kailash 0.8.4__py3-none-any.whl → 0.8.5__py3-none-any.whl

kailash/edge/resource/resource_analyzer.py

@@ -0,0 +1,824 @@
+"""Resource analyzer for intelligent edge resource management.
+
+This module provides real-time resource analysis, pattern identification,
+and bottleneck detection for edge computing infrastructure.
+"""
+
+import asyncio
+import logging
+from collections import defaultdict, deque
+from dataclasses import dataclass, field
+from datetime import datetime, timedelta
+from enum import Enum
+from typing import Any, Dict, List, Optional, Tuple
+
+import numpy as np
+from scipy import stats
+
+
+class ResourceType(Enum):
+    """Types of resources to analyze."""
+
+    CPU = "cpu"
+    MEMORY = "memory"
+    GPU = "gpu"
+    STORAGE = "storage"
+    NETWORK = "network"
+    CUSTOM = "custom"
+
+
+class BottleneckType(Enum):
+    """Types of resource bottlenecks."""
+
+    CAPACITY = "capacity"  # Not enough resources
+    ALLOCATION = "allocation"  # Poor distribution
+    CONTENTION = "contention"  # Resource conflicts
+    FRAGMENTATION = "fragmentation"  # Wasted space
+    THROTTLING = "throttling"  # Rate limiting
+
+
+@dataclass
+class ResourceMetric:
+    """Single resource metric measurement."""
+
+    timestamp: datetime
+    edge_node: str
+    resource_type: ResourceType
+    used: float
+    available: float
+    total: float
+    metadata: Dict[str, Any] = field(default_factory=dict)
+
+    @property
+    def utilization(self) -> float:
+        """Calculate utilization percentage."""
+        if self.total == 0:
+            return 0.0
+        return (self.used / self.total) * 100
+
+    @property
+    def free(self) -> float:
+        """Calculate free resources."""
+        return self.total - self.used
+
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary."""
+        return {
+            "timestamp": self.timestamp.isoformat(),
+            "edge_node": self.edge_node,
+            "resource_type": self.resource_type.value,
+            "used": self.used,
+            "available": self.available,
+            "total": self.total,
+            "utilization": self.utilization,
+            "metadata": self.metadata,
+        }
+
+
+@dataclass
+class ResourcePattern:
+    """Identified resource usage pattern."""
+
+    pattern_type: str
+    confidence: float
+    edge_nodes: List[str]
+    resource_types: List[ResourceType]
+    characteristics: Dict[str, Any]
+    recommendations: List[str]
+
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary."""
+        return {
+            "pattern_type": self.pattern_type,
+            "confidence": self.confidence,
+            "edge_nodes": self.edge_nodes,
+            "resource_types": [rt.value for rt in self.resource_types],
+            "characteristics": self.characteristics,
+            "recommendations": self.recommendations,
+        }
+
+
+@dataclass
+class Bottleneck:
+    """Identified resource bottleneck."""
+
+    bottleneck_type: BottleneckType
+    severity: float  # 0-1 scale
+    edge_node: str
+    resource_type: ResourceType
+    description: str
+    impact: Dict[str, Any]
+    resolution: List[str]
+
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary."""
+        return {
+            "bottleneck_type": self.bottleneck_type.value,
+            "severity": self.severity,
+            "edge_node": self.edge_node,
+            "resource_type": self.resource_type.value,
+            "description": self.description,
+            "impact": self.impact,
+            "resolution": self.resolution,
+        }
+
+
+class ResourceAnalyzer:
+    """Analyzes resource usage patterns and identifies bottlenecks."""
+
+    def __init__(
+        self,
+        history_window: int = 3600,  # 1 hour
+        analysis_interval: int = 60,  # 1 minute
+        anomaly_threshold: float = 2.5,  # Standard deviations
+        pattern_confidence_threshold: float = 0.7,
+    ):
+        """Initialize resource analyzer.
+
+        Args:
+            history_window: Time window for analysis (seconds)
+            analysis_interval: Interval between analyses (seconds)
+            anomaly_threshold: Threshold for anomaly detection
+            pattern_confidence_threshold: Minimum confidence for patterns
+        """
+        self.history_window = history_window
+        self.analysis_interval = analysis_interval
+        self.anomaly_threshold = anomaly_threshold
+        self.pattern_confidence_threshold = pattern_confidence_threshold
+
+        # Resource metrics storage
+        self.metrics: Dict[str, deque] = defaultdict(lambda: deque(maxlen=1000))
+
+        # Analysis results
+        self.patterns: List[ResourcePattern] = []
+        self.bottlenecks: List[Bottleneck] = []
+        self.anomalies: List[Dict[str, Any]] = []
+
+        # Background task
+        self._analysis_task: Optional[asyncio.Task] = None
+
+        self.logger = logging.getLogger(__name__)
+
+    async def start(self):
+        """Start background analysis."""
