Performance Analysis for Web-based Enterprise System

1
Performance Analysisfor Web-based Enterprise
System
Mathematical Approach to

• JavaService Consulting
• 2005.05.10

2
Contents

• - Basic Concept
• - Request/Response System Modeling
• - Queuing Theory
• - Sizing Cases
• - n-Applications, Multiple-bottleneck
• - Benchmark Test
• - Performance Analysis

3
1. Basic Concept
???(m3/sec) ??(m/sec) ???(m2)
? (m3/sec) ?(m/sec) S(m2)
? (m/sec) pr2 (m2)
For ? ?, ?? or r ?, but limited by pressure (and
heat).
4
1.1 Objective
1. Basic Concept
1.How many Clients 2.Reasonable response
time 3.(Cost-effective)
5
1.2 Visit time Think time
1. Basic Concept
Click !
Click !
Click !
Click !
Click !
Click !
NewUser
Leave
Visit Time
Click !
Click !
Response Time
Think Time
Request Interval
Request Interval(sec) Response Time(sec)
Think Time(sec)
6
1.2.1 Think time
1. Basic Concept
????? Think Time ??? TM(Telemarket)
???(Teler?? ?? ???) 10-15? MIS ???? ???
15-20? ????? ??? 25-35? ?????? ??? 30-40?
??? ???? ??? ?? ? ? ??.
Click !
Click !
Response Time
Think Time
Request Interval
7
1.3 Concurrent User
1. Basic Concept
Internet
F/W
B
A
1,200-1,400 sessions
45Mbps x 2
L7(1)
Internet Clients
Inner users
900-1,200 sessions per F/W (Max 7000 per
F/W) Timeout 30 min
900-1,200 sessions per F/W (Max 7000 per F/W)
F/W
F/W
C
System operators
L7(2)
L7
1
600 concurrent sessions per server Timeout 15
min
Backend System
2
3
What is the number of concurrent users?
4
250 user x 3 750 users Timeout 15 min
5
IBM RS6000 SP (4-way, 2GB)
80 port EST 10 x 3 30
6
8
1.3.1 Concurrent User
1. Basic Concept
Web Servers access_log
RequestInterval( Think Time)
Visit Time
Different IP Addresses
NOTE ???????
10min
10min
Time
Concurrent Users 6?
9
1.3.2 Concurrent User Sample
1. Basic Concept
?????? 1,500?
K????? 275?
?? 1,000?
?????? 2,800?
10
1.4 Active User
1. Basic Concept
Concurrent User
Response Time
Time
Worker-Thread
Think Time
Active Service
Request Interval
Click
System
Click
Active User
11
1.5 Throughput
1. Basic Concept
Unit tph, tpm, tps, (pps, rps, ops) 3,600 tph
60 tpm 1 tps
Time
Measure Time
12
2.Request/Response System Model
1.Mathematical Approach 2.Queuing
Theory 3.Quantitative Analysis 4.Measuring
13
2.1 Request and Response
2. Request/Response System Model
Average Response Time
Assumption No ThinkTime
Throughput vs ActiveUesr
Question What is the ability of her ?
Right Answer Maximum throughput
14
2.2 Throughput Measuring
2. Request/Response System Model
Increasing
Response Time
Virtual User
(Assumption ThinkTime0)
Throughput(tps)
ResponseTime(sec)
Virtual User(ThinkTime0)
15
2.3 Littles law
2. Request/Response System Model
Response Time(sec)
Throughput(tps)
T
ResponseTime(sec)
R
ActiveUser(ThinkTime0)
N
ActiveUser
N T x R
Number of ActiveUser
Throughput(tps)
Average Response Time(sec)
Number of ActiveUser Throughput(tps) x Average
Response Time(sec)
16
2.4 ActiveUsers Law
2. Request/Response System Model
Concurrent User
Response Time
Think Time
Active User
Worker-Thread
Active Service
Request Interval
System
Click
ResponseTimes Law
Ave. ResponseTime(sec)
ActiveUser ConcurrentUser x
Ave. ResponseTime(sec) ThinkTime(sec)
17
2.4.1 Proof of ActiveUsers law
2. Request/Response System Model
ActiveUser
Throughput(tps)
Littles Law
Resp.Time(sec)
ConcurrentUser
Throughput(tps)
Request Interval(Resp.TimeThinkTime)
ActiveUser
ConcurrentUser

