kailash 0.1.5__py3-none-any.whl → 0.2.1__py3-none-any.whl

kailash/workflow/cycle_analyzer.py

@@ -0,0 +1,889 @@
+"""
+Cycle Analysis and Performance Monitoring for Cyclic Workflows.
+
+This module provides comprehensive analysis tools that combine debugging and
+profiling capabilities to deliver deep insights into cycle behavior, performance
+characteristics, and optimization opportunities. It serves as the primary
+analysis interface for understanding and improving cyclic workflow execution.
+
+Examples:
+    Comprehensive cycle analysis:
+
+    >>> analyzer = CycleAnalyzer(
+    ...     analysis_level="comprehensive",
+    ...     output_directory="./analysis_results"
+    ... )
+    >>> # Start analysis session
+    >>> session = analyzer.start_analysis_session("optimization_study")
+    >>> # Analyze cycle execution
+    >>> trace = analyzer.start_cycle_analysis("opt_cycle", "workflow_1")
+    >>> analyzer.track_iteration(trace, input_data, output_data, 0.05)
+    >>> analyzer.complete_cycle_analysis(trace, True, "convergence")
+    >>> # Generate comprehensive report
+    >>> report = analyzer.generate_session_report()
+    >>> analyzer.export_analysis_data("analysis_results.json")
+
+    Real-time monitoring:
+
+    >>> # Monitor active cycle
+    >>> metrics = analyzer.get_real_time_metrics(trace)
+    >>> if metrics['health_score'] < 0.5:
+    ...     print("Performance issue detected!")
+    ...     print(f"Alerts: {metrics['alerts']}")
+"""
+
+import json
+import logging
+from datetime import datetime
+from pathlib import Path
+from typing import Any, Dict, List, Optional
+
+from kailash.workflow.cycle_debugger import CycleDebugger, CycleExecutionTrace
+from kailash.workflow.cycle_profiler import CycleProfiler
+
+logger = logging.getLogger(__name__)
+
+
+class CycleAnalyzer:
+    """
+    Comprehensive analysis tool combining debugging and profiling capabilities.
+
+    This class provides a unified interface for cycle analysis, combining
+    the detailed tracking capabilities of CycleDebugger with the performance
+    insights of CycleProfiler to provide comprehensive cycle optimization
+    guidance and health monitoring.
+
+    Examples:
+        >>> analyzer = CycleAnalyzer(analysis_level="comprehensive")
+        >>> # Start analysis
+        >>> session = analyzer.start_analysis_session("optimization_study")
+        >>> trace = analyzer.start_cycle_analysis("cycle_1", "workflow_1")
+        >>> # During execution...
+        >>> analyzer.track_iteration(trace, input_data, output_data)
+        >>> # Complete analysis
+        >>> analyzer.complete_cycle_analysis(trace, converged=True)
+        >>> report = analyzer.generate_comprehensive_report(session)
+    """
+
+    def __init__(
+        self,
+        analysis_level: str = "standard",
+        enable_profiling: bool = True,
+        enable_debugging: bool = True,
+        output_directory: Optional[str] = None,
+    ):
+        """
+        Initialize cycle analyzer.
+
+        Args:
+            analysis_level: Level of analysis ("basic", "standard", "comprehensive").
+            enable_profiling: Whether to enable performance profiling.
+            enable_debugging: Whether to enable detailed debugging.
+            output_directory: Directory for analysis output files.
+        """
+        self.analysis_level = analysis_level
+        self.enable_profiling = enable_profiling
+        self.enable_debugging = enable_debugging
+        self.output_directory = Path(output_directory) if output_directory else None
+
+        # Initialize components based on configuration
+        debug_level = {
+            "basic": "basic",
+            "standard": "detailed",
+            "comprehensive": "verbose",
+        }.get(analysis_level, "detailed")
+
+        self.debugger = (
+            CycleDebugger(debug_level=debug_level, enable_profiling=enable_profiling)
+            if enable_debugging
+            else None
+        )
+
+        self.profiler = (
+            CycleProfiler(enable_advanced_metrics=(analysis_level == "comprehensive"))
+            if enable_profiling
+            else None
+        )
+
+        # Analysis session tracking
+        self.current_session: Optional[str] = None
+        self.session_traces: List[CycleExecutionTrace] = []
+        self.analysis_history: List[Dict[str, Any]] = []
+
+        # Create output directory if specified
+        if self.output_directory:
+            self.output_directory.mkdir(parents=True, exist_ok=True)
+            logger.info(f"Analysis output directory: {self.output_directory}")
+
+    def start_analysis_session(self, session_id: str) -> str:
+        """
+        Start a new analysis session for grouping related cycles.