+        if not self._analysis_task:
+            self._analysis_task = asyncio.create_task(self._analysis_loop())
+            self.logger.info("Resource analyzer started")
+
+    async def stop(self):
+        """Stop background analysis."""
+        if self._analysis_task:
+            self._analysis_task.cancel()
+            try:
+                await self._analysis_task
+            except asyncio.CancelledError:
+                pass
+            self._analysis_task = None
+            self.logger.info("Resource analyzer stopped")
+
+    async def record_metric(self, metric: ResourceMetric):
+        """Record a resource metric.
+
+        Args:
+            metric: Resource metric to record
+        """
+        key = f"{metric.edge_node}:{metric.resource_type.value}"
+        self.metrics[key].append(metric)
+
+        # Check for immediate issues
+        await self._check_immediate_issues(metric)
+
+    async def analyze_resources(self) -> Dict[str, Any]:
+        """Perform comprehensive resource analysis.
+
+        Returns:
+            Analysis results
+        """
+        # Clear previous results
+        self.patterns.clear()
+        self.bottlenecks.clear()
+        self.anomalies.clear()
+
+        # Run all analyses
+        await self._identify_patterns()
+        await self._detect_bottlenecks()
+        await self._detect_anomalies()
+
+        return {
+            "patterns": [p.to_dict() for p in self.patterns],
+            "bottlenecks": [b.to_dict() for b in self.bottlenecks],
+            "anomalies": self.anomalies,
+            "summary": self._generate_summary(),
+        }
+
+    async def get_resource_trends(
+        self,
+        edge_node: Optional[str] = None,
+        resource_type: Optional[ResourceType] = None,
+        duration_minutes: int = 60,
+    ) -> Dict[str, Any]:
+        """Get resource usage trends.
+
+        Args:
+            edge_node: Filter by edge node
+            resource_type: Filter by resource type
+            duration_minutes: Duration to analyze
+
+        Returns:
+            Trend analysis
+        """
+        trends = {}
+        cutoff = datetime.now() - timedelta(minutes=duration_minutes)
+
+        for key, metrics in self.metrics.items():
+            node, rtype = key.split(":")
+
+            # Apply filters
+            if edge_node and node != edge_node:
+                continue
+            if resource_type and rtype != resource_type.value:
+                continue
+
+            # Get recent metrics
+            recent = [m for m in metrics if m.timestamp > cutoff]
+            if not recent:
+                continue
+
+            # Calculate trends
+            utilizations = [m.utilization for m in recent]
+            timestamps = [(m.timestamp - cutoff).total_seconds() for m in recent]
+
+            if len(utilizations) > 1:
+                # Linear regression for trend
+                slope, intercept, r_value, _, _ = stats.linregress(
+                    timestamps, utilizations
+                )
+
+                trends[key] = {
+                    "current": utilizations[-1],
+                    "average": np.mean(utilizations),
+                    "min": np.min(utilizations),
+                    "max": np.max(utilizations),
+                    "std_dev": np.std(utilizations),
+                    "trend_slope": slope,
+                    "trend_direction": (
+                        "increasing"
+                        if slope > 0.1
+                        else "decreasing" if slope < -0.1 else "stable"
+                    ),
+                    "prediction_1h": (
+                        intercept + slope * 3600 if abs(r_value) > 0.5 else None
+                    ),
+                }
+
+        return trends
+
+    async def get_optimization_recommendations(self) -> List[Dict[str, Any]]:
+        """Get resource optimization recommendations.
