SPECweb96 Result

Dell Computer Corporation:	PowerEdge 6350/500
Microsoft:	IIS 4.0 and SWC 1.1

SPECweb96 = 5404

Performance

Throughput (ops/sec)	Response (msec)
546	5.0
1092	5.3
1638	5.8
2184	6.4
2730	6.9
3276	7.5
3822	8.3
4368	9.7
4914	10.5
5404	20.2

Configuration

Availability Dates
Hardware	Mar-99
HTTP Software	Jul-99
Operating System	May-99

Hardware

Vendor Dell Computer Corporation

Model PowerEdge 6350/500

Processor 500MHz Pentium III Xeon

# of Processors 2

Primary Cache 16KBI+16KBD on chip

Secondary Cache 2MB(I+D) off chip

Other Cache None

Memory 4GB

Disk Subsystem 3 9GB 10K RPM Drive

Disk Controllers 1 Adaptec Ultra-SCSI

Other Hardware 2 Alteon ACEswitch 180

Software

Operating System Windows NT Server 4.0 EE SP4

File System Type NTFS

Other Software None

HTTP Software

Vendor Microsoft

HTTP Software IIS 4.0 and SWC 1.1

# of "threads" 1

Server Cache Microsoft Scalable Web Cache 1.1

Log Mode Binary CLF

Test Sponsor
Test Date	May-99
Tested By	Dell Computer Corporation
SPEC License	55

Network

# of Controllers 2

Network Controllers Alteon ACEnic PCI

# of Nets 2

Type of Nets Jumbo Frame Gigabit

Network Speed 1000Mb/s

MSL (sec) 30 (Non RFC1122)

Time-Wait (sec) 60 (Non RFC1122)

Load Generators

# of Load Generators 8

# of Processes Each 15

Model Dell Precision 410

Processor 450MHz Pentium II

# of Processors 2

Memory 128MB

Network Controller Alteon ACEnic PCI

Operating System Microsoft Windows NT 4.0 Workstation SP4

Compiler Microsoft Visual C++ 5.0

Benchmark Configuration

Maximum Requested Load 5460

Maximum Fileset Size (MB) 1616

Notes

OS Notes

Post-SP4 TCP/IP hotfix transport available May-99

Network Notes

Each ACEswitch 180 connects one card in the SUT and four clients.
MTU size = 9000 bytes.

Other Notes

IIS settings: listen port 81, no logging, expected hits fewer than 10000, no ISAPI filters, NT challenge/response disabled.
IIS Home Directory: Access for Read and Directory Browsing Allowed only, Application Settings permissions "none".
Network interrupts affinitized one NIC per CPU.
SWC configured with 1 thread per NIC affinitized to same CPU as NIC.
One 9GB 10KRPM disk for OS and pagefile, 2 9GB 10KRPM disks striped for web pages and logs.
Registry Settings:
- Tcpip/Parameters/MaxHashTableSize=0x10000, sets TCB hash table size to 65,536 entries, default is 512.
- Tcpip/Parameters/TcpTimedWaitDelay=0x3c, sets TIME_WAIT parameter to 60 seconds (non-RFC 1122), default is 240.
- Tcpip/Parameters/MaxFreeTwTcbs=580000, sets maximum number of TIME_WAIT tcbs to 580,000.
- NDIS/Parameters/ProcessorAffinityMask=0, forces processor that handled interrupt to also handle that DPC, default is any processor.
- Inetinfo/Parameters/ObjectCacheTTL=0, sets IIS object cache timeout to 0 seconds, default 10 minutes
- Inetinfo/Parameters/OpenFileInCache=0, sets number of open files in object cache to 0
- Inetinfo/Parameters/MaxPoolThreads=1, sets pool threads per processor, default is 10
- Inetinfo/Parameters/PoolThreadLimit=4, sets total pool threads per system to 4
For each Alteon NIC:
- Altndis(X)/Parameters/SendCoalMax=100, sets number of send buffer descriptors in send ring.
- Altndis(X)/Parameters/RecvCoalMax=20, sets number of frames received before notifying host.
- Altndis(X)/Parameters/SendCoalTicks=1000, sets number of microsec. before send buffers are returned to host.
- Altndis(X)/Parameters/RecvCoalTicks=1000, sets number of microsec. before recv buffers are given to host.
- Altndis(X)/Parameters/JumboFrames=1,enables Jumbo Frame support.
Alternate HTTP/HT.c used. Repeated client-side calls to gethostbyname() omitted, as approved
by SPEC. No change in measured server performance but reduction in client load.

Test Run Details

Detailed Statistics
Load	Throughput	Class	Target Mix	Actual Mix	Error Percent	Mean Response	Std Dev Response
546	546	0	35.0%	35.0%	0.00%	4.13	17.13
		1	50.0%	50.0%	0.00%	4.34	14.52
		2	14.0%	14.0%	0.00%	7.92	7.63
		3	1.0%	1.0%	0.00%	29.98	29.34
1092	1092	0	35.0%	35.0%	0.00%	4.34	7.94
		1	50.0%	50.0%	0.00%	4.63	13.41
		2	14.0%	14.0%	0.00%	8.34	8.04
		3	1.0%	1.0%	0.00%	31.18	30.59
1638	1638	0	35.0%	35.0%	0.00%	4.76	12.07
		1	50.0%	50.0%	0.00%	5.09	16.84
		2	14.0%	14.0%	0.00%	8.99	12.39
		3	1.0%	1.0%	0.00%	32.80	32.22
2184	2184	0	35.0%	35.0%	0.00%	5.30	13.53
		1	50.0%	50.0%	0.00%	5.59	10.39
		2	14.0%	14.0%	0.00%	9.76	9.45
		3	1.0%	1.0%	0.00%	34.73	34.11
2730	2730	0	35.0%	35.0%	0.00%	5.76	14.04
		1	50.0%	50.0%	0.00%	6.10	13.76
		2	14.0%	14.0%	0.00%	10.52	15.02
		3	1.0%	1.0%	0.00%	36.74	36.09
3276	3276	0	35.0%	35.0%	0.00%	6.23	9.15
		1	50.0%	50.0%	0.00%	6.63	12.65
		2	14.0%	14.0%	0.00%	11.26	15.30
		3	1.0%	1.0%	0.00%	38.92	38.26
3822	3822	0	35.0%	35.0%	0.00%	7.02	13.26
		1	50.0%	50.0%	0.00%	7.41	12.99
		2	14.0%	14.0%	0.00%	12.31	15.56
		3	1.0%	1.0%	0.00%	41.37	45.69
4368	4368	0	35.0%	35.0%	0.00%	8.25	12.99
		1	50.0%	50.0%	0.00%	8.65	11.67
		2	14.0%	14.0%	0.00%	14.19	15.75
		3	1.0%	1.0%	0.00%	45.61	44.80
4914	4914	0	35.0%	35.0%	0.00%	9.05	13.39
		1	50.0%	50.0%	0.00%	9.50	12.86
		2	14.0%	14.0%	0.00%	15.19	18.57
		3	1.0%	1.0%	0.00%	47.83	50.59
5460	5404		0.0%	0.0%	0.00%
			0.0%	0.0%	0.00%
			0.0%	0.0%	0.00%
			0.0%	0.0%	0.00%

Thu Jul 15 9:37:15 1999

First published at SPEC.org on 15-Jun-1999