+
+        Analysis sessions allow grouping multiple cycle executions for
+        comparative analysis, trend identification, and comprehensive
+        reporting across related workflow executions.
+
+        Args:
+            session_id: Unique identifier for the analysis session.
+
+        Returns:
+            Session ID for reference.
+
+        Examples:
+            >>> session = analyzer.start_analysis_session("optimization_experiment_1")
+        """
+        self.current_session = session_id
+        self.session_traces = []
+
+        logger.info(f"Started analysis session: {session_id}")
+        return session_id
+
+    def start_cycle_analysis(
+        self,
+        cycle_id: str,
+        workflow_id: str,
+        max_iterations: Optional[int] = None,
+        timeout: Optional[float] = None,
+        convergence_condition: Optional[str] = None,
+    ) -> Optional[CycleExecutionTrace]:
+        """
+        Start analysis for a new cycle execution.
+
+        Begins comprehensive tracking for a cycle execution, including
+        debugging and profiling as configured. Returns a trace object
+        for tracking iteration progress.
+
+        Args:
+            cycle_id: Unique identifier for the cycle.
+            workflow_id: Parent workflow identifier.
+            max_iterations: Configured iteration limit.
+            timeout: Configured timeout limit.
+            convergence_condition: Convergence condition.
+
+        Returns:
+            Trace object for tracking, or None if debugging disabled.
+
+        Examples:
+            >>> trace = analyzer.start_cycle_analysis("opt_cycle", "workflow_1", max_iterations=100)
+        """
+        if not self.debugger:
+            logger.warning("Debugging not enabled - cannot create trace")
+            return None
+
+        trace = self.debugger.start_cycle(
+            cycle_id=cycle_id,
+            workflow_id=workflow_id,
+            max_iterations=max_iterations,
+            timeout=timeout,
+            convergence_condition=convergence_condition,
+        )
+
+        logger.info(
+            f"Started cycle analysis for '{cycle_id}' in session '{self.current_session}'"
+        )
+        return trace
+
+    def track_iteration(
+        self,
+        trace: CycleExecutionTrace,
+        input_data: Dict[str, Any],
+        output_data: Dict[str, Any],
+        convergence_value: Optional[float] = None,
+        node_executions: Optional[List[str]] = None,
+    ):
+        """
+        Track a single cycle iteration with input/output data.
+
+        Records detailed information about a cycle iteration including
+        timing, resource usage, convergence metrics, and execution flow
+        for comprehensive analysis.
+
+        Args:
+            trace: Active trace object.
+            input_data: Input data for the iteration.
+            output_data: Output data from the iteration.
+            convergence_value: Convergence metric if available.
+            node_executions: List of executed nodes.
+
+        Examples:
+            >>> analyzer.track_iteration(trace, input_data, output_data, convergence_value=0.05)
+        """
+        if not self.debugger:
+            return
+
+        iteration = self.debugger.start_iteration(trace, input_data)
+        self.debugger.end_iteration(
+            trace, iteration, output_data, convergence_value, node_executions
+        )
+
+        if self.analysis_level == "comprehensive":
+            logger.debug(
+                f"Tracked iteration {iteration.iteration_number} for cycle '{trace.cycle_id}' "
+                f"with convergence={convergence_value}"
+            )
+
+    def complete_cycle_analysis(
+        self,
+        trace: CycleExecutionTrace,
+        converged: bool,
+        termination_reason: str,
+        convergence_iteration: Optional[int] = None,
+    ):
+        """
+        Complete cycle analysis and generate insights.
+
+        Finalizes cycle tracking and performs comprehensive analysis
+        including performance metrics, optimization recommendations,
+        and comparative insights if multiple cycles are available.
+
+        Args:
+            trace: Cycle trace to complete.
+            converged: Whether the cycle converged successfully.
+            termination_reason: Why the cycle terminated.
+            convergence_iteration: Iteration where convergence occurred.