+
+        Returns:
+            List of recommendations
+        """
+        recommendations = []
+
+        # Analyze current state
+        await self.analyze_resources()
+
+        # Pattern-based recommendations
+        for pattern in self.patterns:
+            if pattern.confidence >= self.pattern_confidence_threshold:
+                recommendations.append(
+                    {
+                        "type": "pattern",
+                        "priority": self._calculate_priority(pattern.confidence),
+                        "pattern": pattern.pattern_type,
+                        "affected_nodes": pattern.edge_nodes,
+                        "recommendations": pattern.recommendations,
+                        "expected_improvement": pattern.characteristics.get(
+                            "improvement", "10-20%"
+                        ),
+                    }
+                )
+
+        # Bottleneck-based recommendations
+        for bottleneck in self.bottlenecks:
+            if bottleneck.severity > 0.5:
+                recommendations.append(
+                    {
+                        "type": "bottleneck",
+                        "priority": self._calculate_priority(bottleneck.severity),
+                        "issue": bottleneck.description,
+                        "node": bottleneck.edge_node,
+                        "resource": bottleneck.resource_type.value,
+                        "resolutions": bottleneck.resolution,
+                        "impact": bottleneck.impact,
+                    }
+                )
+
+        # Sort by priority
+        recommendations.sort(key=lambda x: x["priority"], reverse=True)
+
+        return recommendations
+
+    async def _analysis_loop(self):
+        """Background analysis loop."""
+        while True:
+            try:
+                await asyncio.sleep(self.analysis_interval)
+                await self.analyze_resources()
+            except asyncio.CancelledError:
+                break
+            except Exception as e:
+                self.logger.error(f"Analysis error: {e}")
+
+    async def _check_immediate_issues(self, metric: ResourceMetric):
+        """Check for immediate issues requiring attention.
+
+        Args:
+            metric: Resource metric to check
+        """
+        # Critical utilization check
+        if metric.utilization > 95:
+            self.logger.warning(
+                f"Critical {metric.resource_type.value} utilization "
+                f"on {metric.edge_node}: {metric.utilization:.1f}%"
+            )
+
+        # No available resources
+        if metric.available == 0 and metric.resource_type in [
+            ResourceType.CPU,
+            ResourceType.MEMORY,
+        ]:
+            self.logger.error(
+                f"No {metric.resource_type.value} available " f"on {metric.edge_node}"
+            )
+
+    async def _identify_patterns(self):
+        """Identify resource usage patterns."""
+        # Periodic pattern detection
+        periodic_pattern = await self._detect_periodic_pattern()
+        if periodic_pattern:
+            self.patterns.append(periodic_pattern)
+
+        # Spike pattern detection
+        spike_pattern = await self._detect_spike_pattern()
+        if spike_pattern:
+            self.patterns.append(spike_pattern)
+
+        # Gradual increase pattern
+        growth_pattern = await self._detect_growth_pattern()
+        if growth_pattern:
+            self.patterns.append(growth_pattern)
+
+        # Imbalance pattern
+        imbalance_pattern = await self._detect_imbalance_pattern()
+        if imbalance_pattern:
+            self.patterns.append(imbalance_pattern)
+
+    async def _detect_periodic_pattern(self) -> Optional[ResourcePattern]:
+        """Detect periodic usage patterns."""
+        for key, metrics in self.metrics.items():
+            if len(metrics) < 100:
+                continue
+
+            # Extract utilization time series
+            utilizations = [m.utilization for m in metrics]
+
+            # Simple FFT-based periodicity detection
+            fft = np.fft.fft(utilizations)
+            frequencies = np.fft.fftfreq(len(utilizations))
+
+            # Find dominant frequency
+            dominant_idx = np.argmax(np.abs(fft[1 : len(fft) // 2])) + 1
+            if np.abs(fft[dominant_idx]) > len(utilizations) * 0.1:
+                period = (
+                    1 / frequencies[dominant_idx]
+                    if frequencies[dominant_idx] != 0
+                    else 0
+                )
+
+                if period > 0:
+                    node, rtype = key.split(":")
+                    return ResourcePattern(
+                        pattern_type="periodic",
+                        confidence=min(
+                            np.abs(fft[dominant_idx]) / len(utilizations), 1.0
+                        ),
+                        edge_nodes=[node],
+                        resource_types=[ResourceType(rtype)],
+                        characteristics={
+                            "period_seconds": abs(period * self.analysis_interval),
+                            "amplitude": np.std(utilizations),
+                            "improvement": "15-25%",
+                        },
+                        recommendations=[
+                            f"Implement predictive scaling with {abs(period * self.analysis_interval):.0f}s period",
+                            "Use time-based resource allocation",
+                            "Consider workload scheduling optimization",
+                        ],
+                    )
+
+        return None
+
+    async def _detect_spike_pattern(self) -> Optional[ResourcePattern]:
+        """Detect resource usage spikes."""
