I wrote about my testing of notebook disk drive power usage on ZDnet yesterday (see How much does a flash disk increase battery life?). I pulled the 160 GB WD Scorpio out of my MacBook and ran it on wall power through a Kill-a-watt meter to better understand power usage. I learned – or relearned – a few things.
What surprised me most was the fact that as I measured power usage I saw that I/O, CPU and network usage were all intertwined. I’m surprised I was surprised since they are systems and the pieces work together.
I/O and CPU
I ran a defrag program that exercised the disk while driving CPU usage on a Core Duo to 90%. Since the CPU uses almost 3x the power of the disk it is the CPU and not the disk that is the power hog.
In fact, the biggest power hog is the base system: 13 watts sitting there doing nothing, LCD, Wi-Fi and Bluetooth all turned off, with no CPU load.
Hey, Taiwan, want to build long-battery life notebooks: figure out how to turn more pieces of the system off when not in use.
28 watts max
With the LCD turned on full, Wi-Fi and Bluetooth on, and the CPU at 90% the diskless notebook pulled 28 watts.
Turning on Wi-Fi was good for 2 watts, almost as much as a busy disk, partly due to the CPU load of running the TCP/IP stack.
Optical drives are power hogs too. Maybe not the drive itself, but the associated graphics or CPU processing.
Here are the numbers
The StorageMojo take
The significance of the intertwined nature of I/O, CPU and network usage is this. Flash drives sip power, but in a busy system it is all the other subsystems that chew up the battery.
The power saving advantages of a flash drive are best in a lightly loaded system with a long battery life, i.e. your cell phone, PDA or ultra-light notebook. In a 2-3 hour battery life 15-17 inch notebook the 2-3 watts a disk uses is almost noise level.
I see some evidence that the flash drive makers are adjusting their marketing to these facts. That is all to the good. Given the price/capacity differential you want flash disk customers buying for the right reasons.
Comments welcome, of course.
Interesting, but not really that surprising. When laptops are plugged in they go into “full power” mode where they keep the CPU speed higher for longer and turn off the screen and disk later (if at all).
A more interesting test would be to test on battery, and use runtime as your measurement. If possible, try to do something similar to a normal workload. Even a synthetic disk test (read and write all over the disk) comparing flash vs. magnetic would be useful.
Steve, the OS X system preferences allow you to choose the behavior of better performance or better battery life. If you go to the options you can control screen dimming times, disk standby behavior and such. They aren’t very refined but they did give the basis for 2 CPU measurements noted in the table.
Robin
its not about about power, its not about weight and its not about performance.
atm flash based drives does not seem to have that much advantage over normal disks. i think its more about getting rid of the moving parts and that is a good idea for a notebook. the disk in the ipod was one of the major reasons for me to not get one….
Robin, most of this is pretty well known. If you look online there are quite a few presentations on where all the power goes – the majority is actually into the display.
One problem that I’d point out is that for notebooks you are generally looking for very small amounts of power – a mW there, 4mW in another system, etc. etc.
The power meter you used is extremely unreliable (I know, I own one and tried to use it for metering server power consumption). To get a meter accurate to the 100th of a watt you’ll really need something else.
2-3W is definitely a pretty big win though.
DK
This effect is called Amdahl’s Law. But of course, component manufacturers can’t use balanced system performance as a marketing bullet point.
I suspect the Gigabit Ethernet on your system accounts for a non-negligible proportion of the 13W power draw, probably as muchg as WiFi. You might want to try with the GbE NIC disabled.
This is uneducated.. The TCP/IP stack takes 2W? Damn dude – good job on having a popular blog – but you should research before you post.
David, I’ve looked for the presentations you refer to. Send me the links!
Fazal, I didn’t check to see if the gigE was enabled on the notebook. I never use it, so I doubt it, but you may have a point.
Ben, just reporting what I saw. Are you saying TCP/IP is a lightweight protocol? Compared to what?
Robin
Well, TCP/IP is not light-weight, but that doesnt mean it will use “power” (ie. CPU) unless doing some real heavy lifting? Were you benchmarking the network when the wifi was activated or?