+
+        Examples:
+            >>> analyzer.complete_cycle_analysis(trace, converged=True, termination_reason="convergence")
+        """
+        if not self.debugger:
+            return
+
+        # Complete debugging
+        self.debugger.end_cycle(
+            trace, converged, termination_reason, convergence_iteration
+        )
+
+        # Add to profiler for performance analysis
+        if self.profiler:
+            self.profiler.add_trace(trace)
+
+        # Add to session traces
+        self.session_traces.append(trace)
+
+        logger.info(
+            f"Completed cycle analysis for '{trace.cycle_id}' - "
+            f"converged={converged}, iterations={len(trace.iterations)}"
+        )
+
+        # Generate immediate insights for comprehensive analysis
+        if self.analysis_level == "comprehensive":
+            self._generate_immediate_insights(trace)
+
+    def generate_cycle_report(self, trace: CycleExecutionTrace) -> Dict[str, Any]:
+        """
+        Generate comprehensive report for a single cycle.
+
+        Creates a detailed analysis report for a specific cycle execution
+        including debugging information, performance metrics, and
+        optimization recommendations.
+
+        Args:
+            trace: Completed cycle trace.
+
+        Returns:
+            Comprehensive cycle analysis report.
+
+        Examples:
+            >>> report = analyzer.generate_cycle_report(trace)
+            >>> print(f"Cycle efficiency: {report['performance']['efficiency_score']}")
+        """
+        report = {
+            "analysis_info": {
+                "cycle_id": trace.cycle_id,
+                "workflow_id": trace.workflow_id,
+                "analysis_level": self.analysis_level,
+                "session_id": self.current_session,
+                "generated_at": datetime.now().isoformat(),
+            }
+        }
+
+        # Add debugging information
+        if self.debugger:
+            debug_report = self.debugger.generate_report(trace)
+            report["debugging"] = debug_report
+
+        # Add profiling information
+        if self.profiler:
+            # Create temporary profiler for single trace analysis
+            single_profiler = CycleProfiler(
+                enable_advanced_metrics=(self.analysis_level == "comprehensive")
+            )
+            single_profiler.add_trace(trace)
+
+            performance_metrics = single_profiler.analyze_performance()
+            recommendations = single_profiler.get_optimization_recommendations(trace)
+
+            report["performance"] = performance_metrics.to_dict()
+            report["recommendations"] = recommendations
+
+        # Add analysis-level specific insights
+        if self.analysis_level == "comprehensive":
+            report["advanced_analysis"] = self._generate_advanced_analysis(trace)
+
+        # Export to file if configured
+        if self.output_directory:
+            self._export_cycle_report(report, trace.cycle_id)
+
+        return report
+
+    def generate_session_report(
+        self, session_id: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """
+        Generate comprehensive report for an analysis session.
+
+        Creates a detailed analysis report covering all cycles in a session,
+        including comparative analysis, trend identification, and overall
+        optimization recommendations.
+
+        Args:
+            session_id: Session to analyze, or current session if None.
+
+        Returns:
+            Comprehensive session analysis report.
+
+        Examples:
+            >>> report = analyzer.generate_session_report()
+            >>> print(f"Best cycle: {report['comparative_analysis']['best_cycle']}")
+        """
+        target_session = session_id or self.current_session
+        traces_to_analyze = self.session_traces if session_id is None else []
+
+        report = {
+            "session_info": {
+                "session_id": target_session,
+                "analysis_level": self.analysis_level,
+                "cycles_analyzed": len(traces_to_analyze),
+                "generated_at": datetime.now().isoformat(),
+            },
+            "summary": {
+                "total_cycles": len(traces_to_analyze),
+                "total_iterations": sum(
+                    len(trace.iterations) for trace in traces_to_analyze
+                ),
+                "convergence_rate": (
+                    len([t for t in traces_to_analyze if t.converged])
+                    / len(traces_to_analyze)
+                    if traces_to_analyze
+                    else 0
+                ),
+                "avg_cycle_time": (
+                    sum(t.total_execution_time or 0 for t in traces_to_analyze)
+                    / len(traces_to_analyze)
+                    if traces_to_analyze
+                    else 0
+                ),
+            },
+        }
+
+        if not traces_to_analyze:
+            report["warning"] = "No traces available for analysis"
+            return report
+
+        # Add profiling analysis
+        if self.profiler and traces_to_analyze:
+            # Ensure all traces are in profiler
+            for trace in traces_to_analyze:
+                if trace not in self.profiler.traces:
+                    self.profiler.add_trace(trace)
+
+            performance_report = self.profiler.generate_performance_report()
+            report["performance_analysis"] = performance_report
+
+        # Add comparative analysis
+        if len(traces_to_analyze) >= 2:
+            cycle_ids = [trace.cycle_id for trace in traces_to_analyze]
+            comparison = (
+                self.profiler.compare_cycles(cycle_ids) if self.profiler else {}
+            )
+            report["comparative_analysis"] = comparison
+
+        # Add session-specific insights
+        report["insights"] = self._generate_session_insights(traces_to_analyze)
+
+        # Export to file if configured
+        if self.output_directory:
+            self._export_session_report(report, target_session)
+
+        return report
+
+    def get_real_time_metrics(self, trace: CycleExecutionTrace) -> Dict[str, Any]:
+        """
+        Get real-time metrics for an active cycle.