+        spike_nodes = []
+        spike_resources = set()
+
+        for key, metrics in self.metrics.items():
+            if len(metrics) < 10:
+                continue
+
+            utilizations = [m.utilization for m in metrics]
+            mean = np.mean(utilizations)
+            std = np.std(utilizations)
+
+            # Count spikes
+            spikes = sum(1 for u in utilizations if u > mean + 2 * std)
+
+            if spikes > len(utilizations) * 0.1:  # More than 10% are spikes
+                node, rtype = key.split(":")
+                spike_nodes.append(node)
+                spike_resources.add(ResourceType(rtype))
+
+        if spike_nodes:
+            return ResourcePattern(
+                pattern_type="spike",
+                confidence=0.8,
+                edge_nodes=list(set(spike_nodes)),
+                resource_types=list(spike_resources),
+                characteristics={
+                    "frequency": "frequent",
+                    "impact": "high",
+                    "improvement": "20-30%",
+                },
+                recommendations=[
+                    "Implement burst capacity allocation",
+                    "Use resource pooling for spike handling",
+                    "Consider request rate limiting",
+                    "Enable auto-scaling with aggressive policies",
+                ],
+            )
+
+        return None
+
+    async def _detect_growth_pattern(self) -> Optional[ResourcePattern]:
+        """Detect gradual resource growth patterns."""
+        growth_nodes = []
+        growth_resources = set()
+
+        for key, metrics in self.metrics.items():
+            if len(metrics) < 50:
+                continue
+
+            # Get recent metrics
+            recent = list(metrics)[-50:]
+            utilizations = [m.utilization for m in recent]
+            timestamps = list(range(len(utilizations)))
+
+            # Linear regression
+            slope, _, r_value, _, _ = stats.linregress(timestamps, utilizations)
+
+            # Significant positive trend
+            if slope > 0.1 and abs(r_value) > 0.7:
+                node, rtype = key.split(":")
+                growth_nodes.append(node)
+                growth_resources.add(ResourceType(rtype))
+
+        if growth_nodes:
+            return ResourcePattern(
+                pattern_type="gradual_growth",
+                confidence=0.75,
+                edge_nodes=list(set(growth_nodes)),
+                resource_types=list(growth_resources),
+                characteristics={
+                    "growth_rate": "steady",
+                    "risk": "capacity_exhaustion",
+                    "improvement": "25-35%",
+                },
+                recommendations=[
+                    "Plan capacity expansion",
+                    "Implement predictive scaling",
+                    "Review resource cleanup policies",
+                    "Consider workload migration strategies",
+                ],
+            )
+
+        return None
+
+    async def _detect_imbalance_pattern(self) -> Optional[ResourcePattern]:
+        """Detect resource imbalance across nodes."""
+        # Group by resource type
+        by_type: Dict[ResourceType, List[float]] = defaultdict(list)
+        node_utils: Dict[str, float] = {}
+
+        for key, metrics in self.metrics.items():
+            if not metrics:
+                continue
+
+            node, rtype = key.split(":")
+            recent_util = np.mean([m.utilization for m in list(metrics)[-10:]])
+
+            by_type[ResourceType(rtype)].append(recent_util)
+            node_utils[node] = recent_util
+
+        # Check for imbalance
+        imbalanced_resources = []
+        for rtype, utils in by_type.items():
+            if len(utils) > 1:
+                cv = np.std(utils) / np.mean(utils) if np.mean(utils) > 0 else 0
+                if cv > 0.5:  # Coefficient of variation > 0.5
+                    imbalanced_resources.append(rtype)
+
+        if imbalanced_resources:
+            # Find over and under utilized nodes
+            avg_util = np.mean(list(node_utils.values()))
+            over_utilized = [n for n, u in node_utils.items() if u > avg_util + 20]
+            under_utilized = [n for n, u in node_utils.items() if u < avg_util - 20]
+
+            return ResourcePattern(
+                pattern_type="imbalance",
+                confidence=0.85,
+                edge_nodes=over_utilized + under_utilized,
+                resource_types=imbalanced_resources,
+                characteristics={
+                    "over_utilized": over_utilized,
+                    "under_utilized": under_utilized,
+                    "imbalance_severity": "high",
+                    "improvement": "30-40%",
+                },
+                recommendations=[
+                    "Implement load balancing strategies",
+                    "Use affinity rules for better distribution",
+                    "Consider workload migration from hot nodes",
+                    "Enable cross-node resource sharing",
+                ],
+            )
+
+        return None
+
+    async def _detect_bottlenecks(self):
+        """Detect resource bottlenecks."""
