SSD Benchmark Tests: SATA IDE vs AHCI Mode

In a recent Benchmark Reviews technical article, SSD performance was tested in AHCI and IDE mode using only the HD-Tune software to produce results. It wasn't intentional, but our test results were flawed by this single-threaded benchmark tool, and our conclusion did not properly illustrate IOPS performance. In this article, Benchmark Reviews tests the latest JMicron, Indilinx, and SandForce SSDs using a combination of tools to illustrate the true difference between SATA IDE and AHCI-mode, and demonstrates how one SATA mode is better suited than the other.

Solid State Drive technology is unfamiliar to many consumers, and so long as there are different ways an SSD can operate there always be questions that need answering. Benchmark Reviews offers various SSD tests, but proving performance speeds and matching manufacturer claims is only part of the story. Each SSD processor has unique behaviors, with some working well with TRIM and offering improved performance in AHCI mode, while others include Garbage Collection (GC) and work best in IDE mode. In this article Benchmark Reviews demonstrates how SSDs are tested by the manufacturer, and illustrates how real-world performance is different for end-users.

When Benchmark Reviews first started testing SSD storage products two years ago, we discovered that there were just too many variables to alter benchmark results. For example, the exact same Solid State Drive may offer one specific write speed while connected to the Intel ICH10 controller, yet could operate up to 30% slower on Marvell or JMicron SATA controller. Additionally, immature driver software can further degrade performance, or optimized drivers can restore speeds. Complicating matters even more, SSD-specific firmware can add features and improvements, but may also reduce operational speeds.

Readers familiar with the new technology have learned to read reviews from as many sources as possible; which is exactly what we've been suggesting in our product conclusions for almost two years. Some websites don't use special drive conditioning tools such as DISKPART or Sanitary Erase, and other do. The same is true for SSD owner, which span from hardware enthusiasts to basic computer users. Benchmark Reviews has concentrated on including several different tools for realizing quantitative performance results for our articles, while other websites use real-world file transfers and application routines. It's all relative, subjective, and impossible to determine which method is best.

For this article, our tests will focus on three of the most widely used SSD processors from Indilinx, JMicron, and SandForce. Using the Intel ICH10 SATA-3.0 controller, our SSD tests will benchmark read and write performance speeds with pristine 'fresh' NAND conditions with Sanitary Erase (for Indilinx) and DISKPART with the 'clean all' command for drive conditioning. Operational IOPS performance will be tested in much the same way, and will reveal which SATA controller is best suited for SSDs.

SSD Benchmark Differences

Solid State Drive technology comes in a myriad of shapes and sizes, which is to say that no SSD is the same as the next. This is why Benchmark Reviews has used three different SSD processor for this article, and did not focus on only one. These differences extend to the strengths and weaknesses of each SSD processor, which might translate into better IOPS performance or bandwidth speed, but could also include new features such at native encryption or TRIM support. Just as every SSD is unique, so is every benchmark testing tool.

As Benchmark Reviews would later discover, our singular use of the HD-Tune software to measure operational IOPS performance was a mistake. HD-Tune 4.01 works fine for measuring sequential read and write bandwidth speeds, but it's a single-threaded application that doesn't issue parallel requests in random IOPS tests. Our previous results relied on this tool to help shape our conclusion of AHCI vs IDE performance, but after some additional testing we determined a second and third opinion were necessary.

For this article, Benchmark Reviews has included results from CrystalDiskMark 3.0.0b and AS-SSD Benchmark 1.4.3704 to help compliment benchmark speed tests from HD-Tune. These tests should offer a more concise report, and assist PC hardware enthusiasts on choosing the SSD and SATA controller mode that is best for their needs.

