)
|
||||
- type: Resource
|
||||
resource:
|
||||
name: cpu
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 75 # Higher threshold for batch workload
|
||||
|
||||
# Memory-based scaling
|
||||
- type: Resource
|
||||
resource:
|
||||
name: memory
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 80
|
||||
|
||||
# Custom metric: Redis job queue depth
|
||||
- type: Pods
|
||||
pods:
|
||||
metric:
|
||||
name: redis_queue_length_backtesting
|
||||
target:
|
||||
type: AverageValue
|
||||
averageValue: "5" # Scale when > 5 jobs per worker
|
||||
|
||||
# Scaling behavior (conservative scale-up, slower scale-down)
|
||||
behavior:
|
||||
scaleUp:
|
||||
stabilizationWindowSeconds: 60 # Wait 60s before scaling up
|
||||
policies:
|
||||
- type: Percent
|
||||
value: 50 # Add 50% more replicas
|
||||
periodSeconds: 60
|
||||
- type: Pods
|
||||
value: 2 # Or add 2 pods
|
||||
periodSeconds: 60
|
||||
selectPolicy: Min # Use whichever scales slower
|
||||
|
||||
scaleDown:
|
||||
stabilizationWindowSeconds: 600 # Wait 10 minutes before scaling down
|
||||
policies:
|
||||
- type: Percent
|
||||
value: 25
|
||||
periodSeconds: 120
|
||||
- type: Pods
|
||||
value: 1
|
||||
periodSeconds: 120
|
||||
selectPolicy: Min
|
||||
|
||||
---
|
||||
# ============================================================================
|
||||
# ML TRAINING SERVICE HPA (MANUAL SCALING RECOMMENDED)
|
||||
# ============================================================================
|
||||
|
||||
# NOTE: ML Training Service uses GPU resources, which are expensive and slow to provision.
|
||||
# Recommendation: Use manual scaling (kubectl scale) or scheduled scaling, NOT auto-scaling.
|
||||
#
|
||||
# If auto-scaling is required, use job queue pattern with custom metrics:
|
||||
|
||||
apiVersion: autoscaling/v2
|
||||
kind: HorizontalPodAutoscaler
|
||||
metadata:
|
||||
name: ml-training-service-hpa
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: ml-training-service
|
||||
component: backend
|
||||
spec:
|
||||
scaleTargetRef:
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
name: ml-training-service
|
||||
|
||||
# Replica limits (GPU resources expensive)
|
||||
minReplicas: 1 # Single instance (GPU idle time acceptable)
|
||||
maxReplicas: 3 # Limited by GPU cluster size
|
||||
|
||||
# Scaling metrics
|
||||
metrics:
|
||||
# GPU utilization (requires NVIDIA DCGM exporter)
|
||||
- type: Pods
|
||||
pods:
|
||||
metric:
|
||||
name: dcgm_gpu_utilization
|
||||
target:
|
||||
type: AverageValue
|
||||
averageValue: "80" # Scale when GPU > 80% utilized
|
||||
|
||||
# Custom metric: Training job queue depth
|
||||
- type: Pods
|
||||
pods:
|
||||
metric:
|
||||
name: redis_queue_length_ml_training
|
||||
target:
|
||||
type: AverageValue
|
||||
averageValue: "2" # Scale when > 2 training jobs queued
|
||||
|
||||
# Scaling behavior (very conservative, GPU instances expensive)
|
||||
behavior:
|
||||
scaleUp:
|
||||
stabilizationWindowSeconds: 300 # Wait 5 minutes before scaling up
|
||||
policies:
|
||||
- type: Pods
|
||||
value: 1 # Add 1 pod at a time
|
||||
periodSeconds: 300
|
||||
selectPolicy: Min
|
||||
|
||||
scaleDown:
|
||||
stabilizationWindowSeconds: 1800 # Wait 30 minutes before scaling down
|
||||
policies:
|
||||
- type: Pods
|
||||
value: 1
|
||||
periodSeconds: 300
|
||||
selectPolicy: Min
|
||||
|
||||
---
|
||||
# ============================================================================
|
||||
# CUSTOM METRICS (Prometheus Adapter Configuration)
|
||||
# ============================================================================
|
||||
|
||||
# To use custom metrics (grpc_requests_per_second, redis_queue_length, etc.),
|
||||
# you must install and configure Prometheus Adapter:
|
||||
#
|
||||
# 1. Install Prometheus Adapter:
|
||||
# helm install prometheus-adapter prometheus-community/prometheus-adapter \
|
||||
# --namespace monitoring \
|
||||
# --values prometheus-adapter-values.yaml
|
||||
#
|
||||
# 2. Configure custom metrics (prometheus-adapter-values.yaml):
|
||||
|
||||
# prometheus:
|
||||
# url: http://prometheus-server.monitoring.svc
|
||||
# port: 9090
|
||||
#
|
||||
# rules:
|
||||
# default: false
|
||||
# custom:
|
||||
# # gRPC request rate per pod
|
||||
# - seriesQuery: 'grpc_requests_total{namespace="foxhunt"}'
|
||||
# resources:
|
||||
# overrides:
|
||||
# namespace: {resource: "namespace"}
|
||||
# pod: {resource: "pod"}
|
||||
# name:
|
||||
# matches: "^grpc_requests_total"
|
||||
# as: "grpc_requests_per_second"
|
||||
# metricsQuery: 'sum(rate(grpc_requests_total{<<.LabelMatchers>>}[1m])) by (<<.GroupBy>>)'
|
||||
#
|
||||
# # Redis queue length
|
||||
# - seriesQuery: 'redis_list_length{namespace="foxhunt",queue="backtesting"}'
|
||||
# resources:
|
||||
# overrides:
|
||||
# namespace: {resource: "namespace"}
|
||||
# name:
|
||||
# matches: "^redis_list_length"
|
||||
# as: "redis_queue_length_backtesting"
|
||||
# metricsQuery: 'avg(redis_list_length{<<.LabelMatchers>>}) by (<<.GroupBy>>)'
|
||||
#
|
||||
# # GPU utilization (requires NVIDIA DCGM exporter)
|
||||
# - seriesQuery: 'DCGM_FI_DEV_GPU_UTIL{namespace="foxhunt"}'
|
||||
# resources:
|
||||
# overrides:
|
||||
# namespace: {resource: "namespace"}
|
||||
# pod: {resource: "pod"}
|
||||
# name:
|
||||
# matches: "^DCGM_FI_DEV_GPU_UTIL"
|
||||
# as: "dcgm_gpu_utilization"
|
||||
# metricsQuery: 'avg(DCGM_FI_DEV_GPU_UTIL{<<.LabelMatchers>>}) by (<<.GroupBy>>)'
|
||||
|
||||
# 3. Verify custom metrics are available:
|
||||
# kubectl get --raw "/apis/custom.metrics.k8s.io/v1beta1" | jq .
|
||||
|
||||
# 4. Test HPA with custom metrics:
|
||||
# kubectl get hpa api-gateway-hpa -o yaml
|
||||
# # Check "currentMetrics" section for custom metric values
|
||||
|
||||
---
|
||||
# ============================================================================
|
||||
# TESTING & VALIDATION
|
||||
# ============================================================================
|
||||
|
||||
# 1. Deploy HPA manifests:
|
||||
# kubectl apply -f hpa.yaml
|
||||
#
|
||||
# 2. Verify HPA status:
|
||||
# kubectl get hpa -n foxhunt
|
||||
# kubectl describe hpa api-gateway-hpa -n foxhunt
|
||||
#
|
||||
# 3. Monitor scaling events:
|
||||
# kubectl get events -n foxhunt --field-selector involvedObject.kind=HorizontalPodAutoscaler
|
||||
#
|
||||
# 4. Load test to trigger scaling:
|
||||
# # Generate load (e.g., with ghz)
|
||||
# ghz --insecure \
|
||||
# --proto api_gateway.proto \
|
||||
# --call foxhunt.api_gateway.ApiGateway/SubmitOrder \
|
||||
# --rps 10000 \
|
||||
# --duration 5m \
|
||||
# localhost:50051
|
||||
#
|
||||
# # Watch HPA scale up
|
||||
# watch kubectl get hpa -n foxhunt
|
||||
#
|
||||
# 5. Verify scale-down after load stops:
|
||||
# # Wait 5-10 minutes after load test ends
|
||||
# kubectl get hpa api-gateway-hpa -n foxhunt
|
||||
# # Should scale down to minReplicas
|
||||
|
||||
---
|
||||
# ============================================================================
|
||||
# TROUBLESHOOTING
|
||||
# ============================================================================
|
||||
|
||||
# Issue: HPA shows "unknown" for metrics
|
||||
# Solution: Ensure metrics-server is installed and running
|
||||
# kubectl get deployment metrics-server -n kube-system
|
||||
# kubectl logs -n kube-system deployment/metrics-server
|
||||
#
|
||||
# Issue: Custom metrics not available
|
||||
# Solution: Verify Prometheus Adapter configuration
|
||||
# kubectl get --raw "/apis/custom.metrics.k8s.io/v1beta1" | jq .
|
||||
# kubectl logs -n monitoring deployment/prometheus-adapter
|
||||
#
|
||||
# Issue: HPA not scaling up despite high CPU
|
||||
# Solution: Check stabilizationWindowSeconds (may be waiting)
|
||||
# kubectl describe hpa <name> -n foxhunt
|
||||
# # Look for "ScalingLimited" or "BackoffBoth" events
|
||||
#
|
||||
# Issue: HPA scaling too aggressively (flapping)
|
||||
# Solution: Increase stabilizationWindowSeconds or reduce scale-up policies
|
||||
#
|
||||
# Issue: Pods evicted due to resource pressure
|
||||
# Solution: Increase resource requests/limits in Deployment spec
|
||||
# kubectl describe pod <name> -n foxhunt
|
||||
# # Look for "Evicted" status
|
||||
|
||||
---
|
||||
# ============================================================================
|
||||
# PRODUCTION BEST PRACTICES
|
||||
# ============================================================================
|
||||
|
||||
# 1. Monitor HPA metrics:
|
||||
# - Current replicas vs desired replicas
|
||||
# - Scaling events (up/down)
|
||||
# - Metric values (CPU, memory, custom)
|
||||
#
|
||||
# 2. Alert on scaling issues:
|
||||
# - HPA unable to scale (resource quotas exceeded)
|
||||
# - HPA scaling too frequently (flapping)
|
||||
# - Pods in CrashLoopBackOff after scale-up
|
||||
#
|
||||
# 3. Test scaling under load:
|
||||
# - Run load tests regularly (weekly/monthly)
|
||||
# - Verify scale-up happens within SLA (< 2 minutes)
|
||||
# - Verify scale-down doesn't cause service disruption
|
||||
#
|
||||
# 4. Cost optimization:
|
||||
# - Use Reserved Instances for minReplicas (50% savings)
|
||||
# - Use Spot Instances for auto-scaled replicas (70% savings, non-critical workloads)
|
||||
