SPECweb96 Result

Hewlett-Packard:	NetServer LXe Pro 6/200
Microsoft:	Internet Information Server 4.0

SPECweb96 = 1556

Performance

Throughput (ops/sec)	Response (msec)
160	6.2
320	6.3
480	7.0
640	7.8
800	8.9
960	10.1
1120	11.4
1280	13.5
1440	16.5
1556	25.3

Configuration

Availability Dates
Hardware	Aug-97
HTTP Software	Jun-98
Operating System	Dec-97

Hardware

Vendor Hewlett-Packard

Model NetServer LXe Pro 6/200

Processor 200Mhz PPro

# of Processors 2

Primary Cache 8KBI+8KBD on chip

Secondary Cache 1MB(I+D) off chip

Other Cache None

Memory 2.0 GB

Disk Subsystem 5 4GB 10K RPM Drives

Disk Controllers 2 Ultra-SCSI

Other Hardware None

Software

Operating System NT 4.0 SP3

File System Type NTFS

Other Software None

HTTP Software

Vendor Microsoft

HTTP Software Internet Information Server 4.0

# of "threads" 1

Server Cache See Notes

Log Mode Combined

Test Sponsor
Test Date	Jun-98
Tested By	Hewlett-Packard
SPEC License	3

Network

# of Controllers 4

Network Controllers DEC DEFPA-MB

# of Nets 4

Type of Nets FDDI over Cat5 UTP

Network Speed 100Mb/s

MSL (sec) 30 (Non RFC1122)

Time-Wait (sec) 60 (Non RFC1122)

Load Generators

# of Load Generators 8

# of Processes Each 8

Model Hewlett-Packard Kayak XU

Processor 300MHz PentiumII

# of Processors 2

Memory 64MB

Network Controller DEC DEFPA FDDI over UTP

Operating System NT 4.0 Workstation SP3

Compiler Visual C++ 5.0

Benchmark Configuration

Maximum Requested Load 1600

Maximum Fileset Size (MB) 873

Notes

OS Notes

Microsoft IIS 4.0 hotfixes applied.
12/97 TCP/IP and AFD Transport.
hotfix(ftp://ftp.microsoft.com/bussys/winnt/winnt-public/fixes/usa/NT40/hotfixes-postSP3/)
IIS 4.0 logfix 4/98 (ftp://ftp.microsoft.com/bussys/winnt/winnt-public/fixes/usa/NT40/hotfixes-postSP3/)

Network Notes

The DEC DEFPA-MB cards are dual attach station (DAS) FDDI over category 5 twisted pair.
No hubs were employed, each of the 4 networks is connected as a ring using both attachments.
Each network included one card in the SUT and 1-2 of the clients.
client 1 on 1st segment, clients 2 and 3 on 2nd, 4 and 5 on 3rd, 6 on 4th.
MTU size = 4352 bytes.

Other Notes

Performance Boost for Foreground Application=NONE ---Do not raise the priority of the foreground application, default is halfway.
Disable IIS Windows NT Challenge/Response ---Disables IIS user information encryption, default is enabled.
Added Registry Entry InetInfo/Parameters/ObjectCacheTTL=0xea60 ---Keeps IIS open descriptors active at least 16.6 hours, default is 30 sec.
Added Registry Entry InetInfo/Parameters/MemoryCacheSize=0x400000 ---Ups the cache open descriptor table to 32MB, default is 4MB.
Added Registry Entry InetInfo/Parameters/ListenBacklog=0x19 ---Specifies maximum active connections to hold in server queue, default is 15.
Added Registry Entry InetInfo/Parameters/MaxPoolThreads=1 ---Specifies maximum network request threads per processor, default is 10.
Added Registry Entry InetInfo/Parameters/PoolThreadLimit=2 ---Specifies maximum pool threads in system, default is 2 * #MB.
Added Registry Entry Tcpip/Parameters/MaxFreeTcbs=0x1d4c0 ---Increases the TCB timewait table to 120000 entries, default is 2000.
Added Registry Entry Tcpip/Parameters/MaxHashTableSize=0x10000 ---Sets TCB hash table size to 65,536 entries, default is 512.
Added Registry Entry Tcpip/Parameters/TcpTimedWaitDelay=0x3c ---Sets TIME_WAIT parameter to 60 seconds (non-RFC 1122), default is 240.
Set NDIS/Parameters/ProcessorAffinityMask to 0 ---Forces processor that handled interrupt to also handle that DPC, default is any processor.
One 4GB 7200RPM disk for OS, 3 4GB 10KRPM disk for web pages and logs, 1 4GB 10KRPM Drive for Pagefile
Network interrupts affinitized to one processor.

Test Run Details

Detailed Statistics
Load	Throughput	Class	Target Mix	Actual Mix	Error Percent	Mean Response	Std Dev Response
160	160	0	35.0%	35.0%	0.00%	3.95	6.40
		1	50.0%	49.9%	0.00%	4.73	7.42
		2	14.0%	14.0%	0.00%	11.07	22.60
		3	1.0%	1.0%	0.00%	87.50	98.70
320	320	0	35.0%	34.9%	0.00%	4.33	13.28
		1	50.0%	50.0%	0.00%	5.04	15.58
		2	14.0%	14.0%	0.00%	11.22	11.13
		3	1.0%	1.0%	0.00%	70.61	72.43
480	480	0	35.0%	35.0%	0.00%	4.95	11.42
		1	50.0%	50.0%	0.00%	5.60	5.49
		2	14.0%	14.0%	0.00%	12.54	20.45
		3	1.0%	1.0%	0.00%	71.30	71.17
640	640	0	35.0%	35.0%	0.00%	5.61	10.57
		1	50.0%	50.0%	0.00%	6.33	17.43
		2	14.0%	14.0%	0.00%	13.78	19.44
		3	1.0%	1.0%	0.00%	75.09	74.70
800	800	0	35.0%	35.0%	0.00%	6.61	15.13
		1	50.0%	50.0%	0.00%	7.25	13.54
		2	14.0%	14.0%	0.00%	15.27	15.26
		3	1.0%	1.0%	0.00%	81.58	81.25
960	960	0	35.0%	35.0%	0.00%	7.63	14.62
		1	50.0%	50.0%	0.00%	8.23	11.80
		2	14.0%	14.0%	0.00%	17.45	26.23
		3	1.0%	1.0%	0.00%	88.20	87.91
1120	1120	0	35.0%	35.0%	0.00%	8.72	14.64
		1	50.0%	50.0%	0.00%	9.36	14.61
		2	14.0%	14.0%	0.00%	19.55	20.30
		3	1.0%	1.0%	0.00%	97.55	97.38
1280	1280	0	35.0%	35.0%	0.00%	10.49	16.24
		1	50.0%	50.0%	0.00%	11.20	22.44
		2	14.0%	14.0%	0.00%	22.89	25.01
		3	1.0%	1.0%	0.00%	106.55	106.11
1440	1440	0	35.0%	35.0%	0.00%	13.12	19.30
		1	50.0%	50.0%	0.00%	13.75	18.14
		2	14.0%	14.0%	0.00%	27.21	32.87
		3	1.0%	1.0%	0.00%	119.48	118.89
1600	1556	0	35.0%	35.0%	0.07%	21.54	22.90
		1	50.0%	50.0%	0.07%	22.27	23.11
		2	14.0%	14.0%	0.07%	37.66	38.61
		3	1.0%	1.0%	0.05%	140.49	139.31

Mon Jul 6 19:58:50 1998

First published at SPEC.org on 30-Jun-1998