SPECsfs2008_nfs.v3 Result

Throughput (ops/sec) Response (msec) 148 1.7 300 2.3 450 2.9 601 3.2 750 3.8 904 5.3 1054 6.9 1207 9.2 1356 12.0 1470 18.1

Tested By	NSPLab(SM) Performed Benchmarking
Product Name	SPECsfs2008 Reference Platform (NFSv3)
Hardware Available	January 2003
Software Available	November 2007
Date Tested	December 2007
SFS License Number	2851
Licensee Locations	El Paso, TX USA

The PowerEdge 1600SC Server was setup as a small workgroup file server using AX100 and AX150 arrays to provide RAID 5 storage. The storage was directly connected to the server.

Item No	Qty	Type	Vendor	Model/Name	Description
1	1	Server	Generic	PowerEdge 1600SC	PowerEdge 1600SC server with single Xeon 2.4GHz processor, running Red Hat Enterprise Linux 5.1 software.
2	2	HBA	Generic	LightPulse	Generic LightPulse 4Gb/s LPe1105-n dual-channel HBA
3	1	Disk Controller	Generic	AX100	This array includes 12 146GB/7200 rpm SATA drives.
4	1	Disk Controller	Generic	AX150	This array includes 12 230GB/7200 rpm SATA drives.
5	1	Ethernet Switch	Generic	PowerConnect 2616	16 port Gigabit Ethernet Switch

OS Name and Version	Linux RHEL 5.1
Other Software	Linux RHEL 5.1 kernel 2.6.16
Filesystem Software	EXT3

Name	Value	Description
net.core.rmem_max	1048576	The maximum receive socket buffer size
net.core.wmem_max	1048576	The maximum send socket buffer size
net.core.wmem_default	1048576	The default send socket buffer size
net.core.rmem_default	1048576	The default receive socket buffer size
vm.min_free_kbytes	40960	Force the VM to keep a minimum number of kilobytes free
vm.dirty_background_ratio	1	The percentage of total memory at which the pdflush background writeback daemon will start writing out dirty data.
vm.dirty_expire_centisecs	1	Data that has been in memory for longer than this interval will be written out next time a pdflush daemon wakes up. Expressed in 100ths of a second.
vm.dirty_writeback_centisecs	99	The interval between wakeups of the pdflush daemons. Expressed in 100th of a second.
vm.dirty_ratio	10	The percentage of total memory at which a process that is generating disk writes will start writing out dirty data.

Server Tuning Notes

None

Description	Number of Disks	Usable Size
AX100 array configuration includes 12 146GB disks configured as (10+1)RAID-5 bound as a single LUN on which a 1.4TB EXT3 file system is built. One disk is used as a Hot Spare.	12	1.4 TB
AX150 array configuration includes 12 230GB disks configured as (10+1)RAID-5 bound as a single LUN on which a 2.3TB EXT3 file system is built. One disk is used as a Hot Spare.	12	2.2 TB
This is an internal SCSI disk drive used for the Linux OS for system use.	1	146.0 GB
Total	25	3.8 TB

Number of Filesystems	2
Total Exported Capacity	3.3 TB
Filesystem Type	EXT3
Filesystem Creation Options	EXT3
Filesystem Config	A filesystem is created on a LUN on the AX100 and another on a LUN on the AX150. The filesystems are fs1 and fs2.
Fileset Size	172.0 GB

Each of the arrays, AX100 and AX150 contained 12 disks that were configured as (10+1) RAID-5 groups with 64KB RAID stripe size. The 12th disk was configured as hot spare.

Item No	Network Type	Number of Ports Used	Notes
1	Gigabit Ethernet	1	Used Gigabit Ethernet port on motherboard

Network Configuration Notes

All network interfaces were connected to a the switch.

The PowerEdge 1600SC server single on-board network interface was connected to the switch. Each of the LG single on-board network interface was connected to the switch.

Item No	Qty	Type	Description	Processing Function
1	1	CPU	Intel Xeon 2.4 Ghz, 1 chip, 1 core per chip, 512KB of (I+D) L2 cache.	NFS protocol, EXT3 filesystem
2	4	CPU	1.2 Ghz Intel Celeron, 2 per AX100/AX150 disk array	RAID processing

Processing Element Notes

None

Description	Size in GB	Number of Instances	Total GB	Nonvolatile
NAS server main memory	4	1	4	V
Storage Processor main memory	1	2	2	NV
Grand Total Memory Gigabytes			6

Memory Notes

Each storage array has dual storage processor units that work as an active-active failover pair. The mirrored write cache is backed up with a battery unit capable of saving the write cache to disk in the event of a power failure. In the event of a storage array failure, the second storage processor unit is capable of saving all state that was managed by the first (and vise versa), even with a simultaneous power failure. When one of the storage processors or battery units are off-line, the system turns off the write cache and writes directly to disk before acknowledging any write operations. The battery unit can support the retention of data for 2 minutes which is sufficient to write all necessary data twice. Storage processor A could have written 99% of its memory to disk and then fail. In that case storage processor B has enough battery to store its copy of A's data as well as its own.

2 NFS file systems were used. Each RAID group was bound in a single LUN on each array. Each file system was configured on one LUN: fs1 on AX150 and fs2 on AX100. Each array had 2 Fibre Channel connections to the server although only one connection was enabled. In this configuration, NFS stable write and commit operations are not acknowledged until after the storage array has acknowledged that the related data has been stored in stable storage (i.e. NVRAM or disk).

See above for the tuning params.

A minimal amount of Linux tuning was performed to get the benchmark to run successfully. The goal was to measure performance as close to out of the box (default tunings) as possible.

Item No	Qty	Vendor	Model/Name	Description
1	4	Generic	430SC	Workstation with single Dual Core Xeon 2.8 GHz with 1 GB RAM

LG Type Name	LG1-4
BOM Item #	1
Processor Name	Xeon Dual Core
Processor Speed	2.8 GHz
Number of Processors (chips)	1
Number of Cores/Chip	2
Memory Size	1 GB
Operating System	Linux RHES 4.2
Network Type	BCM95751

Benchmark Parameters

Network Attached Storage Type	NFS V3
Number of Load Generators	4
Number of Processes per LG	8
Biod Max Read Setting	2
Biod Max Write Setting	2
Block Size	AUTO

Testbed Configuration

LG No	LG Type	Network	Target Filesystems	Notes
1..4	LG1	1	/fs1,/fs2	N/A

Load Generator Configuration Notes

Both filesystems were mounted on all clients, which were connected to the same physical and logical network.

Each client has the both file systems mounted from each of the 2 storage arrays: AX150 and AX100.

(No other report notes)

• Config Diagram