# - Monitor auto-scaling costs (CloudWatch/Prometheus)
|
||||
#
|
||||
# 5. Resource limits:
|
||||
# - Set resource requests = expected usage
|
||||
# - Set resource limits = 2x requests (headroom for bursts)
|
||||
# - Monitor actual usage and adjust
|
||||
|
||||
---
|
||||
# ============================================================================
|
||||
# PRODUCTION CHECKLIST
|
||||
# ============================================================================
|
||||
|
||||
# [ ] Install metrics-server (for CPU/memory metrics)
|
||||
# [ ] Install Prometheus Adapter (for custom metrics)
|
||||
# [ ] Configure custom metrics (request rate, queue depth, GPU util)
|
||||
# [ ] Deploy HPA manifests (kubectl apply -f hpa.yaml)
|
||||
# [ ] Verify HPA status (kubectl get hpa)
|
||||
# [ ] Run load test to validate scaling behavior
|
||||
# [ ] Configure Grafana dashboards (HPA metrics, scaling events)
|
||||
# [ ] Configure alerts (scaling issues, resource pressure)
|
||||
# [ ] Document scaling thresholds and replica limits
|
||||
# [ ] Review and adjust HPA configuration quarterly
|
||||
@@ -1,137 +0,0 @@
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
metadata:
|
||||
name: api-gateway
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: api-gateway
|
||||
component: gateway
|
||||
tier: frontend
|
||||
version: v1
|
||||
spec:
|
||||
replicas: 3
|
||||
revisionHistoryLimit: 10
|
||||
strategy:
|
||||
type: RollingUpdate
|
||||
rollingUpdate:
|
||||
maxUnavailable: 1
|
||||
maxSurge: 1
|
||||
selector:
|
||||
matchLabels:
|
||||
app: api-gateway
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
app: api-gateway
|
||||
component: gateway
|
||||
tier: frontend
|
||||
version: v1
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9091"
|
||||
prometheus.io/path: "/metrics"
|
||||
spec:
|
||||
serviceAccountName: foxhunt-api-gateway
|
||||
securityContext:
|
||||
runAsNonRoot: true
|
||||
runAsUser: 1000
|
||||
runAsGroup: 1000
|
||||
fsGroup: 1000
|
||||
containers:
|
||||
- name: api-gateway
|
||||
image: foxhunt/api-gateway:latest
|
||||
imagePullPolicy: Always
|
||||
ports:
|
||||
- name: grpc
|
||||
containerPort: 50051
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
containerPort: 9091
|
||||
protocol: TCP
|
||||
- name: health
|
||||
containerPort: 8080
|
||||
protocol: TCP
|
||||
env:
|
||||
- name: API_GATEWAY_HOST
|
||||
value: "0.0.0.0"
|
||||
- name: API_GATEWAY_PORT
|
||||
value: "50051"
|
||||
- name: METRICS_PORT
|
||||
value: "9091"
|
||||
- name: JWT_SECRET
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: jwt-secret
|
||||
- name: DATABASE_URL
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: database-url
|
||||
- name: REDIS_URL
|
||||
valueFrom:
|
||||
configMapKeyRef:
|
||||
name: foxhunt-config
|
||||
key: redis-url
|
||||
- name: RUST_LOG
|
||||
value: "info,api_gateway=debug"
|
||||
- name: RUST_BACKTRACE
|
||||
value: "1"
|
||||
resources:
|
||||
requests:
|
||||
cpu: "1000m"
|
||||
memory: "512Mi"
|
||||
limits:
|
||||
cpu: "2000m"
|
||||
memory: "1Gi"
|
||||
livenessProbe:
|
||||
httpGet:
|
||||
path: /health/liveness
|
||||
port: 8080
|
||||
initialDelaySeconds: 30
|
||||
periodSeconds: 10
|
||||
timeoutSeconds: 5
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
readinessProbe:
|
||||
httpGet:
|
||||
path: /health/readiness
|
||||
port: 8080
|
||||
initialDelaySeconds: 10
|
||||
periodSeconds: 5
|
||||
timeoutSeconds: 3
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
startupProbe:
|
||||
httpGet:
|
||||
path: /health/startup
|
||||
port: 8080
|
||||
initialDelaySeconds: 0
|
||||
periodSeconds: 5
|
||||
timeoutSeconds: 3
|
||||
successThreshold: 1
|
||||
failureThreshold: 30
|
||||
volumeMounts:
|
||||
- name: config
|
||||
mountPath: /app/config
|
||||
readOnly: true
|
||||
- name: logs
|
||||
mountPath: /app/logs
|
||||
volumes:
|
||||
- name: config
|
||||
configMap:
|
||||
name: api-gateway-config
|
||||
- name: logs
|
||||
emptyDir: {}
|
||||
affinity:
|
||||
podAntiAffinity:
|
||||
preferredDuringSchedulingIgnoredDuringExecution:
|
||||
- weight: 100
|
||||
podAffinityTerm:
|
||||
labelSelector:
|
||||
matchExpressions:
|
||||
- key: app
|
||||
operator: In
|
||||
values:
|
||||
- api-gateway
|
||||
topologyKey: kubernetes.io/hostname
|
||||
@@ -1,143 +0,0 @@
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
metadata:
|
||||
name: backtesting-service
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: backtesting-service
|
||||
component: backtesting
|
||||
tier: backend
|
||||
version: v1
|
||||
spec:
|
||||
replicas: 2
|
||||
revisionHistoryLimit: 10
|
||||
strategy:
|
||||
type: RollingUpdate
|
||||
rollingUpdate:
|
||||
maxUnavailable: 1
|
||||
maxSurge: 1
|
||||
selector:
|
||||
matchLabels:
|
||||
app: backtesting-service
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
app: backtesting-service
|
||||
component: backtesting
|
||||
tier: backend
|
||||
version: v1
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9093"
|
||||
prometheus.io/path: "/metrics"
|
||||
spec:
|
||||
serviceAccountName: foxhunt-backtesting-service
|
||||
securityContext:
|
||||
runAsNonRoot: true
|
||||
runAsUser: 1000
|
||||
runAsGroup: 1000
|
||||
fsGroup: 1000
|
||||
containers:
|
||||
- name: backtesting-service
|
||||
image: foxhunt/backtesting-service:latest
|
||||
imagePullPolicy: Always
|
||||
ports:
|
||||
- name: grpc
|
||||
containerPort: 50053
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
containerPort: 9093
|
||||
protocol: TCP
|
||||
- name: health
|
||||
containerPort: 8082
|
||||
protocol: TCP
|
||||
env:
|
||||
- name: BACKTESTING_SERVICE_HOST
|
||||
value: "0.0.0.0"
|
||||
- name: BACKTESTING_SERVICE_PORT
|
||||
value: "50053"
|
||||
- name: METRICS_PORT
|
||||
value: "9093"
|
||||
- name: DATABASE_URL
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: database-url
|
||||
- name: REDIS_URL
|
||||
valueFrom:
|
||||
configMapKeyRef:
|
||||
name: foxhunt-config
|
||||
key: redis-url
|
||||
- name: VAULT_ADDR
|
||||
valueFrom:
|
||||
configMapKeyRef:
|
||||
name: foxhunt-config
|
||||
key: vault-addr
|
||||
- name: VAULT_TOKEN
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: vault-token
|
||||
- name: RUST_LOG
|
||||
value: "info,backtesting_service=debug"
|
||||
- name: RUST_BACKTRACE
|
||||
value: "1"
|
||||
resources:
|
||||
requests:
|
||||
cpu: "1000m"
|
||||
memory: "1Gi"
|
||||
limits:
|
||||
cpu: "2000m"
|
||||
memory: "2Gi"
|
||||
livenessProbe:
|
||||
httpGet:
|
||||
path: /health
|
||||
port: 8082
|
||||
initialDelaySeconds: 30
|
||||
periodSeconds: 10
|
||||
timeoutSeconds: 5
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
readinessProbe:
|
||||
httpGet:
|
||||
path: /ready
|
||||
port: 8082
|
||||
initialDelaySeconds: 10
|
||||
periodSeconds: 5
|
||||
timeoutSeconds: 3
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
volumeMounts:
|
||||
- name: config
|
||||
mountPath: /app/config
|
||||
readOnly: true
|
||||
- name: logs
|
||||
mountPath: /app/logs
|
||||
- name: data
|
||||
mountPath: /app/data
|
||||
- name: results
|
||||
mountPath: /app/results
|
||||
volumes:
|
||||
- name: config
|
||||
configMap:
|
||||
name: backtesting-service-config
|
||||
- name: logs
|
||||
emptyDir: {}
|
||||
- name: data
|
||||
persistentVolumeClaim:
|
||||
claimName: backtesting-data-pvc
|
||||
- name: results
|
||||
persistentVolumeClaim:
|
||||
claimName: backtesting-results-pvc
|
||||
affinity:
|
||||
podAntiAffinity:
|
||||
preferredDuringSchedulingIgnoredDuringExecution:
|
||||
- weight: 100
|
||||
podAffinityTerm:
|
||||
labelSelector:
|
||||
matchExpressions:
|
||||
- key: app
|
||||
operator: In
|
||||
values:
|
||||
- backtesting-service
|
||||
topologyKey: kubernetes.io/hostname
|
||||
@@ -1,153 +0,0 @@
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
metadata:
|
||||
name: ml-training-service
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: ml-training-service
|
||||
component: ml-training
|
||||
tier: backend
|
||||
version: v1
|
||||
spec:
|
||||
replicas: 1
|
||||
revisionHistoryLimit: 10
|
||||
strategy:
|
||||
type: RollingUpdate
|
||||
rollingUpdate:
|
||||
maxUnavailable: 1
|
||||
maxSurge: 1
|
||||
selector:
|
||||
matchLabels:
|
||||
app: ml-training-service
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
app: ml-training-service
|
||||
component: ml-training
|
||||
tier: backend
|
||||
version: v1
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9094"
|
||||
prometheus.io/path: "/metrics"
|
||||
spec:
|
||||
serviceAccountName: foxhunt-ml-training-service
|
||||
securityContext:
|
||||
runAsNonRoot: true
|
||||
runAsUser: 1000
|
||||
runAsGroup: 1000
|
||||
fsGroup: 1000
|
||||
containers:
|
||||
- name: ml-training-service
|
||||
image: foxhunt/ml-training-service:latest
|
||||
imagePullPolicy: Always
|
||||
ports:
|
||||
- name: grpc
|
||||
containerPort: 50054
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
containerPort: 9094
|
||||
protocol: TCP
|
||||
- name: health
|
||||
containerPort: 8095
|
||||
protocol: TCP
|
||||
env:
|
||||
- name: ML_TRAINING_SERVICE_HOST
|
||||
value: "0.0.0.0"
|
||||
- name: ML_TRAINING_SERVICE_PORT
|
||||
value: "50054"
|
||||
- name: METRICS_PORT
|
||||
value: "9094"
|
||||
- name: DATABASE_URL
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: database-url
|
||||
- name: REDIS_URL
|
||||
valueFrom:
|
||||
configMapKeyRef:
|
||||
name: foxhunt-config
|
||||
key: redis-url
|
||||
- name: VAULT_ADDR
|
||||
valueFrom:
|
||||
configMapKeyRef:
|
||||
name: foxhunt-config
|
||||
key: vault-addr
|
||||
- name: VAULT_TOKEN
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: vault-token
|
||||
- name: AWS_ACCESS_KEY_ID
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: aws-access-key-id
|
||||
- name: AWS_SECRET_ACCESS_KEY
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: aws-secret-access-key
|
||||
- name: AWS_REGION
|
||||
value: "us-east-1"
|
||||
- name: S3_MODEL_BUCKET
|
||||
value: "foxhunt-models"
|
||||
- name: RUST_LOG
|
||||
value: "info,ml_training_service=debug"
|
||||
- name: RUST_BACKTRACE
|
||||
value: "1"
|
||||
resources:
|
||||
requests:
|
||||
cpu: "2000m"
|
||||
memory: "4Gi"
|
||||
nvidia.com/gpu: 1
|
||||
limits:
|
||||
cpu: "4000m"
|
||||
memory: "8Gi"
|
||||
nvidia.com/gpu: 1
|
||||
livenessProbe:
|
||||
httpGet:
|
||||
path: /health
|
||||
port: 8095
|
||||
initialDelaySeconds: 30
|
||||
periodSeconds: 10
|
||||
timeoutSeconds: 5
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
readinessProbe:
|
||||
httpGet:
|
||||
path: /ready
|
||||
port: 8095
|
||||
initialDelaySeconds: 10
|
||||
periodSeconds: 5
|
||||
timeoutSeconds: 3
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
volumeMounts:
|
||||
- name: config
|
||||
mountPath: /app/config
|
||||
readOnly: true
|
||||
- name: logs
|
||||
mountPath: /app/logs
|
||||
- name: models
|
||||
mountPath: /app/models
|
||||
- name: checkpoints
|
||||
mountPath: /app/checkpoints
|
||||
volumes:
|
||||
- name: config
|
||||
configMap:
|
||||
name: ml-training-service-config
|
||||
- name: logs
|
||||
emptyDir: {}
|
||||
- name: models
|
||||
persistentVolumeClaim:
|
||||
claimName: ml-models-pvc
|
||||
- name: checkpoints
|
||||
persistentVolumeClaim:
|
||||
claimName: ml-checkpoints-pvc
|
||||
nodeSelector:
|
||||
nvidia.com/gpu: "true"
|
||||
tolerations:
|
||||
- key: nvidia.com/gpu
|
||||
operator: Exists
|
||||
effect: NoSchedule
|
||||
@@ -1,137 +0,0 @@
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
metadata:
|
||||
name: trading-agent-service
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: trading-agent-service
|
||||
component: trading-agent
|
||||
tier: backend
|
||||
version: v1
|
||||
spec:
|
||||
replicas: 2
|
||||
revisionHistoryLimit: 10
|
||||
strategy:
|
||||
type: RollingUpdate
|
||||
rollingUpdate:
|
||||
maxUnavailable: 1
|
||||
maxSurge: 1
|
||||
selector:
|
||||
matchLabels:
|
||||
app: trading-agent-service
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
app: trading-agent-service
|
||||
