Performance Comparison of Windows Server 2012 Essentials on the HP ProLiant N40L MicroServer and the HP ProLiant N54L G7 MicroServer

By: John Stutsman

One of the things I wanted to do with the HP ProLiant N54L G7 MicroServer was compare it’s performance to its predecessor HP ProLiant N40L MicroServer. In a previous writing I compared the rated performance of the CPU’s that showed the N54L scoring 43% higher than the N40L on benchmark tests accumulated through February 5, 2013.

Both machines were flashed with the latest BIOS-MOD based on the most current stock BIOS (#O41 July 29, 2011) and all settings were as identical as I could make them. Additionally, I updated the NIC drivers in both machines with the latest HP Broadcom 1Gb Driver (#15.4.0.19 February 19, 2013) for Microsoft Windows Server 2012 Essentials and Microsoft Home Server 2011.

Note: the NASPT Benchmarks discussed later are useful for Comparisons with My specific testing setup, network, and Client Workstation implementation and should not be considered absolute benchmarks for comparison against another person’s results.

Contents:

HP ProLiant N54L G7 MicroServer and HP ProLiant N40L MicroServer
Description of NAS Performance Toolkit (“NASPT”) Tests
Specifications of Client Workstation that NASPT will operate from
Specifications of Target: Windows Server 2012 Essentials on N40L
Specifications of Target: Windows Server 2012 Essentials on N54L
Specifications of Network between Client and Targets
Test Results
Conclusion
References

= = = = = = = =

HP ProLiant N54L G7 MicroServer and HP ProLiant N40L MicroServer

The HP ProLiant N54L G7 MicroServer (“N54L”) is the third variation in the MicroServer line at HP (N36L w/system board #620826-001, N40L w/system board #661787-001, and N54L w/system board #708503-001). In a previous writing I compared the rated performance of the CPU’s and having worked with the N40L in the past I was anxious to try out the N54L. This is the same MicroServer I introduced in:

HP ProLiant N54L G7 MicroServer – First Look

The HP ProLiant N40L MicroServer (“N40L”) is the same machine that was exhibited at the 2012 Home Server Show Meet-Up later used in test set-ups described in the articles:

Description of NAS Performance Toolkit (“NASPT”) Tests

The NAS Performance Toolkit (“NASPT”) was developed and described by Tony Bock, Mason Cabot, Frank Hady, and Matthew Shopsin of the Storage Technologies Group, Intel Corporation, in the paper Measuring and Improving Single-User NAS Performance.

Portions of their abstract states:

NAS devices are increasingly entering the home and small business as centralized storage resources for large collections of documents, pictures, music and videos. Increasingly these devices are used for more than background tasks like backup. Newer interactive usages, like media access/creation, expose the performance of the NAS directly to the user. Unlike the enterprise NAS, the home and small business NAS will be judged primarily by single user performance as seen in user wait time.

We introduce a new tool, the NAS Performance Toolkit (NASPT), uniquely built to measure the single user NAS Performance seen by a user of a mainstream personal computer. NASPT includes a wide range of workloads identified by our analysis of media, productivity and bulk data operations likely to drive single user NAS performance.

We’ve made NASPT very easy to use and freely available. ….

A number of tests are provided with the NASPT. Below is a listing of those tests used in this report:

HD Video Playback: Traced from a commonly available video playback application, this trace represents about ten minutes of 720p high definition MPEG-2 video playback. A single 1.3GB file is accessed sequentially with 256kB user level reads. As is true in many of the workloads the NAS itself sees smaller reads since the SMB client and file system break these 256kB requests into smaller requests.

HD Video Record: This trace represents recording roughly fifteen minutes of a broadcast 720p MPEG-2. A single 1.6GB file is written sequentially with 256kB access. The bit rate is somewhat lower than the playback test, they contain different video.

HD Video Play & Record: This test was algorithmically constructed from the above video playback and record traces. To combine we introduced a 50ms offset into the record stream then merged the two streams. The 1GB file represents four minutes twenty seconds of application run time. Because the two streams have differing bit rates and because of variation in original trace periodicity, there is not a strict alternation of accesses. About 20% of the transactions are sequential.

Two HD Video Playback Streams: Constructed from two copies of the above HD Video Playback test, this trace transfers 1.4GB of data representing two video streams played back for about six minutes. Again, sometimes one stream will issue two transactions in rapid succession so about 18% of the transactions are sequential.

Four HD Video Playback Streams: This workload is constructed from four copies of the video playback test. The 1.3GB trace represents about three minutes forty-five seconds of video playback for each stream. About 11% of the accesses are sequential.