+        # Capacity bottlenecks
+        await self._detect_capacity_bottlenecks()
+
+        # Allocation bottlenecks
+        await self._detect_allocation_bottlenecks()
+
+        # Contention bottlenecks
+        await self._detect_contention_bottlenecks()
+
+        # Fragmentation bottlenecks
+        await self._detect_fragmentation_bottlenecks()
+
+    async def _detect_capacity_bottlenecks(self):
+        """Detect capacity bottlenecks."""
+        for key, metrics in self.metrics.items():
+            if not metrics:
+                continue
+
+            node, rtype = key.split(":")
+            recent = list(metrics)[-10:]
+
+            # Check sustained high utilization
+            high_util_count = sum(1 for m in recent if m.utilization > 85)
+
+            if high_util_count > len(recent) * 0.8:
+                avg_util = np.mean([m.utilization for m in recent])
+
+                self.bottlenecks.append(
+                    Bottleneck(
+                        bottleneck_type=BottleneckType.CAPACITY,
+                        severity=min((avg_util - 85) / 15, 1.0),
+                        edge_node=node,
+                        resource_type=ResourceType(rtype),
+                        description=f"Sustained high {rtype} utilization ({avg_util:.1f}%)",
+                        impact={
+                            "performance_degradation": "high",
+                            "request_failures": avg_util > 95,
+                            "user_impact": "significant",
+                        },
+                        resolution=[
+                            f"Increase {rtype} capacity on {node}",
+                            "Migrate workloads to other nodes",
+                            "Optimize resource-intensive operations",
+                            "Enable vertical scaling",
+                        ],
+                    )
+                )
+
+    async def _detect_allocation_bottlenecks(self):
+        """Detect allocation bottlenecks."""
+        # Check for poor allocation patterns
+        for key, metrics in self.metrics.items():
+            if len(metrics) < 20:
+                continue
+
+            node, rtype = key.split(":")
+
+            # Look for allocation/deallocation patterns
+            utils = [m.utilization for m in metrics]
+            changes = np.diff(utils)
+
+            # High variation suggests allocation issues
+            if np.std(changes) > 10:
+                self.bottlenecks.append(
+                    Bottleneck(
+                        bottleneck_type=BottleneckType.ALLOCATION,
+                        severity=min(np.std(changes) / 20, 1.0),
+                        edge_node=node,
+                        resource_type=ResourceType(rtype),
+                        description=f"Inefficient {rtype} allocation patterns",
+                        impact={
+                            "resource_waste": "moderate",
+                            "response_variance": "high",
+                            "efficiency": "low",
+                        },
+                        resolution=[
+                            "Implement resource pooling",
+                            "Use allocation caching",
+                            "Optimize allocation algorithms",
+                            "Review resource lifecycle management",
+                        ],
+                    )
+                )
+
+    async def _detect_contention_bottlenecks(self):
+        """Detect resource contention."""
+        # Look for patterns indicating contention
+        for key, metrics in self.metrics.items():
+            if len(metrics) < 30:
+                continue
+
+            node, rtype = key.split(":")
+
+            # Get wait times from metadata if available
+            wait_times = []
+            for m in metrics:
+                if "wait_time" in m.metadata:
+                    wait_times.append(m.metadata["wait_time"])
+
+            if wait_times and np.mean(wait_times) > 100:  # 100ms average wait
+                self.bottlenecks.append(
+                    Bottleneck(
+                        bottleneck_type=BottleneckType.CONTENTION,
+                        severity=min(np.mean(wait_times) / 500, 1.0),
+                        edge_node=node,
+                        resource_type=ResourceType(rtype),
+                        description=f"High {rtype} contention (avg wait: {np.mean(wait_times):.0f}ms)",
+                        impact={
+                            "latency_increase": f"{np.mean(wait_times):.0f}ms",
+                            "throughput_reduction": "significant",
+                            "user_experience": "degraded",
+                        },
+                        resolution=[
+                            "Implement resource locking optimization",
+                            "Use lock-free data structures",
+                            "Increase resource pool size",
+                            "Review concurrent access patterns",
+                        ],
+                    )
+                )
+
+    async def _detect_fragmentation_bottlenecks(self):
+        """Detect resource fragmentation."""
