Test Methodology
Network Storage Configurations |
|||||
---|---|---|---|---|---|
CPU | 1.6GHz Marvell |
1.3GHz Marvell Armada CP (dual-core) |
2.13GHz Intel Atom D2700 (dual-core, 1MB cache) |
1.86GHz Intel Atom D2550 (dual-core, 1MB cache) |
1.8GHz Intel Atom D525 (dual-core, 1MB cache) |
Memory | TBC |
512MB DDR3 |
1GB DDR3 |
2GB DDR3 |
2GB DDR3 |
Disk drives | 4x 1TB Seagate ST31000528AS (included) |
4x 1TB Seagate Barracuda ST1000DM003 (included) |
4x 1TB Samsung HD103SJ (optional) |
5x 1TB Samsung HD103SJ (optional) |
4x 4TB Hitachi HDS5C4040ALE630 (included) |
RAID mode | RAID 5 |
RAID 5 |
RAID 5 |
RAID 5 |
RAID 5 |
Network connection | Gigabit Ethernet |
Gigabit Ethernet |
Gigabit Ethernet |
Gigabit Ethernet |
Gigabit Ethernet |
Firmware | 1.54 |
3.3.4.29856 |
3.6.0 Build 0220T |
2.03.08 |
01.05.08.31 |
Benchmarks |
---|
Iometer v2008.06.18-RC2 Intel NAS Performance Toolkit v1.7.1 Power Consumption |
Notes
All of our comparison NAS servers are tested using a single Gigabit Ethernet link with Jumbo frames disabled, and are connected to our Intel client machine and network via a Netgear GS108 switch.
Iometer, a familiar benchmark in our test suite, is an I/O subsystem measurement tool. For the purpose of our NAS testing, we set a queue depth of 16 (outstanding I/Os) and run the following three benchmarks:
- 64KB transfer, 100% sequential, 100% read
- 64KB transfer, 100% sequential, 100% write
- 1MB transfer, 100% sequential, 100% read
Our second benchmark - the NAS Performance Toolkit, developed by Intel - is a file-system exerciser specifically designed to provide performance comparisons between NAS devices. Intel's utility focuses on user-level performance using real-world workload traces gathered from typical digital home applications such as HD video playback and record. Intel NASPT reproduces the file-system traffic observed in these traces onto a chosen storage solution and records performance in MB/s.
Last but not least, we measure NAS power draw during three states; under load, idle and standby mode if applicable.