component: trading-agent
|
||||
tier: backend
|
||||
version: v1
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9095"
|
||||
prometheus.io/path: "/metrics"
|
||||
spec:
|
||||
serviceAccountName: foxhunt-trading-agent-service
|
||||
securityContext:
|
||||
runAsNonRoot: true
|
||||
runAsUser: 1000
|
||||
runAsGroup: 1000
|
||||
fsGroup: 1000
|
||||
containers:
|
||||
- name: trading-agent-service
|
||||
image: foxhunt/trading-agent-service:latest
|
||||
imagePullPolicy: Always
|
||||
ports:
|
||||
- name: grpc
|
||||
containerPort: 50055
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
containerPort: 9095
|
||||
protocol: TCP
|
||||
- name: health
|
||||
containerPort: 8083
|
||||
protocol: TCP
|
||||
env:
|
||||
- name: TRADING_AGENT_SERVICE_HOST
|
||||
value: "0.0.0.0"
|
||||
- name: TRADING_AGENT_SERVICE_PORT
|
||||
value: "50055"
|
||||
- name: METRICS_PORT
|
||||
value: "9095"
|
||||
- name: DATABASE_URL
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: database-url
|
||||
- name: REDIS_URL
|
||||
valueFrom:
|
||||
configMapKeyRef:
|
||||
name: foxhunt-config
|
||||
key: redis-url
|
||||
- name: VAULT_ADDR
|
||||
valueFrom:
|
||||
configMapKeyRef:
|
||||
name: foxhunt-config
|
||||
key: vault-addr
|
||||
- name: VAULT_TOKEN
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: vault-token
|
||||
- name: RUST_LOG
|
||||
value: "info,trading_agent_service=debug"
|
||||
- name: RUST_BACKTRACE
|
||||
value: "1"
|
||||
resources:
|
||||
requests:
|
||||
cpu: "1000m"
|
||||
memory: "1Gi"
|
||||
limits:
|
||||
cpu: "2000m"
|
||||
memory: "2Gi"
|
||||
livenessProbe:
|
||||
httpGet:
|
||||
path: /health
|
||||
port: 8083
|
||||
initialDelaySeconds: 30
|
||||
periodSeconds: 10
|
||||
timeoutSeconds: 5
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
readinessProbe:
|
||||
httpGet:
|
||||
path: /ready
|
||||
port: 8083
|
||||
initialDelaySeconds: 10
|
||||
periodSeconds: 5
|
||||
timeoutSeconds: 3
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
volumeMounts:
|
||||
- name: config
|
||||
mountPath: /app/config
|
||||
readOnly: true
|
||||
- name: logs
|
||||
mountPath: /app/logs
|
||||
- name: data
|
||||
mountPath: /app/data
|
||||
volumes:
|
||||
- name: config
|
||||
configMap:
|
||||
name: trading-agent-service-config
|
||||
- name: logs
|
||||
emptyDir: {}
|
||||
- name: data
|
||||
emptyDir: {}
|
||||
affinity:
|
||||
podAntiAffinity:
|
||||
preferredDuringSchedulingIgnoredDuringExecution:
|
||||
- weight: 100
|
||||
podAffinityTerm:
|
||||
labelSelector:
|
||||
matchExpressions:
|
||||
- key: app
|
||||
operator: In
|
||||
values:
|
||||
- trading-agent-service
|
||||
topologyKey: kubernetes.io/hostname
|
||||
@@ -1,137 +0,0 @@
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
metadata:
|
||||
name: trading-service
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: trading-service
|
||||
component: trading
|
||||
tier: backend
|
||||
version: v1
|
||||
spec:
|
||||
replicas: 2
|
||||
revisionHistoryLimit: 10
|
||||
strategy:
|
||||
type: RollingUpdate
|
||||
rollingUpdate:
|
||||
maxUnavailable: 1
|
||||
maxSurge: 1
|
||||
selector:
|
||||
matchLabels:
|
||||
app: trading-service
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
app: trading-service
|
||||
component: trading
|
||||
tier: backend
|
||||
version: v1
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9092"
|
||||
prometheus.io/path: "/metrics"
|
||||
spec:
|
||||
serviceAccountName: foxhunt-trading-service
|
||||
securityContext:
|
||||
runAsNonRoot: true
|
||||
runAsUser: 1000
|
||||
runAsGroup: 1000
|
||||
fsGroup: 1000
|
||||
containers:
|
||||
- name: trading-service
|
||||
image: foxhunt/trading-service:latest
|
||||
imagePullPolicy: Always
|
||||
ports:
|
||||
- name: grpc
|
||||
containerPort: 50052
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
containerPort: 9092
|
||||
protocol: TCP
|
||||
- name: health
|
||||
containerPort: 8081
|
||||
protocol: TCP
|
||||
env:
|
||||
- name: TRADING_SERVICE_HOST
|
||||
value: "0.0.0.0"
|
||||
- name: TRADING_SERVICE_PORT
|
||||
value: "50052"
|
||||
- name: METRICS_PORT
|
||||
value: "9092"
|
||||
- name: DATABASE_URL
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: database-url
|
||||
- name: REDIS_URL
|
||||
valueFrom:
|
||||
configMapKeyRef:
|
||||
name: foxhunt-config
|
||||
key: redis-url
|
||||
- name: VAULT_ADDR
|
||||
valueFrom:
|
||||
configMapKeyRef:
|
||||
name: foxhunt-config
|
||||
key: vault-addr
|
||||
- name: VAULT_TOKEN
|
||||
valueFrom:
|
||||
secretKeyRef:
|
||||
name: foxhunt-secrets
|
||||
key: vault-token
|
||||
- name: RUST_LOG
|
||||
value: "info,trading_service=debug"
|
||||
- name: RUST_BACKTRACE
|
||||
value: "1"
|
||||
resources:
|
||||
requests:
|
||||
cpu: "2000m"
|
||||
memory: "2Gi"
|
||||
limits:
|
||||
cpu: "4000m"
|
||||
memory: "4Gi"
|
||||
livenessProbe:
|
||||
httpGet:
|
||||
path: /health
|
||||
port: 8081
|
||||
initialDelaySeconds: 30
|
||||
periodSeconds: 10
|
||||
timeoutSeconds: 5
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
readinessProbe:
|
||||
httpGet:
|
||||
path: /ready
|
||||
port: 8081
|
||||
initialDelaySeconds: 10
|
||||
periodSeconds: 5
|
||||
timeoutSeconds: 3
|
||||
successThreshold: 1
|
||||
failureThreshold: 3
|
||||
volumeMounts:
|
||||
- name: config
|
||||
mountPath: /app/config
|
||||
readOnly: true
|
||||
- name: logs
|
||||
mountPath: /app/logs
|
||||
- name: data
|
||||
mountPath: /app/data
|
||||
volumes:
|
||||
- name: config
|
||||
configMap:
|
||||
name: trading-service-config
|
||||
- name: logs
|
||||
emptyDir: {}
|
||||
- name: data
|
||||
emptyDir: {}
|
||||
affinity:
|
||||
podAntiAffinity:
|
||||
preferredDuringSchedulingIgnoredDuringExecution:
|
||||
- weight: 100
|
||||
podAffinityTerm:
|
||||
labelSelector:
|
||||
matchExpressions:
|
||||
- key: app
|
||||
operator: In
|
||||
values:
|
||||
- trading-service
|
||||
topologyKey: kubernetes.io/hostname
|
||||
@@ -1,431 +0,0 @@
|
||||
# Foxhunt HFT Trading System - Complete Kubernetes Deployment
|