+
+        Provides current performance metrics and health indicators
+        for a cycle that is currently executing, enabling real-time
+        monitoring and early intervention if issues are detected.
+
+        Args:
+            trace: Active cycle trace.
+
+        Returns:
+            Dict[str, Any]: Real-time metrics and health indicators
+
+        Side Effects:
+            None - this is a pure analysis method
+
+        Example:
+            >>> metrics = analyzer.get_real_time_metrics(trace)
+            >>> if metrics['health_score'] < 0.5:
+            ...     print("Cycle performance issue detected!")
+        """
+        if not trace.iterations:
+            return {"status": "no_iterations", "health_score": 0.5}
+
+        recent_iterations = trace.iterations[-5:]  # Last 5 iterations
+
+        # Calculate real-time performance indicators
+        avg_recent_time = sum(
+            iter.execution_time or 0 for iter in recent_iterations
+        ) / len(recent_iterations)
+
+        # Memory trend (if available)
+        memory_values = [
+            iter.memory_usage_mb for iter in recent_iterations if iter.memory_usage_mb
+        ]
+        memory_trend = "stable"
+        if len(memory_values) >= 2:
+            if memory_values[-1] > memory_values[0] * 1.2:
+                memory_trend = "increasing"
+            elif memory_values[-1] < memory_values[0] * 0.8:
+                memory_trend = "decreasing"
+
+        # Convergence trend
+        convergence_values = [
+            iter.convergence_value
+            for iter in recent_iterations
+            if iter.convergence_value
+        ]
+        convergence_trend = "unknown"
+        if len(convergence_values) >= 2:
+            if convergence_values[-1] < convergence_values[0]:
+                convergence_trend = "improving"
+            elif convergence_values[-1] > convergence_values[0]:
+                convergence_trend = "degrading"
+            else:
+                convergence_trend = "stable"
+
+        # Health score calculation
+        health_score = self._calculate_real_time_health_score(trace, recent_iterations)
+
+        return {
+            "status": "active",
+            "current_iteration": len(trace.iterations),
+            "avg_recent_iteration_time": avg_recent_time,
+            "memory_trend": memory_trend,
+            "convergence_trend": convergence_trend,
+            "health_score": health_score,
+            "alerts": self._generate_real_time_alerts(trace, recent_iterations),
+        }
+
+    def export_analysis_data(
+        self,
+        filepath: Optional[str] = None,
+        format: str = "json",
+        include_traces: bool = True,
+    ):
+        """
+        Export comprehensive analysis data.
+
+        Exports all analysis data including traces, performance metrics,
+        and reports for external analysis, archival, or sharing.