Content Creation: This is a trace of commercially available video and photo editing software products executing a scripted set of operations to produce a video from a collection of different source materials. It contains a single very large file, apparently containing the video output, which is written in bits and pieces. About 11% of accesses within this file are sequential. There are many smaller files that are read and written more or less sequentially. Overall, about 40% of the accesses are issued sequentially. The test transfers 155MBs, 90% of transactions are writes. The median read size is 1300 bytes. The median write is 12kB. Transfers include a wide range of different sized accesses.

Office Productivity: Scripted sequences of typical workday operations from a commonly available office productivity suite make up this trace. This test is the largest of the collection, transferring 2.8GB of data evenly divided between reads and writes. Eighty percent of these accesses are logically sequential, scattered across six hundred files ranging from 12 bytes in length to over 200MB. The median read size is 2.2kB whereas the median write size is 1.8kB.

File Copy To NAS: This trace includes accesses executed when copying a 1.4GB file to a NAS. Data is written in 64kB sequential transactions.

File Copy From NAS: Identical to File Copy To NAS, but in the opposite direction. All transactions are sequential 64kB reads.

Directory Copy To NAS: This trace represents a bulk copy of a complex directory tree containing 2833 files, a transfer a large collection of files to the NAS. The directory used represented a typical installation of a commercially available office productivity suite. 247MBs is transferred with an average write size of 41.4kB. Only 52% of the writes are logically sequential as many files are small.

Directory Copy From NAS: Identical to File Copy To NAS, but in the opposite direction creating many read accesses.

The N54L and N40L are nearly identical machines and hardware except for the differences in the CPUs in the machines so I was not expecting any difference in the performance of the machines, if I was able to set them up identically, except for those processes that relied heavily on CPU performance to complete.

So, in addition to the tests above I also timed the loading of several applications in Windows Server 2012 Essentials to gauge the impact of going from the N40L to the N54L. These applications were:

Load DashBoard
Load Allway Sync
Load Server Manager
Load Computer Management

I expected these tests to provide a metric on the “snappiness” or “feel” when moving from the N40L to the N54L.

Specifications of Client Workstation that NASPT will operate from

The Client Workstation that NASRT v1.7.1 will be operated from for these tests is specified as follows:

Hyper-V Client: Windows 7 Professional SP1, x86,
2 CPU Cores have been allocated to the Client
2GB RAM (locked to 2GB with dynamic memory turned off).
Client VM and VHDX is the only client running on VHD Drive: Crucial M4 256GB SSD attached to a GSATA III Port on the Host platform GA-Z77X-UD5H running Windows Server 2012 functioning as a Hyper-V Server.
This is essentially the same Z77 Hyper-V Server exhibited at the 2012 HSS Meet-up with additional VHD Drives including the Crucial M4 256GB SSD listed above.
All VM’s on the Host have been shut down to provide the Client for this test exclusive access to the Virtual NIC bound with the Intel 82579V Gigabit NIC on the Z77 Motherboard.
The directory path to Windows Server 2012 Essentials (\\Ranger\Shares\NASPTxPerfTest\NASPT) has been bound to the letter R: with Read/Write access granted

Figure 1 — ATTO Performance of the Client Workstation

Figure 2 — CrystalDiskMark of the Client Workstation

Specifications of Target: Windows Server 2012 Essentials on N40L

The Target: Windows Server 2012 Essentials – on the N40L was specified as follows:

Windows Server 2012 Essentials
HP ProLiant N40L MicroServer
BIOS-MOD Flashed to unlock BIOS and enable AHCI and SATA II Maximum Speed for on-board SATA Port and eSATA Port
16GB G.Skill F3-1333C9D-16GAO RAM
OS Drive: Crucial M4 128GB SSD SATA III
HighPoint Rocket 640L PCIe Card providing SATA III Port for OS Drive
Data Drives: 5 x ST3000DM001
Western Digital SuperSpeed PCIe USB 3.0 Card
Icy Dock DuoSwap MB971SP-B
Other Software Running: MicroSoft Storage Spaces, AllWay Sync, MicroSoft ForeFront Client, Server Backup
DropBox and SkyDrive also loaded
Data is on a drive created in Storage Spaces that is: 2-Way Mirror, thin provisioning, NTFS
The Target Folder NASPTxPerfTest was created on the Data Drive
NIC updated with HP Broadcom 1Gb Driver for Windows Server 2008 x64 Editions version 15.4.0.19 (19 Feb 2013) cp017879.exe
Backup Drive for Server Backup: STBV3000100