AHCI & IDE SATA-Mode Differences

Most desktop computer motherboards offers hardware enthusiasts the option to adjust the SATA controller interface settings in their BIOS between AHCI and IDE mode. Traditionally, IDE is the time-tested standard for desktop hard drive products. The IDE interface offers basic control over the storage device, but unfortunately it does not include enterprise-level features. Advanced Host Controller Interface (AHCI) on the other hand, was created as a means to deliver additional storage support for server backplanes and high-end performance desktop computers.

Between these two disk IO settings, there are a few specific features that differ. AHCI-mode includes hot-swap drive support and Native Command Queuing that is not available in IDE mode. IDE does not support RAID arrays, and does not include the extended power management features that AHCI offers. For OpenBSD servers, port multiplier support is also enabled through AHCI.

Bandwidth Speed vs Operational Performance

The argument between AHCI and IDE SATA-modes revolves around two other dynamics: bandwidth speed (MB/s) and operational performance (IOPS). The two metrics work together, but one is more important than the other. Consider this analogy: bandwidth determines how much cargo a ship can transport in one voyage, and operational IOPS performance is how fast the ship moves. By understanding this and applying it to SSD storage, there is a clear importance set on each variable depending on the task at hand.

For casual users, especially those with laptop or desktop computers that have been upgraded to use an SSD, the naturally quick response time is enough to automatically improve the user experience. Bandwidth speed is important, but only to the extent that operational performance meets the minimum needs of the system. If an SSD has a very high bandwidth speed but a low operational performance, it will take longer to load applications and boot the computer into Windows than if the SSD offered a higher IOPS performance.

SSD Test Methodology

Solid State Drive storage products arrive with unused 'clean' NAND flash, and as data is written these modules become used and 'dirty'. For the purpose of this article, Benchmark Reviews used the DISKPART command line program with the 'clean all' instruction to condition each storage device prior to testing. In the case of products using the Indilinx Barefoot controller, the SSD was further prepared with the Sanitary Erase application.

As a word of caution, applications such as Sanitary Erase (SE) and Wiper offer immediate but temporary restoration of original 'pristine' performance levels. These performance levels are not representative to real-world performance. SSDs attached to TRIM-enabled Operating System's will benefit from continuously restored like-new NAND performance, whereas older O/S's will require a garbage collection (GC) tool to avoid 'dirty NAND' performance degradation.

It's critically important to understand that no software can accurately measure SSD performance in relation to the real-world experience. Synthetic benchmark tools are helpful indicators, but should not be considered the ultimate determining factor for any SSD purchase. Benchmark Reviews includes both bandwidth speed and operational IOPS tests to present a conclusive measurement of product performance under 'pristine' NAND conditions.

Test System

• Motherboard: Gigabyte GA-X58A-UD7 with version F6 BIOS
• Processor: Intel Core i7-920 BX80601920 @ 2.667 GHz
• System Memory: 6GB Tri-Channel DDR3 1600MHz CL6-6-6-18
• SATA 3Gb/s Storage HBA: Integrated Intel ICH10R SATA-3.0 Gbps Controller
• Operating System: Microsoft Windows 7 Ultimate Edition 64-Bit

Drive Hardware Tested

The following storage hardware has been used in our benchmark performance testing, and may be included in portions of this article:

• ADATA S599 100GB SATA MLC SSD
• Corsair Nova V128 CMSSSD-V128GB2 SATA MLC SSD
• WD SiliconEdge-Blue SSC-D0256SC-2100 256GB SATA SSD

Test Tools

• AS-SSD Benchmark 1.4.3704: Tests SSD file transfer and operational performance
• CrystalDiskMark 3.0.0b: Records sequential read and write speeds at various loads
• HD Tune Pro 4.01 by EFD Software: Measures sequential file transfer bandwidth speed

Test Results Disclaimer

This article utilizes benchmark software tools to produce operational IOPS performance and bandwidth speed results. Each test was conducted in a specific fashion, and repeated for all products. These test results are not comparable to any other benchmark application, neither on this website or another, regardless of similar IOPS or MB/s terminology in the scores. The test results in this project are only intended to be compared to the other test results conducted in identical fashion for this article.