Resp.Time(sec)
Request Interval(Resp.TimeThinkTime)
Resp.Time(sec)
ActiveUser ConcurrentUser x
Request Interval(Resp.TimeThinkTime)
18
2.4.2 Meaning of Active User
1. Basic Concept
Ave. ResponseTime(sec)
ActiveUser ConcurrentUser x
Ave. ResponseTime(sec) ThinkTime(sec)
Service Rate
Request Rate
Active User
19
2.5 Equation summary
2. Request/Response System Model
Definition
Def1 Def2 Def3
Throughput(???)(tps) ????? / ????(time) Request
Interval(sec) ResponseTime(sec)
ThinkTime(sec) Concurrent User ActiveUser
InactiveUser
Equations
ActiveUser
Throughput(tps)
Littles Law
Resp.Time(sec)
ConcurrentUser
Throughput(tps)
Request Interval(Resp.TimeThinkTime)
Resp.Time(sec)
ActiveUser ConcurrentUser x
Request Interval(Resp.TimeThinkTime)
20
2.6 Throughput Formula
2. Request/Response System Model
ResponseTime
of ActiveUser
Throughput(tps)
Ave. Resp.Time(sec)
Max Worker-Thread 30
Throughput(tps) 30/2 15 tps
SourceBlcok A
2 sec
Throughput(tps) 30/3 10 tps
SourceBlcok B
3 sec
Throughput Formula
1
6
1
1
SourceBlcok A
Throughput(tps) 30/(32) 6 tps
23 5 sec
15
10
SourceBlcok B
21
2.6.1 Throughput Formula
2. Request/Response System Model
22
2.6.2 Throughput Formula
2. Request/Response System Model
23
2.6.3 Throughput Measure
2. Request/Response System Model
Program P
SourceBlock A
Throughput of ABC 6 tps
Servlet/JSP
SourceBlock B
SourceBlock C
Question Throughput of B ? tps
SourceBlock A
Deleted
SourceBlock B
SourceBlock C
1
6
1
1
15
x
Throughput of AC 15 tps
x 10 tps
24
3. Queuing Theory
G/G/1 G/G/m G/M/1 M/M/1 M/M/m M/G/1 G/M/m
Queuing Networks
Single-Class Open Queuing Networks Multiple-Class
Open Queuing Networks Closed Model Single-Class
Closed Model Multiple-Class Closed Model
25
3.1 Queuing Theory Concept
3. Queuing Theory
Arrival Rate (n/sec)
Service Time
Queue Size
Queuing Time
How many Queues ?
26
3.2 Queuing Theory - G/G/1
3. Queuing Theory
Infinite Queue Size
1 Queue
Arrival rate
Service rate
? µ
Arbitrary(infinite) Population
(?ltµ)
27
3.3 Queuing Theory - G/M/1
3. Queuing Theory
Finite Queue Size
1 Queue
Arrival rate
Service rate
? µ
Arbitrary(infinite) Population
28
3.4 Queuing Theory - G/G/1
3. Queuing Theory
Make sense ?
Throughput (?ltµ)
Arrival rate
Service rate
? µ
(?ltµ)
ActiveUser
Utilization
Response Time
29
3.5 ActiveUser/Resp.Time vs ConcurrentUser
3. Queuing Theory
Response Time
ActiveUser
Arrival Rate(?)
Arrival rate
Arrival rate
req/sec
Response Time(R)
Active User(N)
Concurrent User
?
?
Concurrent User
Concurrent User
30
3.6 Saturation Point
3. Queuing Theory
ResponseTime(sec)
Throughput(tps)
Critical Ave. Response Time
Saturation Point
p
ActiveUser(ThinkTime0)
31
3.6.1 Saturation Point
3. Queuing Theory
ResponseTime(sec)
Critical Ave. Response Time
Saturation Point
ActiveUser(ThinkTime0)
32
3.7 Maximum Concurrent User
3. Queuing Theory
ActiveUser
Throughput(tps)
Littles Law
Resp.Time(sec)
Resp.Time(sec)
ActiveUser ConcurrentUser x
Request Interval(Resp.TimeThinkTime)
ConcurrentUser
Throughput(tps)
Request Interval(Resp.TimeThinkTime)
ConcurrentUser ActiveUser Throughput(tps) x
ThinkTime(sec)
ThinkTime(sec)
ConcurrentUser ActiveUser x 1