The performance of the OS Drive C and the Storage Space Drive D on the S2012E on the N54L using ATTO and CrystalDiskMark is as follows:

Figure 3 — OS Drive C on S2012E on N40L

Figure 4 — OS Drive C on S2012E on N40L

Figure 5 — Storage Spaces 2-Way Mirror Drive D on S2012E on N40L

Figure 6 — Storage Spaces 2-Way Mirror Drive D on S2012E on N40L

The performance of the connection from the Client WorkStation to the Target Directory Mapped on S2012E on the N40L using ATTO and CrystalDiskMark is as follows:

Figure 7 — ATTO from Client Workstation to Target Mapped on S2012E on the N40L

Figure 8 — CrystalDiskMark from Client WorkStation to Target Directory Mapped on S2012E on the N40L

Specifications of Target: Windows Server 2012 Essentials on N54L

The Target: Windows Server 2012 Essentials – on the N54L was specified as follows:

Exactly the same as Windows Server 2012 Essentials – N40L: everything was moved to the N54L and all settings in BIOS were the same to the best of my knowledge

The performance of the OS Drive C and the Storage Space Drive D on the S2012E on the N54L using ATTO and CrystalDiskMark is as follows:

Figure 9 — OS Drive C on S2012E on N54L

Figure 10 — OS Drive C on S2012E on N54L

Figure 11 — Storage Spaces 2-Way Mirror Drive D on S2012E on N54L

Figure 12 — Storage Spaces 2-Way Mirror Drive D on S2012E on N54L

The performance of the connection from the Client WorkStation to the Target Directory Mapped on S2012E on the N54L using ATTO and CrystalDiskMark is as follows:

Figure 13 — ATTO from Client Workstation to Target Mapped on S2012E on the N54L

Figure 14 — CrystalDiskMark from Client WorkStation to Target Directory Mapped on S2012E on the N54L

Specifications of Network between Client and Targets

The Network between the Client and Targets was:

The Client and all Targets are attached to a single D-Link 24 Port Gigabit Switch DGS-1024D via CAT6 Cable.

Figure 15 — Network for NAS Performance Tests

Figure 16 — LAN Speed Test between Client WorkStation and S2012E on N40L

Figure 17 — LAN Speed Test between Client WorkStation and S2012E on N54L

Test Results

I should preface this section by noting that I am not a NAS performance expert but I did notice that the N54L was “snappier” than the N40L. I tried to quantify what “snappier” was by measuring with a stop watch the start-up of several applications in S2012E. Figure 18 shows that I measured a 40% or 16 seconds improvement in the loading time of the DashBoard in S2012E when moving from the N40L to the N54L. This corresponds favorably with the 43% performance improvement in Table 4 of HP ProLiant MicroServer CPU Performance Index Comparisons using PassMark Index Scores.

Figure 18 — Performance Improvement as measured with a Stop Watch

At Chris Kenney’s suggestion, I looked at Intel’s NAS Performance Tests as another way to quantify the difference between the N40L and the N54L.

Figure 19 — NASPT Performance of S2012E on the N40L and N54L Normalized to the N40L

As Figures 19 and 20 illustrate the Intel NASPT of the N54L is only slightly better overall than the N40L’s. I believe this could be expected, as shown later in Figure 21, a number of the NASPT results (HD Playback’s) are approaching the theoretical limit of transfer rates over a Gigabit Network. So, there just wouldn’t be a lot to gain in that respect for transfer rates when moving from the N40L to the N54L.

Figure 20 — NASPT Performance of S2012E on the N40L and N54L Normalized to the N40L

Conclusion

On a Gigabit Ethernet the maximum theoretical transfer is 125MB/s. Many of the average scores (See Figure 21 & 22 especially for HD Playback’s) for the N40L and N54L with S2012E are approaching that theoretical transfer limit. While the N54L provides a slight performance boost with NASPT scores the real benefit of the N54L is the improved responsiveness (See Figure 18) of applications in S2012E.

Figure 21 — The Throughput of Many Media Tests is approaching the Maximum Theoretical of 125MB/s for a Gigabit Ethernet – Note: these Benchmarks are useful for Comparisons with My specific testing setup, network, and Client Workstation implementation and should not be considered absolute benchmarks for comparison against another person’s results

Figure 22 — NASPT Throughput Comparison of S2012E on a N40L and a N54L – Note: these Benchmarks are useful for Comparisons with My specific testing setup, network, and Client Workstation implementation and should not be considered absolute benchmarks for comparison against another person’s results

References

Latest Images

Trending Articles

Latest Images