+
+        Args:
+            filepath (Optional[str]): Output file path, auto-generated if None
+            format (str): Export format ("json", "csv")
+            include_traces (bool): Whether to include detailed trace data
+
+        Side Effects:
+            Creates export file with analysis data
+
+        Example:
+            >>> analyzer.export_analysis_data("cycle_analysis.json", include_traces=True)
+        """
+        if filepath is None:
+            timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
+            filepath = f"cycle_analysis_{self.current_session or 'session'}_{timestamp}.{format}"
+            if self.output_directory:
+                filepath = str(self.output_directory / filepath)
+
+        export_data = {
+            "analysis_metadata": {
+                "session_id": self.current_session,
+                "analysis_level": self.analysis_level,
+                "export_timestamp": datetime.now().isoformat(),
+                "cycles_count": len(self.session_traces),
+            }
+        }
+
+        # Include session report
+        if self.session_traces:
+            export_data["session_report"] = self.generate_session_report()
+
+        # Include individual cycle reports
+        if include_traces:
+            export_data["cycle_reports"] = [
+                self.generate_cycle_report(trace) for trace in self.session_traces
+            ]
+
+        # Include performance history if available
+        if self.profiler:
+            export_data["performance_history"] = [
+                metrics.to_dict() for metrics in self.profiler.performance_history
+            ]
+
+        # Export to file
+        if format == "json":
+            with open(filepath, "w") as f:
+                json.dump(export_data, f, indent=2)
+        elif format == "csv":
+            # For CSV, export summary data only
+            import csv
+
+            with open(filepath, "w", newline="") as f:
+                writer = csv.writer(f)
+
+                # Write header
+                writer.writerow(
+                    [
+                        "cycle_id",
+                        "workflow_id",
+                        "iterations",
+                        "execution_time",
+                        "converged",
+                        "efficiency_score",
+                    ]
+                )
+
+                # Write cycle data
+                for trace in self.session_traces:
+                    stats = trace.get_statistics()
+                    writer.writerow(
+                        [
+                            trace.cycle_id,
+                            trace.workflow_id,
+                            len(trace.iterations),
+                            trace.total_execution_time,
+                            trace.converged,
+                            stats.get("efficiency_score", 0.0),
+                        ]
+                    )
+        else:
+            raise ValueError(f"Unsupported export format: {format}")
+
+        logger.info(f"Exported analysis data to {filepath} in {format} format")
+
+    def _generate_immediate_insights(self, trace: CycleExecutionTrace):
+        """Generate immediate insights for a completed cycle."""
+        stats = trace.get_statistics()
+
+        # Log key insights
+        if stats["efficiency_score"] > 0.8:
+            logger.info(
+                f"Excellent performance for cycle '{trace.cycle_id}' - efficiency: {stats['efficiency_score']:.2f}"
+            )
+        elif stats["efficiency_score"] < 0.3:
+            logger.warning(
+                f"Poor performance for cycle '{trace.cycle_id}' - efficiency: {stats['efficiency_score']:.2f}"
+            )
+
+        if not trace.converged:
+            logger.warning(
+                f"Cycle '{trace.cycle_id}' failed to converge - reason: {trace.termination_reason}"
+            )
+
+        # Check for performance issues
+        if stats["avg_iteration_time"] > 1.0:
+            logger.warning(
+                f"Slow iterations detected for cycle '{trace.cycle_id}' - avg: {stats['avg_iteration_time']:.3f}s"
+            )
+
+    def _generate_advanced_analysis(self, trace: CycleExecutionTrace) -> Dict[str, Any]:
+        """Generate advanced analysis insights for comprehensive mode."""
+        convergence_trend = trace.get_convergence_trend()
+
+        # Convergence pattern analysis
+        convergence_analysis = {}
+        if convergence_trend:
+            values = [value for _, value in convergence_trend if value is not None]
+            if len(values) >= 3:
+                # Calculate convergence velocity
+                velocity = (
+                    (values[0] - values[-1]) / len(values) if len(values) > 1 else 0
+                )
+                convergence_analysis = {
+                    "convergence_velocity": velocity,
+                    "convergence_pattern": (
+                        "fast"
+                        if velocity > 0.1
+                        else "slow" if velocity > 0.01 else "minimal"
+                    ),
+                    "stability_score": self._calculate_convergence_stability(values),
+                }
+
+        # Iteration pattern analysis
+        iteration_times = [
+            iter.execution_time for iter in trace.iterations if iter.execution_time
+        ]
+        iteration_analysis = {}
+        if iteration_times:
+            import statistics
+
+            iteration_analysis = {
+                "time_distribution": {
+                    "mean": statistics.mean(iteration_times),
+                    "median": statistics.median(iteration_times),
+                    "mode": (
+                        statistics.mode(iteration_times)
+                        if len(set(iteration_times)) != len(iteration_times)
+                        else None
+                    ),
+                    "skewness": self._calculate_skewness(iteration_times),
+                },
+                "performance_trend": self._analyze_performance_trend(iteration_times),
+            }
+
+        return {
+            "convergence_analysis": convergence_analysis,
+            "iteration_analysis": iteration_analysis,
+            "resource_efficiency": self._analyze_resource_efficiency(trace),
+        }
+
+    def _generate_session_insights(
+        self, traces: List[CycleExecutionTrace]
+    ) -> Dict[str, Any]:
+        """Generate insights across multiple cycles in a session."""
