It has been demonstrated that shouting at a disk drive will cause it to stop. But what about the constant nagging they get in noisy data centers or busy towers? That may be a bigger problem.
A study published in 2005 Performance Impact of External Vibration on Consumer-grade and Enterprise-class Disk Drives by Thomas Ruwart and Yingping Lu found that
. . . that CS disk drives are more sensitive to the vibration from physically coupled adjacent disk drives. . . . [G]reater care needs to be taken in enclosure design, particularly for the 3.5-inch form factor disk drives due to their higher-energy seek operations when compared to seek operations on a 2.5-inch form factor disk drive.
The mechanical engineers who design system and drive enclosures know that vibration is a serious problem. The problem is that most civilians don’t understand the problem and are not willing to pay to solve it.
Hitachi GST, a major drive manufacturer, put it nicely in a pdf on their Rotational Vibration Safeguard feature:
One of the greatest hindrances to hard disk performance is vibration. Like a needle on a record, the disk drive’s head must try to follow narrow data tracks in order to read (or write) information.
Physical disturbances can throw the head off-track and cause a delay while the actuator repositions it. . . . In modern, balanced hard drives, linear back-and-forth vibration barely affects the operation of the drive head. However, circular movements — rotational vibration — can cause serious disruption.
. . . The drop in performance can be significant.
Current disk drives use accelerometers to compensate for shock and linear and rotational vibration. Accelerometer info isn’t available to us so it’s difficult to quantify the vibration’s impact. But data centers are noisy places quivering with the vibration of fans, air conditioners, disk drives and 60 cycle hum.
A limited study
In a paper presented at the USENIX SustainIT ‘10 conference Julian Turner reported on limited tests of a prototype anti-vibration rack. The AVR-1000 is made of engineered carbon fiber composite designed to dissipate vibration across a wide frequency range.
His observations included these:
Performance improvements for random reads ranged from 56% to 246% while improvements for random writes ranged from 34% to 88% for a defined set of industry benchmarks. Streaming sequential reads and writes had a much smaller performance improvement. . . .
Evidently the combination of random head movements and vibration has a substantial effect on disk read performance.
For the enterprise and Internet-scale data centers the implications are substantial:
- Energy savings. Anti-vibration racks have the potential to save significant power - not only by improving disk performance - but reducing run times.
- SSD value. Flash SSDs have fast random read access. But disks can improve their performance by 50% through vibration damping, that changes value proposition for SSDs.
- Array sizing. Enterprise arrays are over-configured to improve performance. If disks were suddenly 50% faster that could be reduced or, alternatively, the utilization could be increased.
The Storage Bits take
The research is limited, but everything we know about disks and vibration today suggests this is real. Assuming further research finds similar results we could see an explosion of products using engineered materials to improve disk performance.
This is already the norm in chip fabs, where nanometer precision requires extensive vibration damping. Disk feature sizes are even smaller, so why not?
For home users of multi-drive towers it may be that stiff carbon fiber disk mounts could improve performance. It may not be economic, but it sure would look cool.
Comments welcome, of course.