Indilinx: AS-SSD Performance

Using an NTFS formatted Corsair Nova V128 Solid State Drive, sequential bandwidth speed and operational performance tests were conducted on the Indilinx 'Eco' Barefoot SSD controller with AS-SSD Benchmark 1.4.3704:

Indilinx "Eco" Barefoot SSD Performance on ICH10 AHCI

Indilinx "Eco" Barefoot SSD Performance on ICH10 IDE

Indilinx AS-SSD Benchmark Summary

Using AS-SSD Benchmark the reported sequential read speeds favored AHCI, while sequential write performance was similar between SATA modes. Reported AHCI sequential read/write speeds: 236/187 MB/s. Reported IDE speeds: 220/188 MB/s.

4K single thread write tests were very similar, but 4K 64-thread tests undoubtedly favored the Indilinx SSD controller in AHCI mode.

SATA Mode Recommendation: AHCI

Indilinx: CrystalDiskMark

Using an NTFS formatted Corsair Nova V128 Solid State Drive, sequential bandwidth speed and operational performance tests were conducted on the Indilinx 'Eco' Barefoot SSD controller:

Indilinx "Eco" Barefoot SSD Performance on ICH10 AHCI

Indilinx "Eco" Barefoot SSD Performance on ICH10 IDE

Indilinx CrystalDiskMark Summary

CrystalDiskMark 3.0.0b reported sequential read speeds in favor of AHCI mode, however sequential write performance was virtually identical between the two SATA modes. Reported AHCI sequential read/write speeds: 250/174 MB/s. Reported IDE speeds: 232/174 MB/s.

512K tests indicate a marginal advantage for AHCI over IDE performance, and 4K performance follows the same trend.

4K tests with a queue depth of 32 produced similar write performance results, but the 4K QD32 read tests undoubtedly favored the Indilinx SSD controller in AHCI mode and reported nearly twice the performance.

SATA Mode Recommendation: AHCI

Indilinx: HD-Tune Speed

Although HD-Tune is a single-threaded application that cannot offer deep queue depth, it does offer a basic interpretation of linear bandwidth speed. Despite the limitations, we've included this test to provide an additional perspective into our research.

Using a clean and conditioned Corsair Nova V128 Solid State Drive, bandwidth speed tests were conducted on the Indilinx 'Eco' Barefoot SSD controller using HD-Tune 4.01:

Indilinx "Eco" Barefoot SSD Write Speed on ICH10 AHCI

Indilinx "Eco" Barefoot SSD Read Speed on ICH10 IDE

Indilinx Read Speed Summary

Average AHCI read speed: 192.6 MB/s. Average IDE read speed: 194.2 MB/s. Favored SATA controller interface for Indilinx SSD read speeds: Intel ICH10 IDE-Mode with a 1.6 MB/s advantage.

Indilinx "Eco" Barefoot SSD Write Speed on ICH10 AHCI

Indilinx "Eco" Barefoot SSD Write Speed on ICH10 IDE

Indilinx Write Speed Summary

Average AHCI write speed: 164.5 MB/s. Average IDE write speed: 192.3 MB/s. Favored SATA controller interface for Indilinx SSD write speeds: Intel ICH10 IDE-Mode with a 27.8 MB/s advantage.

Indilinx HD-Tune Summary

Although speed is not the most critical measurement of SSDs, HD-Tune reports basic single-threaded performance in favor of IDE mode over AHCI on the Indilinx controller. Read speeds were very similar but leaned towards IDE mode, while write performance appeared to wane while using AHCI mode and remained strong in IDE mode. If not for the impact of AHCI mode write operations on the Indilinx controller the differences appear minimal, so bandwidth speeds could be comparable SSD-wide.