Resp.Time(sec)
ConcurrentUser Throughput(tps) x
Resp.Time(sec)ThinkTime(sec)
33
3.7.1 Maximum Concurrent User
3. Queuing Theory
ThinkTime(sec)
ConcurrentUser ActiveUser Throughput(tps) x
ThinkTime(sec)
ThinkTime(sec)
ConcurrentUser ActiveUser x 1

Resp.Time(sec)
34
4. Capacity/Sizing Strategy
Saturation Point
Throughput ActiveUser(service) /
ResponseTime
35
4.1 Horizontal Scaling
4. Capacity/Sizing Strategy
T tps
Throughput T tps
Throughput Graph
Horizontal Scaling
n x T tps
ResponseTime - SaturationPoint
Throughput Graph
Throughput n x T tps
36
4.1.1 Horizontal Scaling
4. Capacity/Sizing Strategy
37
4.2 Vertical Scaling
4. Capacity/Sizing Strategy
T tps
t tpmC
Throughput Graph
Throughput T tps
Vertical Scaling
n x t tpmC
n x T tps
ResponseTime SaturationPoint
Throughput Graph
NOTE For only CPU bounded Job
Throughput n x T tps
38
4.2.1 Vertical Scaling
4. Capacity/Sizing Strategy
Horizontal - nxT - Expensive?
Vertical - nxT (Only CPU bounded App) - Cheap?
39
4.3 Distribution - Division
4. Capacity/Sizing Strategy
CPU 100
10 tps
2 sec
Saturation Point 20
Throughput 20 / 2 10 tps
2 sec
13.3 tps
CPU 100
CPU 70
3 sec
ActiveUser
Saturation Point 40
T
2 1 3 sec
ResponseTime
Throughput 40 / 3 13.3 tps
40
4.3.1 Division Bad Case
4. Capacity/Sizing Strategy
85 tps
85 tps
Servlet/JSP
EJB
SUN E10000(4-way,4GB)
Throughput 85 tps
CPU 100
Throughput Graph
Servlet/JSP
25 tps
SUN R420 (2-way,2GB)
EJB
25 tps
Servlet/JSP
SUN E10000(4-way,4GB)
SUN R420(2-way,2GB)
NOTE Remote EJB Call
Throughput 25 25 50 tps
41
4.5 Sizing Case Study 4.5.1 K? ???????
4. Capacity/Sizing Strategy
S80(12-way,8GB, 78,126 tpmC) S80(6-way,6GB,
41,140 tpmC) ?????? 4,506? Peak???????
275? Peak Arrival Rate 18.3 tps RequestInterval
18 sec ???? vs ???????? ?? 6.1 CPU ???
?(CPU??? ???) 119,266tpmC/18.3tps 6,517
tpmC/tps
42
4.5.2 ??
4. Capacity/Sizing Strategy
M80 (4-way, 3GB, 34,588 tpmC) x 2 S80 (12-way,
8GB, 78,126 tpmC) x 2 6F1 (4-way, 4GB, 44,500
tpmC) x 1 H80 (2-way, 1GB) x 8 ????M80(4-way)x2
69,176 tpmC ?????? 1,621?(??
3,435?) Peak??????? 600? ??????? 466,639? Peak
Arrival Rate 20.0 tps RequestInterval 30
sec Visit time 1?? 25? 43? ???? vs ???????? ??
31 CPU ??? 70-100 69,176tpmC/20tps 3,459
tpmC/tps 6F1(4-way,4GB, 44,500 tpmC )?? ???
CPU??? 60 113,676tpmC/20tps 5,684 tpmC/tps
43
4.5.3 ? ?????
4. Capacity/Sizing Strategy
M80(4-way, 4GB, 34,588 tpmC) x 3 ??????
96,753? Peak??????? 1,500-2,000? ???????
1,795,867? Peak Arrival Rate 58.3
tps RequestInterval 25.7 sec Visit time 6?
25? ????? ????? 18.6? ???? vs ???????? ??
1.6-2.0 CPU ??? 70 103,764 tpmC/58.3tps
1,780tpmC/tps
44
4.5.4 J?? CRM???
4. Capacity/Sizing Strategy
S85(12-way, 32GB, 124,818tpmC) X 2 HOST
DB ?????? 5,028 ? Peak??????? 250 ? ???????
264,060? Peak Arrival Rate 10
tps RequestInterval 25.3 sec Visit time 4?