||||
# Wave 5 Production Deployment Infrastructure
|
||||
#
|
||||
# This manifest includes:
|
||||
# - Namespace
|
||||
# - ConfigMaps (3)
|
||||
# - Secrets (2)
|
||||
# - PersistentVolumeClaims (5)
|
||||
# - Services (5)
|
||||
# - HorizontalPodAutoscalers (3)
|
||||
#
|
||||
# Usage: kubectl apply -f k8s/foxhunt-complete.yaml
|
||||
|
||||
---
|
||||
# Namespace
|
||||
apiVersion: v1
|
||||
kind: Namespace
|
||||
metadata:
|
||||
name: foxhunt
|
||||
labels:
|
||||
name: foxhunt
|
||||
environment: production
|
||||
|
||||
---
|
||||
# ConfigMap: Main configuration
|
||||
apiVersion: v1
|
||||
kind: ConfigMap
|
||||
metadata:
|
||||
name: foxhunt-config
|
||||
namespace: foxhunt
|
||||
data:
|
||||
redis-url: "redis://redis:6379"
|
||||
vault-addr: "http://vault:8200"
|
||||
postgres-host: "postgres"
|
||||
postgres-port: "5432"
|
||||
postgres-db: "foxhunt"
|
||||
|
||||
---
|
||||
# ConfigMap: Prometheus configuration
|
||||
apiVersion: v1
|
||||
kind: ConfigMap
|
||||
metadata:
|
||||
name: prometheus-config
|
||||
namespace: foxhunt
|
||||
data:
|
||||
prometheus.yml: |
|
||||
global:
|
||||
scrape_interval: 15s
|
||||
evaluation_interval: 15s
|
||||
scrape_configs:
|
||||
- job_name: 'api-gateway'
|
||||
static_configs:
|
||||
- targets: ['api-gateway:9091']
|
||||
- job_name: 'trading-service'
|
||||
static_configs:
|
||||
- targets: ['trading-service:9092']
|
||||
- job_name: 'backtesting-service'
|
||||
static_configs:
|
||||
- targets: ['backtesting-service:9093']
|
||||
- job_name: 'ml-training-service'
|
||||
static_configs:
|
||||
- targets: ['ml-training-service:9094']
|
||||
- job_name: 'trading-agent-service'
|
||||
static_configs:
|
||||
- targets: ['trading-agent-service:9095']
|
||||
|
||||
---
|
||||
# ConfigMap: Redis configuration
|
||||
apiVersion: v1
|
||||
kind: ConfigMap
|
||||
metadata:
|
||||
name: redis-config
|
||||
namespace: foxhunt
|
||||
data:
|
||||
redis.conf: |
|
||||
maxmemory 256mb
|
||||
maxmemory-policy allkeys-lru
|
||||
appendonly yes
|
||||
save 900 1
|
||||
save 300 10
|
||||
save 60 10000
|
||||
|
||||
---
|
||||
# Secret: Database credentials (base64 encoded)
|
||||
# IMPORTANT: Replace with actual values from Vault in production
|
||||
apiVersion: v1
|
||||
kind: Secret
|
||||
metadata:
|
||||
name: foxhunt-secrets
|
||||
namespace: foxhunt
|
||||
type: Opaque
|
||||
stringData:
|
||||
jwt-secret: "REPLACE_WITH_VAULT_VALUE"
|
||||
database-url: "postgresql://foxhunt:REPLACE_PASSWORD@postgres:5432/foxhunt"
|
||||
vault-token: "REPLACE_WITH_VAULT_TOKEN"
|
||||
aws-access-key-id: "REPLACE_WITH_AWS_KEY"
|
||||
aws-secret-access-key: "REPLACE_WITH_AWS_SECRET"
|
||||
|
||||
---
|
||||
# Secret: Vault credentials
|
||||
apiVersion: v1
|
||||
kind: Secret
|
||||
metadata:
|
||||
name: vault-secrets
|
||||
namespace: foxhunt
|
||||
type: Opaque
|
||||
stringData:
|
||||
vault-root-token: "REPLACE_WITH_VAULT_ROOT_TOKEN"
|
||||
vault-unseal-key: "REPLACE_WITH_VAULT_UNSEAL_KEY"
|
||||
|
||||
---
|
||||
# PVC: PostgreSQL data
|
||||
apiVersion: v1
|
||||
kind: PersistentVolumeClaim
|
||||
metadata:
|
||||
name: postgres-pvc
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
accessModes:
|
||||
- ReadWriteOnce
|
||||
resources:
|
||||
requests:
|
||||
storage: 100Gi
|
||||
storageClassName: fast-ssd
|
||||
|
||||
---
|
||||
# PVC: Redis data
|
||||
apiVersion: v1
|
||||
kind: PersistentVolumeClaim
|
||||
metadata:
|
||||
name: redis-pvc
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
accessModes:
|
||||
- ReadWriteOnce
|
||||
resources:
|
||||
requests:
|
||||
storage: 10Gi
|
||||
storageClassName: fast-ssd
|
||||
|
||||
---
|
||||
# PVC: Prometheus data
|
||||
apiVersion: v1
|
||||
kind: PersistentVolumeClaim
|
||||
metadata:
|
||||
name: prometheus-pvc
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
accessModes:
|
||||
- ReadWriteOnce
|
||||
resources:
|
||||
requests:
|
||||
storage: 50Gi
|
||||
storageClassName: standard
|
||||
|
||||
---
|
||||
# PVC: ML Models
|
||||
apiVersion: v1
|
||||
kind: PersistentVolumeClaim
|
||||
metadata:
|
||||
name: ml-models-pvc
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
accessModes:
|
||||
- ReadWriteOnce
|
||||
resources:
|
||||
requests:
|
||||
storage: 50Gi
|
||||
storageClassName: fast-ssd
|
||||
|
||||
---
|
||||
# PVC: ML Checkpoints
|
||||
apiVersion: v1
|
||||
kind: PersistentVolumeClaim
|
||||
metadata:
|
||||
name: ml-checkpoints-pvc
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
accessModes:
|
||||
- ReadWriteOnce
|
||||
resources:
|
||||
requests:
|
||||
storage: 100Gi
|
||||
storageClassName: fast-ssd
|
||||
|
||||
---
|
||||
# PVC: Backtesting Data
|
||||
apiVersion: v1
|
||||
kind: PersistentVolumeClaim
|
||||
metadata:
|
||||
name: backtesting-data-pvc
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
accessModes:
|
||||
- ReadWriteOnce
|
||||
resources:
|
||||
requests:
|
||||
storage: 200Gi
|
||||
storageClassName: standard
|
||||
|
||||
---
|
||||
# PVC: Backtesting Results
|
||||
apiVersion: v1
|
||||
kind: PersistentVolumeClaim
|
||||
metadata:
|
||||
name: backtesting-results-pvc
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
accessModes:
|
||||
- ReadWriteOnce
|
||||
resources:
|
||||
requests:
|
||||
storage: 50Gi
|
||||
storageClassName: standard
|
||||
|
||||
---
|
||||
# Service: Backtesting Service
|
||||
apiVersion: v1
|
||||
kind: Service
|
||||
metadata:
|
||||
name: backtesting-service
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: backtesting-service
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9093"
|
||||
spec:
|
||||
type: ClusterIP
|
||||
selector:
|
||||
app: backtesting-service
|
||||
ports:
|
||||
- name: grpc
|
||||
port: 50053
|
||||
targetPort: 50053
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
port: 9093
|
||||
targetPort: 9093
|
||||
protocol: TCP
|
||||
- name: health
|
||||
port: 8082
|
||||
targetPort: 8082
|
||||
protocol: TCP
|
||||
|
||||
---
|
||||
# Service: ML Training Service
|
||||
apiVersion: v1