+        for key, metrics in self.metrics.items():
+            if not metrics:
+                continue
+
+            node, rtype = key.split(":")
+            recent = list(metrics)[-5:]
+
+            # Check for fragmentation indicators
+            for m in recent:
+                if m.used < m.total * 0.7 and m.available < m.total * 0.2:
+                    # Used is low but available is also low = fragmentation
+                    fragmentation_pct = (1 - (m.used + m.available) / m.total) * 100
+
+                    if fragmentation_pct > 10:
+                        self.bottlenecks.append(
+                            Bottleneck(
+                                bottleneck_type=BottleneckType.FRAGMENTATION,
+                                severity=min(fragmentation_pct / 30, 1.0),
+                                edge_node=node,
+                                resource_type=ResourceType(rtype),
+                                description=f"{rtype} fragmentation ({fragmentation_pct:.1f}%)",
+                                impact={
+                                    "wasted_resources": f"{fragmentation_pct:.1f}%",
+                                    "allocation_failures": fragmentation_pct > 20,
+                                    "efficiency": "reduced",
+                                },
+                                resolution=[
+                                    "Implement defragmentation routine",
+                                    "Use contiguous allocation strategies",
+                                    "Review resource allocation sizes",
+                                    "Enable resource compaction",
+                                ],
+                            )
+                        )
+                        break
+
+    async def _detect_anomalies(self):
+        """Detect resource anomalies."""
+        cutoff = datetime.now() - timedelta(seconds=self.history_window)
+
+        for key, metrics in self.metrics.items():
+            if len(metrics) < 20:
+                continue
+
+            node, rtype = key.split(":")
+
+            # Get historical data
+            historical = [m for m in metrics if m.timestamp < cutoff]
+            recent = [m for m in metrics if m.timestamp >= cutoff]
+
+            if len(historical) < 10 or len(recent) < 3:
+                continue
+
+            # Calculate statistics
+            hist_utils = [m.utilization for m in historical]
+            recent_utils = [m.utilization for m in recent]
+
+            mean = np.mean(hist_utils)
+            std = np.std(hist_utils)
+
+            # Check for anomalies
+            for i, util in enumerate(recent_utils):
+                z_score = (util - mean) / std if std > 0 else 0
+
+                if abs(z_score) > self.anomaly_threshold:
+                    self.anomalies.append(
+                        {
+                            "timestamp": recent[i].timestamp.isoformat(),
+                            "edge_node": node,
+                            "resource_type": rtype,
+                            "value": util,
+                            "expected_range": [mean - 2 * std, mean + 2 * std],
+                            "z_score": z_score,
+                            "severity": "high" if abs(z_score) > 4 else "medium",
+                            "description": f"Unusual {rtype} utilization: {util:.1f}% (expected: {mean:.1f}±{std:.1f}%)",
+                        }
+                    )
+
+    def _generate_summary(self) -> Dict[str, Any]:
+        """Generate analysis summary."""
+        # Calculate overall health score
+        pattern_score = 100 - len(self.patterns) * 10
+        bottleneck_score = 100 - sum(b.severity * 20 for b in self.bottlenecks)
+        anomaly_score = 100 - len(self.anomalies) * 5
+
+        health_score = max(
+            0, min(100, (pattern_score + bottleneck_score + anomaly_score) / 3)
+        )
+
+        return {
+            "health_score": health_score,
+            "health_status": self._get_health_status(health_score),
+            "total_patterns": len(self.patterns),
+            "total_bottlenecks": len(self.bottlenecks),
+            "total_anomalies": len(self.anomalies),
+            "critical_issues": len([b for b in self.bottlenecks if b.severity > 0.8]),
+            "top_recommendations": self._get_top_recommendations(),
+        }
+
+    def _get_health_status(self, score: float) -> str:
+        """Get health status from score."""
+        if score >= 90:
+            return "excellent"
+        elif score >= 75:
+            return "good"
+        elif score >= 60:
+            return "fair"
+        elif score >= 40:
+            return "poor"
+        else:
+            return "critical"
+
+    def _get_top_recommendations(self) -> List[str]:
+        """Get top recommendations."""
+        recommendations = []
+
+        # From patterns
+        for pattern in sorted(self.patterns, key=lambda p: p.confidence, reverse=True)[
+            :2
+        ]:
+            recommendations.extend(pattern.recommendations[:1])
+
+        # From bottlenecks
+        for bottleneck in sorted(
+            self.bottlenecks, key=lambda b: b.severity, reverse=True
+        )[:2]:
+            recommendations.extend(bottleneck.resolution[:1])
+
+        return recommendations[:5]  # Top 5 recommendations
+
+    def _calculate_priority(self, score: float) -> str:
+        """Calculate priority from score."""
+        if score >= 0.8:
+            return "critical"
+        elif score >= 0.6:
+            return "high"
+        elif score >= 0.4:
+            return "medium"
+        else:
+            return "low"