10? ????? ????? 9.9 ? ???? vs ???????? ?? 4.97
CPU ??? 30-40 249,636tpmC / 10 tps 24,964
tpmC/tps
45
4.5.5 D?? e-Hanaro???
4. Capacity/Sizing Strategy
WAS 6H1(4-Way 4 GB, 40,763 tpmC) DB H70(4-Way
4GB, 17,134 tpmC) HOST CICS ?????? 2,800
?(???? 3,300?) Peak??????? 350 ? ???????
301,190 ? Peak Arrival Rate 33
tps RequestInterval 9 sec Visit time 16
? ????? ????? 120.5 ? ???? vs ???????? ?? 12.5
CPU ??? 50 (??App??) 40,763 tpmC / 33 tps
1,235 tpmC/tps (App??? CPU?? ?????)
46
4.5.6 ?? ???????
4. Capacity/Sizing Strategy
WAS H80(2-way, 2GB, 14,756(?) tpmC) DB
S80(12-way, 8GB, 67,908tpmC) ?????? 258
? Peak??????? 45 ? ??????? 65,192 ?(??????
20) Peak Arrival Rate 4.2 tps RequestInterval
11-18 sec ???? vs ???????? ?? 17.4 CPU
??? 50-60 (DB?? 35) (CPU?? ?? APP/Sybase SQL
??) 14,756 tpmC / 4.2 tps 3,513 tpmC/tps
47
4.5.7 K?? eCRM ??? ???
4. Capacity/Sizing Strategy
WAS H80 (2-way, 4GB, 14,756(?) tpmC) x 2 DB
M80(2way-4GB, 18,647(?) tpmC) ?????? 37,951
? Peak??????? 230 ? ??????? 235,527 ? Peak
Arrival Rate 5.83 tps RequestInterval 24.7
sec Visit time 2 ? 33 ? ????? ????? 6.2 ? ????
vs ???????? ?? 0.6 CPU ??? ?(CPU??
???) 29,512 tpmC / 5.83 tps 5,062 tpmC/tps
48
4.5.8 K?? ??? ??? ???
4. Capacity/Sizing Strategy
M80 (4-way, 4GB, 34,588 tpmC) x 9 WSBCC ,
Servlet/JSP, CTG ?????? 5,323 ? ????????
4,500? Peak??????? 2,800 ? ???????
6,198,133? Peak Arrival Rate 217
tps RequestInterval 23 sec ???? vs ???????? ??
53 CPU ??? ???? ?? 311,292tpmC / 217 tps
1,435 tpmC/tps 2?? Peak Arrival Rate 31.7
tps Peak??????? 600 ? 34,588 tpmC / 31.7 tps
1,091 tpmC/tps 4?? Peak Arrival Rate 50
tps 34,588 tpmC / 50 tps 692 tpmC/tps
CPU ???
49
4.5.9 Statistic Result?
4. Capacity/Sizing Strategy
?
Performance Aspect ? SI Project Scope
50
4.6 Capacity/Performance Approach
4. Capacity/Sizing Strategy
??
Pilot BMT
??/??
?????
?Open ?????
Perf. Predictio\n
Capacity Estimation
Unit App. Perf.Test
?????
BMT???
???? HVWS Simulator
??App ??
??App ??
???Open
????
App Tuning
Unit App. Perf.Test
Perf. Prediction
Perf. Data Logging
Perf Analysis
??? ?? ??\??
Perf Analysis
51
5. Unit Benchmark Test
SLA(Service Level Agreement) - Concurrent User
5,000 - Average Response Time less than 3 sec
How to prove or test it ?
52
5.1 Benchmark Test Bad Case
5. Unit Benchmark Test
??? ???? - ?? ?? ???(?) 5,000? - ??????? ???
3? ??
???? ??? ?? ???? - ?? ?????(?) 5,000 x 20
1,000 ? - ?????? Virtual User 1,000 ? ?? -
Think Time 0 - ?????? 3? ?? ??
??? ?? ?????? 20-30? ????
??? ??? ??? ?? !!
53
5.2 Refine SLA with ThinkTime
5. Unit Benchmark Test
SLA(Service Level Agreement) - Concurrent User
5,000 - Average Response Time less than 3 sec
ThinkTime Agreement - TM(Telemarket) ???
10-15? - MIS ???? ??? 15-20? - ????? ???
25-35? - ?????? ??? 30-40? - ??? ???? ??? ??
? ? ??.

