Compaq ProLiant 1000 Performance Analysis and Tuning of Raptor's Eagle NT 3.06 - Page 21

Base Run, DNS Lookups for HTTPD and HTTP Cache - proliant 1500

Page 21 highlights

278A/0497 WHITE PAPER (cont.) ... Base Run The base system, test run 1, consists of the ProLiant 5000 system, 1Pentium Pro 200 MHz, 512 cache processor, 64Mb RAM, 2-EISA NetFlx-3 10/100, PCI S2-Array Controller Raid 0, 1 SCSI Disk, MaxReceive Buffers is 100, HTTPD cache is on, DNS Lookups for HTTPD is on, and 100Mb Network. The graph of the base system run is displayed in Graph 8. 2400 2100 1800 1500 TPM 1200 900 600 300 0 1 Base Run, HTTP Only, Run #19 12 24 32 36 48 56 72 Number of Virtual Clients Graph 8: Base Run, HTTP Only As you may notice HTTP only Transactions Per Minute are three times the TPM as seen with HTTP and FTP transactions. The overall TPM for 2 to 72 virtual clients 2 was 1966. As the load increases the TPM increases to a high 2146 TPM at 72 virtual clients. The failure rate was 0.0%. The calculated highest throughput is 15Mb for 72 virtual clients. DNS Lookups for HTTPD and HTTP Cache Run #20 turns the DNS Lookup for HTTP switch off and Run #21 turns off the HTTP Cache switch. Refer to Graph 9 for the results. DNS Lookups off yields in a 9% increase from the base where not caching rules for HTTP results in 7% degradation from the base system. Failures remained at 0.0%. Run #20 and Run #21 TPM 2400 2100 1800 1500 1200 900 600 300 0 1 12 24 32 36 48 56 72 V irtual C lients Base Run, HTTP O nly DNS Lookups O ff HTTP Cache Off Graph 9: Base Run, DNS Lookups Off, HTTP Cache Off 21

  • 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

W
HITE
P
APER
(cont.)
21
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
278A/0497
Base Run
The base system, test run 1, consists of the ProLiant 5000 system, 1Pentium Pro 200 MHz, 512
cache processor, 64Mb RAM, 2-EISA NetFlx-3 10/100, PCI S2-Array Controller Raid 0, 1 SCSI
Disk, MaxReceive Buffers is 100, HTTPD cache is on, DNS Lookups for HTTPD is on, and
100Mb Network.
The graph of the base system run is displayed in Graph 8.
Graph 8:
Base Run, HTTP Only
As you may notice HTTP only Transactions Per Minute are three times the TPM as seen with
HTTP and FTP transactions. The overall TPM for 2 to 72 virtual clients 2 was 1966. As the load
increases the TPM increases to a high 2146 TPM at 72 virtual clients. The failure rate was 0.0%.
The calculated highest throughput is 15Mb for 72 virtual clients.
DNS Lookups for HTTPD and HTTP Cache
Run #20 turns the DNS Lookup for HTTP switch off and Run #21 turns off the HTTP Cache
switch. Refer to Graph 9 for the results.
DNS Lookups off yields in a 9% increase from the base
where not caching rules for HTTP results in 7% degradation from the base system.
Failures
remained at 0.0%.
Graph 9:
Base Run, DNS Lookups Off, HTTP Cache Off
Base Run, HTTP Only, Run #19
0
300
600
900
1200
1500
1800
2100
2400
1
12
24
32
36
48
56
72
Number of Virtual Clients
TPM
0
300
600
900
1200
1500
1800
2100
2400
Base Run, HTTP
Only
DNS Lookups Off
HTTP Cache Off
1
12
24
32
36
48
56
72
Virtual Clients
Run #20 and Run #21 TPM