JMicron: AS-SSD Performance

Using the WD SiliconEdge-Blue Solid State Drive, sequential bandwidth speed and operational performance tests were conducted on the JMicron JMF612 SSD controller with AS-SSD Benchmark 1.4.3704:

JMicron JMF612 SSD Performance on ICH10 AHCI

JMicron JMF612 SSD Performance on ICH10 IDE

JMicron AS-SSD Benchmark Summary

Using AS-SSD Benchmark on the JMicron controller, the reported sequential speeds both heavily favored AHCI mode. Reported AHCI sequential read/write speeds: 238/180 MB/s. Reported IDE speeds: 218/155 MB/s.

4K single thread read tests were very similar between AHCI and IDE, however write tests marginally favored IDE mode. 4K 64-thread testing revealed that AHCI mode offered the best read performance, while the write performance leaned in the opposite direction toward IDE.

The JMicron JMF612 SSD controller offered mixed results, but the AS-SSD Benchmark tests generally conclude that AHCI mode prevails.

SATA Mode Recommendation: AHCI

JMicron: CrystalDiskMark

Using the WD SiliconEdge-Blue Solid State Drive, sequential bandwidth speed and operational performance tests were conducted on the JMicron JMF612 SSD controller:

JMicron JMF612 SSD Performance on ICH10 AHCI

JMicron JMF612 SSD Performance on ICH10 IDE

JMicron CrystalDiskMark Summary

CrystalDiskMark 3.0.0b reported sequential read speeds in favor of AHCI mode, however sequential write performance was very similar between the two SATA modes on the JMicron JMF612 SSD controller. Reported AHCI sequential read/write speeds: 241/188 MB/s. Reported IDE speeds: 228/187 MB/s.

512K read tests indicate a marginal advantage for AHCI over IDE performance, while 512K write performance was similar but managed to lean towards IDE mode. 4K read performance was comparable between SATA modes, while write performance slightly favored IDE mode.

4K tests with a queue depth of 32 produced nearly identical write performance results, but the 4K QD32 read tests favored the JMicron JMF612 SSD controller in AHCI mode.

SATA Mode Recommendation: AHCI

JMicron: HD-Tune Speed

Although HD-Tune is a single-threaded application that cannot offer deep queue depth, it does offer a basic interpretation of linear bandwidth speed. Despite the limitations, we've included this test to provide an additional perspective into our research.

Using the WD SiliconEdge-Blue Solid State Drive, sequential bandwidth speed and operational performance tests were conducted on the JMicron JMF612 SSD controller with HD-Tune 4.01:

JMicron JMF612 SSD Read Speed on ICH10 AHCI

JMicron JMF612 SSD Read Speed on ICH10 IDE

JMicron Read Speed Summary

Average AHCI read speed: 171.1 MB/s. Average IDE read speed: 189.5 MB/s. Favored SATA controller interface for JMicron SSD read speeds: Intel ICH10 IDE-Mode with a 18.4 MB/s advantage.

JMicron JMF612 SSD Write Speed on ICH10 AHCI

JMicron JMF612 SSD Write Speed on ICH10 IDE

JMicron Write Speed Summary

Average AHCI write speed: 90.2 MB/s. Average IDE write speed: 92.8 MB/s. Favored SATA controller interface for Indilinx SSD write speeds: Intel ICH10 IDE-Mode with a 2.6 MB/s advantage.

JMicron HD-Tune Summary

HD-Tune bandwidth speed benchmarks report basic single-threaded performance in favor of IDE-mode over AHCI on the JMicron controller. Read speeds were much faster when tested in IDE-mode, while write performance was equally poor on the JMicron controller in either mode.

SandForce: AS-SSD Performance

Using the ADATA S599 Solid State Drive, sequential bandwidth speed and operational performance tests were conducted on the JMicron JMF612 SSD controller with AS-SSD Benchmark 1.4.3704:

SandForce SF-1200 SSD Performance on ICH10 AHCI

SandForce SF-1200 SSD Performance on ICH10 IDE

SandForce AS-SSD Benchmark Summary

Using AS-SSD Benchmark on the SandForce SF-1200 controller, the reported sequential speeds both heavily favored AHCI mode. Reported AHCI sequential read/write speeds: 206/94 MB/s. Reported IDE speeds: 157/68 MB/s.