?? ThinkTime ???
SLA(Service Level Agreement) - Concurrent User
5,000 - Think Time 30 (Example) - Average
Response Time less than 3 sec
54
5.3 StressTest with ThinkTime 30
5. Unit Benchmark Test
Throughput Graph
235.3
Throughput(tps)
Response Time
158.7
AverageResponseTime Graph
4.0
1.5
5,000
8,000
Virtual User(ThinkTime30 ? ConcurrentUser)
ConcurrentUser
Throughput(tps)
AverageResposeTime ThinkTime(30)
55
5.3.1 StressTest with ThinkTime0
5. Unit Benchmark Test
Throughput Graph
235.3
AverageResponseTime Graph
Response Time
Throughput(tps)
158.7
4.0
1.5
941
238
Virtual User(ThinkTime0 ? ActiveUser)
ConcurrentUser 5,000 8,000
ConcurrentUser ActiveUser Throughput(tps) x
ThinkTime(sec)
ThinkTime(sec)
ConcurrentUser ActiveUser x 1

Resp.Time(sec)
ConcurrentUser Throughput(tps) x
Resp.Time(sec)ThinkTime(sec)
56
6. Multiple Applications

• - Different Hit Ratio
• - Different Performance
• Homogeneous/heterogeneous
• Bottleneck Condition

57
6.1 2-Applications
6. Multiple Applications
TPSmax
A application
B application
TPS
TPS
40
40
40
20
20
20
Saturation Point
Saturation Point
A
B
Active Threads
40
Active Threads
40
1 1
3 1
1 3
TPS
TPS
TPS
40
40
40
A
30
A
B
20
15
B
10
10
B
5
A
Active Threads
40
10
30
Active Threads
40
10
30
Active Threads
40
20
TPSmax
TPSmax
TPSmax
40
40
40
30
20
20
15
10
10
5
A
B
A
B
A
B
Graph 1
Graph 2
Graph 3
58
6.1.1 2-Applications
6. Multiple Applications
A
40
Different Arrival Rate
graph 2
30
graph 1
20
graph 3
10
Limited by same resource bottleneck
? Homogeneous bottleneck condition
20
10
O
B
59
6.1.2 2-Applications
6. Multiple Applications
y
A r1 (req/sec) B r2 (req/sec)
T2
y
x

r2
r1
R2
Q
(r1,r2)
x
y
r2
1
P

T1
T2
x
O
T1
r1
R1
60
6.2 3-Applications
6. Multiple Applications
z
(0, 0, T3)
T3
y
z
x


r1
r2
r3
Q
P
r3
(r1, r2, r3)
(0, T2, 0)
(0, 0)
O
r2
y
T2
r1
y
x
z
T1

1

T1
T2
T3
(T1, 0, 0)
x
61
6.3 n-Applications
6. Multiple Applications
NOTE Homogeneous Bottleneck Condition
62
6.4 Performance Utilization
6. Multiple Applications
z
App. Ti TPSmax ri req/sec ri / Ti
x 400 48.2 0.12
y 9 1.2 0.13
z 50 22.7 0.45
T3
x
y
z

1

T1
T2
T3
Q
P
0.12 0.13 0.45 0.70 ? ri / Ti x 100
70 Critical Perf. Utilization (???????)
y
T2
T1
? ri / Ti 1.0
x
63
6.5 Hit Ratio Performance Utilization
6. Multiple Applications
App Ti TPSmax ri req/sec ri / Ti
a1 42 5.4 0.129
a2 9 1.9 0.210
a3 70 10.7 0.153
a4 15 4.5 0.300
Sub ?ri 22.5 ? ri / Ti 0.792

TOTAL TOTAL 25(req/s)
Arrival Rate
a.jsp 5.4 req/sec b.jsp 1.9 req/sec c.jsp
10.7 req/sec d.jsp 4.5 req/sec ..
? ri / Ti ? 1.0 (? is total hit ratio of
?ai) ? ? ri / Ti / ? ? ri / Ti /
(22.5/ 25) 88
64
6.6 Multiple Bottleneck Theory
6. Multiple Applications
z
Bottleneck surface
Arrival rate line
x
y
z