+        if not traces:
+            return {}
+
+        # Find best and worst performing cycles
+        cycle_scores = {
+            trace.cycle_id: trace.get_statistics()["efficiency_score"]
+            for trace in traces
+        }
+        best_cycle = max(cycle_scores.items(), key=lambda x: x[1])
+        worst_cycle = min(cycle_scores.items(), key=lambda x: x[1])
+
+        # Identify patterns
+        convergence_rate = len([t for t in traces if t.converged]) / len(traces)
+        avg_iterations = sum(len(t.iterations) for t in traces) / len(traces)
+
+        insights = {
+            "best_performing_cycle": {"id": best_cycle[0], "score": best_cycle[1]},
+            "worst_performing_cycle": {"id": worst_cycle[0], "score": worst_cycle[1]},
+            "overall_convergence_rate": convergence_rate,
+            "avg_iterations_per_cycle": avg_iterations,
+            "performance_consistency": best_cycle[1]
+            - worst_cycle[1],  # Lower is more consistent
+            "session_quality": (
+                "excellent"
+                if convergence_rate > 0.9 and cycle_scores[best_cycle[0]] > 0.8
+                else "good" if convergence_rate > 0.7 else "needs_improvement"
+            ),
+        }
+
+        return insights
+
+    def _calculate_real_time_health_score(
+        self, trace: CycleExecutionTrace, recent_iterations: List
+    ) -> float:
+        """Calculate real-time health score for an active cycle."""
+        score_components = []
+
+        # Performance component
+        if recent_iterations:
+            avg_time = sum(
+                iter.execution_time or 0 for iter in recent_iterations
+            ) / len(recent_iterations)
+            time_score = max(
+                0.0, 1.0 - min(1.0, avg_time / 2.0)
+            )  # Penalty after 2s per iteration
+            score_components.append(time_score)
+
+        # Error rate component
+        error_count = len([iter for iter in recent_iterations if iter.error])
+        error_score = (
+            max(0.0, 1.0 - (error_count / len(recent_iterations)))
+            if recent_iterations
+            else 1.0
+        )
+        score_components.append(error_score)
+
+        # Memory trend component (if available)
+        memory_values = [
+            iter.memory_usage_mb for iter in recent_iterations if iter.memory_usage_mb
+        ]
+        if memory_values and len(memory_values) >= 2:
+            memory_growth = (memory_values[-1] - memory_values[0]) / memory_values[0]
+            memory_score = max(
+                0.0, 1.0 - max(0.0, memory_growth)
+            )  # Penalty for memory growth
+            score_components.append(memory_score)
+
+        return (
+            sum(score_components) / len(score_components) if score_components else 0.5
+        )
+
+    def _generate_real_time_alerts(
+        self, trace: CycleExecutionTrace, recent_iterations: List
+    ) -> List[str]:
+        """Generate real-time alerts for potential issues."""
+        alerts = []
+
+        # Check for slow iterations
+        if recent_iterations:
+            avg_time = sum(
+                iter.execution_time or 0 for iter in recent_iterations
+            ) / len(recent_iterations)
+            if avg_time > 2.0:
+                alerts.append(f"Slow iterations detected: {avg_time:.2f}s average")
+
+        # Check for errors
+        error_count = len([iter for iter in recent_iterations if iter.error])
+        if error_count > 0:
+            alerts.append(
+                f"Errors detected in {error_count}/{len(recent_iterations)} recent iterations"
+            )
+
+        # Check for memory growth
+        memory_values = [
+            iter.memory_usage_mb for iter in recent_iterations if iter.memory_usage_mb
+        ]
+        if len(memory_values) >= 2:
+            memory_growth = (memory_values[-1] - memory_values[0]) / memory_values[0]
+            if memory_growth > 0.2:
+                alerts.append(
+                    f"Memory usage increasing: {memory_growth*100:.1f}% growth"
+                )
+
+        # Check for potential non-convergence
+        if len(trace.iterations) > (trace.max_iterations_configured or 100) * 0.8:
+            alerts.append(
+                f"Approaching max iterations: {len(trace.iterations)}/{trace.max_iterations_configured}"
+            )
+
+        return alerts
+
+    def _calculate_convergence_stability(self, values: List[float]) -> float:
+        """Calculate stability score for convergence values."""
