1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
|
package stats
import (
"math"
"time"
)
// PatternAnalysis represents the analysis of I/O patterns
type PatternAnalysis struct {
// Burst detection
BurstCount int
BurstDuration time.Duration
BurstThreshold int64
// Periodicity detection
IsPeriodic bool
PeriodDuration time.Duration
// Trend analysis
TrendSlope float64
TrendDirection string // "increasing", "decreasing", "stable"
}
// AnalyzePatterns analyzes the I/O patterns from the statistics
func (s *IOStats) AnalyzePatterns() *PatternAnalysis {
s.mu.RLock()
defer s.mu.RUnlock()
analysis := &PatternAnalysis{
BurstThreshold: 1000, // Configurable threshold
}
// Analyze bursts
analysis.analyzeBursts(s.bytesHistory, s.timestamps)
// Analyze periodicity
analysis.analyzePeriodicity(s.bytesHistory, s.timestamps)
// Analyze trends
analysis.analyzeTrend(s.bytesHistory, s.timestamps)
return analysis
}
func (a *PatternAnalysis) analyzeBursts(bytes []int64, timestamps []time.Time) {
if len(bytes) < 2 {
return
}
var currentBurstSize int64
var burstStart time.Time
inBurst := false
for i := 0; i < len(bytes); i++ {
if bytes[i] > a.BurstThreshold {
if !inBurst {
burstStart = timestamps[i]
inBurst = true
}
currentBurstSize += bytes[i]
} else if inBurst {
a.BurstCount++
a.BurstDuration += timestamps[i].Sub(burstStart)
inBurst = false
currentBurstSize = 0
}
}
}
func (a *PatternAnalysis) analyzePeriodicity(bytes []int64, timestamps []time.Time) {
if len(bytes) < 4 {
return
}
// Simple periodicity detection using autocorrelation
mean := 0.0
for _, b := range bytes {
mean += float64(b)
}
mean /= float64(len(bytes))
var maxCorrelation float64
var bestPeriod int
// Check for periods up to half the data length
for period := 1; period < len(bytes)/2; period++ {
correlation := 0.0
count := 0
for i := 0; i < len(bytes)-period; i++ {
correlation += (float64(bytes[i]) - mean) * (float64(bytes[i+period]) - mean)
count++
}
if count > 0 {
correlation /= float64(count)
if correlation > maxCorrelation {
maxCorrelation = correlation
bestPeriod = period
}
}
}
// If we found a strong correlation, consider it periodic
if maxCorrelation > 0.7 { // Threshold for periodicity
a.IsPeriodic = true
if bestPeriod > 0 && bestPeriod < len(timestamps)-1 {
a.PeriodDuration = timestamps[bestPeriod].Sub(timestamps[0])
}
}
}
func (a *PatternAnalysis) analyzeTrend(bytes []int64, timestamps []time.Time) {
if len(bytes) < 2 {
return
}
// Simple linear regression
var sumX, sumY, sumXY, sumX2 float64
n := float64(len(bytes))
for i := 0; i < len(bytes); i++ {
x := float64(timestamps[i].UnixNano())
y := float64(bytes[i])
sumX += x
sumY += y
sumXY += x * y
sumX2 += x * x
}
// Calculate slope
a.TrendSlope = (n*sumXY - sumX*sumY) / (n*sumX2 - sumX*sumX)
// Determine trend direction
if math.Abs(a.TrendSlope) < 0.1 {
a.TrendDirection = "stable"
} else if a.TrendSlope > 0 {
a.TrendDirection = "increasing"
} else {
a.TrendDirection = "decreasing"
}
}
|