|
||||
kind: Service
|
||||
metadata:
|
||||
name: ml-training-service
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: ml-training-service
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9094"
|
||||
spec:
|
||||
type: ClusterIP
|
||||
selector:
|
||||
app: ml-training-service
|
||||
ports:
|
||||
- name: grpc
|
||||
port: 50054
|
||||
targetPort: 50054
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
port: 9094
|
||||
targetPort: 9094
|
||||
protocol: TCP
|
||||
- name: health
|
||||
port: 8095
|
||||
targetPort: 8095
|
||||
protocol: TCP
|
||||
|
||||
---
|
||||
# Service: Trading Agent Service
|
||||
apiVersion: v1
|
||||
kind: Service
|
||||
metadata:
|
||||
name: trading-agent-service
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: trading-agent-service
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9095"
|
||||
spec:
|
||||
type: ClusterIP
|
||||
selector:
|
||||
app: trading-agent-service
|
||||
ports:
|
||||
- name: grpc
|
||||
port: 50055
|
||||
targetPort: 50055
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
port: 9095
|
||||
targetPort: 9095
|
||||
protocol: TCP
|
||||
- name: health
|
||||
port: 8083
|
||||
targetPort: 8083
|
||||
protocol: TCP
|
||||
|
||||
---
|
||||
# HPA: API Gateway
|
||||
apiVersion: autoscaling/v2
|
||||
kind: HorizontalPodAutoscaler
|
||||
metadata:
|
||||
name: api-gateway-hpa
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
scaleTargetRef:
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
name: api-gateway
|
||||
minReplicas: 3
|
||||
maxReplicas: 10
|
||||
metrics:
|
||||
- type: Resource
|
||||
resource:
|
||||
name: cpu
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 70
|
||||
- type: Resource
|
||||
resource:
|
||||
name: memory
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 80
|
||||
behavior:
|
||||
scaleDown:
|
||||
stabilizationWindowSeconds: 300
|
||||
policies:
|
||||
- type: Percent
|
||||
value: 50
|
||||
periodSeconds: 60
|
||||
scaleUp:
|
||||
stabilizationWindowSeconds: 0
|
||||
policies:
|
||||
- type: Percent
|
||||
value: 100
|
||||
periodSeconds: 30
|
||||
- type: Pods
|
||||
value: 2
|
||||
periodSeconds: 30
|
||||
selectPolicy: Max
|
||||
|
||||
---
|
||||
# HPA: Trading Service
|
||||
apiVersion: autoscaling/v2
|
||||
kind: HorizontalPodAutoscaler
|
||||
metadata:
|
||||
name: trading-service-hpa
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
scaleTargetRef:
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
name: trading-service
|
||||
minReplicas: 2
|
||||
maxReplicas: 5
|
||||
metrics:
|
||||
- type: Resource
|
||||
resource:
|
||||
name: cpu
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 80
|
||||
- type: Resource
|
||||
resource:
|
||||
name: memory
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 85
|
||||
behavior:
|
||||
scaleDown:
|
||||
stabilizationWindowSeconds: 300
|
||||
policies:
|
||||
- type: Pods
|
||||
value: 1
|
||||
periodSeconds: 120
|
||||
scaleUp:
|
||||
stabilizationWindowSeconds: 0
|
||||
policies:
|
||||
- type: Pods
|
||||
value: 1
|
||||
periodSeconds: 60
|
||||
|
||||
---
|
||||
# HPA: ML Training Service
|
||||
apiVersion: autoscaling/v2
|
||||
kind: HorizontalPodAutoscaler
|
||||
metadata:
|
||||
name: ml-training-service-hpa
|
||||
namespace: foxhunt
|
||||
spec:
|
||||
scaleTargetRef:
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
name: ml-training-service
|
||||
minReplicas: 1
|
||||
maxReplicas: 3
|
||||
metrics:
|
||||
- type: Resource
|
||||
resource:
|
||||
name: nvidia.com/gpu
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 90
|
||||
- type: Resource
|
||||
resource:
|
||||
name: memory
|
||||
target:
|
||||
type: Utilization
|
||||
averageUtilization: 85
|
||||
behavior:
|
||||
scaleDown:
|
||||
stabilizationWindowSeconds: 600
|
||||
policies:
|
||||
- type: Pods
|
||||
value: 1
|
||||
periodSeconds: 300
|
||||
scaleUp:
|
||||
stabilizationWindowSeconds: 60
|
||||
policies:
|
||||
- type: Pods
|
||||
value: 1
|
||||
periodSeconds: 120
|
||||
@@ -1,33 +0,0 @@
|
||||
apiVersion: v1
|
||||
kind: Service
|
||||
metadata:
|
||||
name: api-gateway
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: api-gateway
|
||||
component: gateway
|
||||
tier: frontend
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9091"
|
||||
spec:
|
||||
type: LoadBalancer
|
||||
sessionAffinity: ClientIP
|
||||
sessionAffinityConfig:
|
||||
clientIP:
|
||||
timeoutSeconds: 10800
|
||||
selector:
|
||||
app: api-gateway
|
||||
ports:
|
||||
- name: grpc
|
||||
port: 50051
|
||||
targetPort: 50051
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
port: 9091
|
||||
targetPort: 9091
|
||||
protocol: TCP
|
||||
- name: health
|
||||
port: 8080
|
||||
targetPort: 8080
|
||||
protocol: TCP
|
||||
@@ -1,29 +0,0 @@
|
||||
apiVersion: v1
|
||||
kind: Service
|
||||
metadata:
|
||||
name: trading-service
|
||||
namespace: foxhunt
|
||||
labels:
|
||||
app: trading-service
|
||||
component: trading
|
||||
tier: backend
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9092"
|
||||
spec:
|
||||
type: ClusterIP
|
||||
selector:
|
||||
app: trading-service
|
||||
ports:
|
||||
- name: grpc
|
||||
port: 50052
|
||||
targetPort: 50052
|
||||
protocol: TCP
|
||||
- name: metrics
|
||||
port: 9092
|
||||
targetPort: 9092
|
||||
protocol: TCP
|
||||
- name: health
|
||||
port: 8081
|
||||
targetPort: 8081
|
||||
protocol: TCP
|
||||
@@ -1,233 +0,0 @@
|
||||
# =============================================================================
|
||||
# FOXHUNT TRADING SERVICE - ULTRA-PERFORMANCE KUBERNETES DEPLOYMENT
|
||||
# =============================================================================
|
||||
# This manifest deploys the Trading Service with specialized performance
|
||||
# optimizations for 14ns latency requirements
|
||||
|
||||
apiVersion: v1
|
||||
kind: Namespace
|
||||
metadata:
|
||||
name: foxhunt-trading
|
||||
labels:
|
||||
app.kubernetes.io/name: foxhunt
|
||||
app.kubernetes.io/component: trading
|
||||