+        if len(values) < 2:
+            return 1.0
+
+        import statistics
+
+        mean_val = statistics.mean(values)
+        if mean_val == 0:
+            return 1.0
+
+        stddev = statistics.stdev(values)
+        cv = stddev / mean_val  # Coefficient of variation
+
+        # Lower CV means more stable
+        return max(0.0, 1.0 - min(1.0, cv))
+
+    def _calculate_skewness(self, data: List[float]) -> float:
+        """Calculate skewness of data distribution."""
+        if len(data) < 3:
+            return 0.0
+
+        import statistics
+
+        mean_val = statistics.mean(data)
+        n = len(data)
+        variance = sum((x - mean_val) ** 2 for x in data) / n
+        if variance == 0:
+            return 0.0
+
+        std_dev = variance**0.5
+        skewness = sum((x - mean_val) ** 3 for x in data) / (n * std_dev**3)
+        return skewness
+
+    def _analyze_performance_trend(self, iteration_times: List[float]) -> str:
+        """Analyze performance trend over iterations."""
+        if len(iteration_times) < 3:
+            return "insufficient_data"
+
+        # Simple trend analysis
+        first_half = iteration_times[: len(iteration_times) // 2]
+        second_half = iteration_times[len(iteration_times) // 2 :]
+
+        import statistics
+
+        first_avg = statistics.mean(first_half)
+        second_avg = statistics.mean(second_half)
+
+        improvement = (first_avg - second_avg) / first_avg
+
+        if improvement > 0.1:
+            return "improving"
+        elif improvement < -0.1:
+            return "degrading"
+        else:
+            return "stable"
+
+    def _analyze_resource_efficiency(
+        self, trace: CycleExecutionTrace
+    ) -> Dict[str, Any]:
+        """Analyze resource usage efficiency."""
+        memory_values = [
+            iter.memory_usage_mb for iter in trace.iterations if iter.memory_usage_mb
+        ]
+        cpu_values = [
+            iter.cpu_usage_percent
+            for iter in trace.iterations
+            if iter.cpu_usage_percent
+        ]
+
+        efficiency = {}
+
+        if memory_values:
+            import statistics
+
+            efficiency["memory_efficiency"] = {
+                "peak_usage": max(memory_values),
+                "avg_usage": statistics.mean(memory_values),
+                "efficiency_score": max(
+                    0.0, 1.0 - (max(memory_values) / 2000)
+                ),  # Penalty after 2GB
+            }
+
+        if cpu_values:
+            import statistics
+
+            efficiency["cpu_efficiency"] = {
+                "peak_usage": max(cpu_values),
+                "avg_usage": statistics.mean(cpu_values),
+                "efficiency_score": min(
+                    1.0, statistics.mean(cpu_values) / 100
+                ),  # Higher CPU usage is better utilization
+            }
+
+        return efficiency
+
+    def _export_cycle_report(self, report: Dict[str, Any], cycle_id: str):
+        """Export cycle report to file."""
+        if not self.output_directory:
+            return
+
+        filename = (
+            f"cycle_report_{cycle_id}_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json"
+        )
+        filepath = self.output_directory / filename
+
+        with open(filepath, "w") as f:
+            json.dump(report, f, indent=2)
+
+        logger.debug(f"Exported cycle report to {filepath}")
+
+    def _export_session_report(self, report: Dict[str, Any], session_id: str):
+        """Export session report to file."""
+        if not self.output_directory:
+            return
+
+        filename = f"session_report_{session_id}_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json"
+        filepath = self.output_directory / filename
+
+        with open(filepath, "w") as f:
+            json.dump(report, f, indent=2)
+
+        logger.debug(f"Exported session report to {filepath}")