4K single thread read and write tests were very similar, although they did lean slightly towards IDE mode. The 4K 64-thread tests undoubtedly favored AHCI mode on the SandForce SF-1200 SSD controller, reported AHCI read performance at 123 MB/s compared to only 22 MB/s for IDE.

SATA Mode Recommendation: AHCI

SandForce: CrystalDiskMark

Using an NTFS formatted ADATA S599 Solid State Drive, bandwidth speed tests were conducted on the SandForce SF-1200 SSD controller:

SandForce SF-1200 SSD Performance on ICH10 AHCI

SandForce SF-1200 SSD Performance on ICH10 IDE

SandForce CrystalDiskMark Summary

CrystalDiskMark 3.0.0b reported sequential read and write speeds that heavily favored AHCI over IDE mode on the SandForce SF-1200 controller. Reported AHCI sequential read/write speeds: 211/90 MB/s. Reported IDE speeds: 167/70 MB/s.

512K read tests also indicate a substantial advantage for AHCI over IDE performance in both the 512K read and write tests. 4K performance was similar between SATA modes, although the results slightly favored IDE mode.

4K tests with a queue depth of 32 produced made a significant difference on the SandForce controller. 4K QD32 read tests favored AHCI mode with 115/82 over IDE mode with 20/66.

SATA Mode Recommendation: AHCI

SandForce: HD-Tune Speed

Although HD-Tune is a single-threaded application that cannot offer deep queue depth, it does offer a basic interpretation of linear bandwidth speed. Despite the limitations, we've included this test to provide an additional perspective into our research.

Using a clean and conditioned ADATA S599 Solid State Drive, bandwidth speed tests were conducted on the SandForce SF-1200 SSD controller using HD-Tune 4.01:

SandForce SF-1200 SSD Read Speed on ICH10 AHCI

SandForce SF-1200 SSD Read Speed on ICH10 IDE

SandForce Read Speed Summary

Average AHCI read speed: 183.6 MB/s. Average IDE read speed: 168.8 MB/s. Favored SATA controller interface for SandForce SSD read speeds: Intel ICH10 AHCI-Mode. In one of the very few time AHCI performance exceeds IDE-mode, the SandForce SF-1200 SSD processor offered better read speed and smoother transactions.

SandForce SF-1200 SSD Write Speed on ICH10 AHCI

SandForce SF-1200 SSD Write Speed on ICH10 IDE

SandForce Write Speed Summary

Average AHCI read speed: 192.2 MB/s. Average IDE read speed: 195.1 MB/s. Favored SATA controller interface for SandForce SSD write speeds: Intel ICH10 AHCI-Mode with a 1.1 MB/s advantage.

SandForce HD-Tune Summary

Unlike the Indilinx and JMicron controllers the SandForce SF-1200 prefers AHCI mode over IDE, even if it's only to a slight advantage. Read speeds favored AHCI mode by nearly 15 MB/s, while write performance delivered a 1.1 MB/s benefit to AHCI over IDE mode.

AHCI vs IDE Final Thoughts

Even before the benchmarks make their report, there's a lot of weight behind using AHCI over the less fortunate IDE mode. While the core readership of Benchmark Reviews consists of PC enthusiast users who might own only a single SSD, there are many other users who read our articles that have two or more SSDs combined into RAID arrays on server computer systems. This is why the benchmark performance results may not necessarily impact the final decision: purpose supersedes performance. If your application requires hot-plug drive support or redundant disks, then AHCI is the only choice.

AHCI feature-set not withstanding, the only argument IDE-mode offers in its defense is the occasional benefit of slightly faster read and write speeds. Remembering my analogy of the cargo ship and the speed it moves from the article's introduction, there's a lot more weight to be given towards operational IOPS performance over file transfer bandwidth speeds.