performance-tier: ultra-high
|
||||
|
||||
---
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
metadata:
|
||||
name: trading-service
|
||||
namespace: foxhunt-trading
|
||||
labels:
|
||||
app: trading-service
|
||||
version: v1
|
||||
performance-tier: ultra-high
|
||||
spec:
|
||||
replicas: 2 # Active-standby for HA
|
||||
strategy:
|
||||
type: RollingUpdate
|
||||
rollingUpdate:
|
||||
maxSurge: 1
|
||||
maxUnavailable: 0 # Zero-downtime deployment
|
||||
selector:
|
||||
matchLabels:
|
||||
app: trading-service
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
app: trading-service
|
||||
version: v1
|
||||
performance-tier: ultra-high
|
||||
annotations:
|
||||
# Performance annotations
|
||||
scheduler.alpha.kubernetes.io/critical-pod: "true"
|
||||
cluster-autoscaler.kubernetes.io/safe-to-evict: "false"
|
||||
spec:
|
||||
# ULTRA-PERFORMANCE NODE AFFINITY
|
||||
nodeSelector:
|
||||
performance-tier: ultra-high
|
||||
workload-type: trading
|
||||
|
||||
affinity:
|
||||
nodeAffinity:
|
||||
requiredDuringSchedulingIgnoredDuringExecution:
|
||||
nodeSelectorTerms:
|
||||
- matchExpressions:
|
||||
- key: performance-tier
|
||||
operator: In
|
||||
values: ["ultra-high"]
|
||||
- key: kubernetes.io/arch
|
||||
operator: In
|
||||
values: ["amd64"]
|
||||
podAntiAffinity:
|
||||
preferredDuringSchedulingIgnoredDuringExecution:
|
||||
- weight: 100
|
||||
podAffinityTerm:
|
||||
labelSelector:
|
||||
matchExpressions:
|
||||
- key: app
|
||||
operator: In
|
||||
values: ["trading-service"]
|
||||
topologyKey: kubernetes.io/hostname
|
||||
|
||||
# PERFORMANCE-CRITICAL CONFIGURATION
|
||||
hostNetwork: true # Direct host networking for minimal latency
|
||||
dnsPolicy: ClusterFirstWithHostNet
|
||||
|
||||
# PRIVILEGED ACCESS FOR HARDWARE OPTIMIZATION
|
||||
securityContext:
|
||||
runAsNonRoot: false
|
||||
runAsUser: 0
|
||||
fsGroup: 0
|
||||
|
||||
# PRIORITY AND PREEMPTION
|
||||
priorityClassName: system-critical
|
||||
|
||||
# RESOURCE TOPOLOGY
|
||||
topologySpreadConstraints:
|
||||
- maxSkew: 1
|
||||
topologyKey: kubernetes.io/hostname
|
||||
whenUnsatisfiable: DoNotSchedule
|
||||
labelSelector:
|
||||
matchLabels:
|
||||
app: trading-service
|
||||
|
||||
containers:
|
||||
- name: trading-service
|
||||
image: foxhunt/trading-service:latest
|
||||
imagePullPolicy: Always
|
||||
|
||||
# ULTRA-PERFORMANCE SECURITY CONTEXT
|
||||
securityContext:
|
||||
privileged: true # Required for RDTSC hardware access
|
||||
allowPrivilegeEscalation: true
|
||||
readOnlyRootFilesystem: false
|
||||
capabilities:
|
||||
add:
|
||||
- SYS_NICE # CPU scheduling priority
|
||||
- SYS_TIME # High-resolution timers
|
||||
- SYS_RAWIO # Raw I/O operations
|
||||
- NET_RAW # Raw network access
|
||||
- IPC_LOCK # Memory locking
|
||||
- SYS_RESOURCE # Resource limits
|
||||
|
||||
# CPU AFFINITY AND RESOURCE LIMITS
|
||||
resources:
|
||||
requests:
|
||||
memory: "8Gi"
|
||||
cpu: "4"
|
||||
hugepages-2Mi: "2Gi"
|
||||
limits:
|
||||
memory: "8Gi" # Hard limit to prevent swap
|
||||
cpu: "4" # Dedicated CPU cores
|
||||
hugepages-2Mi: "2Gi"
|
||||
|
||||
# PERFORMANCE-CRITICAL ENVIRONMENT
|
||||
env:
|
||||
- name: RUST_LOG
|
||||
value: "error" # Minimal logging for performance
|
||||
- name: MALLOC_CONF
|
||||
value: "background_thread:false,dirty_decay_ms:0,muzzy_decay_ms:0"
|
||||
- name: CPU_AFFINITY_CORES
|
||||
value: "0,1,2,3" # Pinned to specific cores
|
||||
- name: GOMAXPROCS
|
||||
value: "4"
|
||||
- name: OMP_NUM_THREADS
|
||||
value: "4"
|
||||
|
||||
# VOLUME MOUNTS FOR PERFORMANCE
|
||||
volumeMounts:
|
||||
- name: hugepages
|
||||
mountPath: /dev/hugepages
|
||||
- name: proc
|
||||
mountPath: /host/proc
|
||||
readOnly: true
|
||||
|
||||
# gRPC PORT
|
||||
ports:
|
||||
- containerPort: 50051
|
||||
name: grpc
|
||||
protocol: TCP
|
||||
|
||||
# KUBERNETES PROBES (lightweight for performance)
|
||||
livenessProbe:
|
||||
tcpSocket:
|
||||
port: 50051
|
||||
initialDelaySeconds: 10
|
||||
periodSeconds: 30
|
||||
timeoutSeconds: 5
|
||||
failureThreshold: 3
|
||||
|
||||
readinessProbe:
|
||||
tcpSocket:
|
||||
port: 50051
|
||||
initialDelaySeconds: 5
|
||||
periodSeconds: 10
|
||||
timeoutSeconds: 3
|
||||
failureThreshold: 1
|
||||
|
||||
# PERFORMANCE-OPTIMIZED VOLUMES
|
||||
volumes:
|
||||
- name: hugepages
|
||||
emptyDir:
|
||||
medium: HugePages-2Mi
|
||||
- name: proc
|
||||
hostPath:
|
||||
path: /proc
|
||||
type: Directory
|
||||
|
||||
# SCHEDULING CONSTRAINTS
|
||||
tolerations:
|
||||
- key: "performance-tier"
|
||||
operator: "Equal"
|
||||
value: "ultra-high"
|
||||
effect: "NoSchedule"
|
||||
- key: "node.kubernetes.io/not-ready"
|
||||
operator: "Exists"
|
||||
effect: "NoExecute"
|
||||
tolerationSeconds: 30
|
||||
|
||||
---
|
||||
apiVersion: v1
|
||||
kind: Service
|
||||
metadata:
|
||||
name: trading-service
|
||||
namespace: foxhunt-trading
|
||||
labels:
|
||||
app: trading-service
|
||||
annotations:
|
||||
# Performance service annotations
|
||||
service.beta.kubernetes.io/aws-load-balancer-type: "nlb"
|
||||
service.beta.kubernetes.io/aws-load-balancer-backend-protocol: "tcp"
|
||||
spec:
|
||||
type: ClusterIP # Internal service only
|
||||
sessionAffinity: ClientIP # Session stickiness for performance
|
||||
selector:
|
||||
app: trading-service
|
||||
ports:
|
||||
- name: grpc
|
||||
port: 50051
|
||||
targetPort: 50051
|
||||
protocol: TCP
|
||||
|
||||
---
|
||||
apiVersion: policy/v1
|
||||
kind: PodDisruptionBudget
|
||||
metadata:
|
||||
name: trading-service-pdb
|
||||
namespace: foxhunt-trading
|
||||
spec:
|
||||
minAvailable: 1 # Always keep at least one instance running
|
||||
selector:
|
||||
matchLabels:
|
||||
app: trading-service
|
||||
|
||||
---
|
||||
apiVersion: scheduling.k8s.io/v1
|
||||
kind: PriorityClass
|
||||
metadata:
|
||||
name: system-critical
|
||||
value: 1000000
|
||||
globalDefault: false
|
||||
description: "Ultra-high priority for critical trading services"
|
||||
Reference in New Issue
Block a user