r2
r3
r1
Q
P
r3
(r1, r2, r3)
bottleneck saturation flat
(0, 0)
r2
O
? ri / Ti 1.0 e(egt0)
Heterogeneous bottleneck condition
x
65
7. Performance Analysis
Workload Patterns
Test Users -Arrival Rate -ThinkTime
Modeling StressTest
Real Performance Prediction
66
7.1 Performance Analysis
7. Performance Analysis
??? ????? ???? ???
??????? (?????) 65,192?
????????? (TPH) 8,500?(?? ?? 13)
Peak? ???????? (TPM) 150tpm (2.5tps)
Peak? ???????? (TPMmax) 250tpm (4.2tps)
??????? 258?
Peak? ?????????? 40?
Peak? ?????????? 46?
???? ??????(sec) (RequestInterval) ????????/TPS 40(?)/2.5(tps) 16(sec)
???? ??????(sec) (Visit Time) ???? T(sec)??? ??, T/(?????/??????????)
???? ??????? ???? ??? ??, ?????/????? 65,192?/258? 252?
?? ????? ???? ???
???????? ???
67
7.2 Arrival Rate Information
7. Performance Analysis
Performance Matrix
?? ????? ??? ???
?
N
???????? ???
Arrival Rate ?
RequestInterval 28.34 sec
68
7.3 Performance Matrix Stress Test
7. Performance Analysis
NOTE 82 Rule
App. Arrival Rate(tps) Arrival Ratio Max Throughput ?i /?i
a1 ?1 ?1 /??1 ?1 ?1/?1
a2 ?2 ?2 /??2 ?2 ?2/?2
a3 ?3 ?3 /??3 ?3 ?3/?3

an ?n ?n /??n ?n ?n/?n
(Others) ?others 1-?(Ex.20) N/A N/A
SUM ? ?i ?others ? ??i 0.2 1.0 N/A ? ?i/?i
? ?i/?i
? ?i/?i ?(1.0e) (e0) (? is total hit ratio of
?ai)
Critical Performance Utilization
1.0
?(1.0e)
(When ConcurrentUser N)
(e 0)
69
7.4 Performance Prediction
7. Performance Analysis
Arrival Rate
Concurrent Users
kN
k?
Future
Future
N
Current
Current
?
? ?i/?i
x k
1.0 (100)
Current Perf. Utilization
?(1.0e)
70
7.5 Max. Concurrent User Prediction
7. Performance Analysis
? ?i/?i
1.0 (100)
x k
?(1.0e)
?(1.0e)
(if e0)
k
? ?i/?i
?
(if e 0)
k
? ?i/?i
Maximum Concurrent User kN
Maximum Arrival Rate k? (hit/sec)
71
8. n-Applications Benchmark Test
- Web-Stress Test Tool - Arrival Rate - ThinkTime
72
8.1 Virtual User Simulation Test
8. Multiple Applications Benchmark Test
App. Ratio ?i
a1 0.5
a2 0.2
a3 0.1

(Others) 0.2(20)
SUM 0.8(80)
RequestInterval Resp.Time ThinkTime ? ThinkTim
e 28.5 sec(Example)
Response Time Graph
942
Virtual User(ThinkTime 28.5 sec)
73
8.2 n-Applications Benchmark Test
8. Multiple Applications Benchmark Test
App. Ratio ?i Req/sec ? TPS Ti ?i /Ti
a1 0.5 0.5? ?1 T1400 ?1/T1 0.5?/400 0.00125?
a2 0.2 0.2? ?2 T2 9 ?2/T2 0.2?/9 0.02222?
a3 0.1 0.1? ?3 T3 50 ?3/T3 0.1?/50 0.00200?
(Others) 0.2(20) 0.2? ?others N/A
SUM 0.8(80) ? ?i? N/A ? ?i /Ti (0.5/4000.2/90.1/50) ?
? ?i /Ti ?(1.0e) (e0) ? (0.5/4000.2/90.1/5
0) ? ?(1.0e)0.8 ? 0.8/0.5/4000.2/90.1/50
31.4 (req/sec)
ConcurrentUser Throughput x
RequestInterval ? 31.4(req/sec) 30(sec) 942?
??????? ???? (?, 30??? ???? ??)
74
Summary Q/A
- Throughput, ResponseTime, VisitTime,
ThinkTime - ConcurrentUser, ActiveUser - Littles
law, Performance Formula - Queuing Theory -
Sizing Case Study - Multiple bottleneck Queuing
Theory - ? ?i / Ti ?(1.0 e) (egt0) - Web
site Performance Analysis - Benchmark Test
Approach