As Solid State Drive storage devices mature and adoption grows, hardware enthusiasts can hope to see less separation between these two SATA interface technologies. Driver software continually improves compatibility, while SSD processor firmware steadily adds functionality and device stability. As we've seen recently with TRIM support in Microsoft Windows 7, the Operating System can also offer valuable new features that interact nicely with SSDs.

Keeping this in mind, the results in this article may become obsolete as weaknesses in the design are corrected. For the foreseeable future of 2010, this information can aid enthusiasts in receiving the fastest speeds and highest operational performance possible from their expensive SSD storage devices.

SATA Controller Mode Conclusion

First and foremost, please accept my apology for previously publishing misleading results. The benchmark tests were all conducted properly, but the test tool I used did not accurately illustrate the difference between SATA controller mode performance. After conducting several new tests using additional benchmark tools, my opinion on the matter has changed.

Based on the results obtained in these tests, here is how the various SSD controllers function in connection with the Intel ICH10 SATA-3.0 controller:

Indilinx "Eco" Barefoot SSD:

The Indilinx SSD processor has found its way into more SSDs than any other, thanks to respectable performance at an affordable price. File transfer read speeds were very similar between SATA controller modes in HD-Tune, while operational IOPS performance in AS-SSD and CrystalDiskMark was much faster while using the Indilinx SSD in AHCI-mode on the Intel ICH10 SATA-3.0 controller. AHCI mode added some overhead to our HD-Tune write results, but AS-SSD and CrystalDiskMark write results appeared very similar in performance. Suggested SATA control mode: AHCI.

JMicron JMF612 SSD:

The JMicron JMF612 SSD processor delivered mixed results in most of our tests, but still showed a sincere preference for AHCI-mode in our sequential read and write bandwidth speed tests with AS-SSD and CrystalDiskMark; although HD-Tune's single-threaded read tests leaned towards IDE mode. Operational IOPS performance was mixed between AHCI and IDE mode when tested with AS-SSD Benchmark and CrystalDiskMark, yet more often than not the results favored the AHCI control mode. Suggested SATA control mode: AHCI.

SandForce SF-1200 SSD:

The SandForce SF-1200 is expected to replace the Indilinx SSD processor for performance and price throughout 2010, making it an important product to concentrate on. In all of our benchmark tests, the SandForce SF-1200 SSD delivered substantially better performance in AHCI mode. File transfer speeds were much faster in AHCI mode for read and write bandwidth regardless of the test tool. Operational IOPS performance also clearly favored AHCI mode, delivering 4K QD32 reads of 115/82 MB/s over IDE mode with 20/66. Suggested SATA control mode: AHCI.

But what about SSD TRIM?

TRIM is non-specific to either SATA controller mode, and will work equally well in both IDE and AHCI-mode. Benchmark Reviews will soon publish a detailed report that illustrates the recovery levels for each SSD controller tested here.

Epilogue

In an unfortunate circumstance, our audience was at odds with the previously published results. Please remember to offer constructive feedback in our comments, so that it can be taken with merit.

Benchmark Reviews encourages you to leave comments (below), or ask questions and join the discussion in our Forum.

Related Articles:

• Android USB Tethering Shared Internet Access
• MSI N460GTX HAWK GeForce GTX 460
• Palit GeForce 9600 GT 1GB Sonic NE/960TSX0202
• ASUS Radeon EAH5870 V2 Video Card
• Palit Radeon HD 4870 Sonic Dual Edition
• NVIDIA GeForce GTX 770 Video Card
• OCZ RevoDrive X2 PCI-Express SSD
• Hanns-G HZ281HPB WideScreen LCD Monitor
• OCZ Summit MLC SSD OCZSSD2-1SUM120G
• Corsair Force F100 SandForce SSD CSSD-F100GB2