[kwlug-disc] USB3 expectations? (was: Re: USB3 on Linux)

unsolicited unsolicited at swiz.ca
Fri Jun 18 14:31:21 EDT 2010

Khalid Baheyeldin wrote, On 06/18/2010 11:54 AM:
> Something I would be interested in is transfer speed from a hard drive. Is
> it really that different on USB3 from USB2, or will it be the same?

Help me understand the question.

If USB3 is faster than USB2, why would it be the same? Hw/sw not 
catching up to the specs yet?

Or are you expecting the performance increase not worth the price?

Until this thread, I don't think I even knew there was such as thing 
as USB 3. So I poked wikipedia for some info. Thieving from wikipedia 
- http://en.wikipedia.org/wiki/Usb3#USB_3.0:


A new major feature is the "SuperSpeed" bus, which provides a fourth 
transfer mode at 5.0 Gbit/s. The raw throughput is 4 Gbit/s, and the 
specification considers it reasonable to achieve 3.2 Gbit/s (0.4 
GByte/s or 400 MByte/s), or more, after protocol overhead.[60]

When operating in SuperSpeed mode, full-duplex signaling occurs over 
two differential pairs separate from the non-SuperSpeed differential 
pair. This results in USB 3.0 cables containing two wires for power 
and ground, two wires for non-SuperSpeed data, and four wires for 
SuperSpeed data, and a shield that was not required in previous 
specifications." (It goes on to talk about a new, but backwards 
compatible, plug. Interesting.)

Hmm. OK, so USB2 480 Mbit/s, USB3 3.2 Gbit/s effective (if we ever hit 
even that. And let's not forget about the whole sharing devices / the 
bus, mixing USB1/2/3 issues. So, we can expect hardware to come with 
even more individual USB ports, and a plethora of add in cards, to 
come?) And USB3 ~= SATA2.

Curious, I then flipped over to SATA, 
http://en.wikipedia.org/wiki/SATA2, "While even the fastest 
conventional hard disk drives can barely saturate the original SATA 
1.5 Gbit/s bandwidth," - huh. Interesting. It continues "Solid-State 
Drives have already saturated the SATA 3 Gbit/s limit at 250 MB/s net 
read speed. Ten channels of fast flash can reach well over 500 MB/s 
with new ONFI drives, so a move from SATA 3 Gbit/s to SATA 6 Gbit/s 
would benefit the flash read speeds. As for the standard hard disks, 
the reads from their built-in DRAM cache will end up faster across the 
new interface.[12] SATA 6 Gbit/s hard drives and Motherboards are now 
shipping from several suppliers."

So, now is there not only a USB3, there's a SATA3 (6Gbits/s, 600MB/s 
effective)! And most drives can't keep up to SATA 1, begging the 
question ...

Now it gets a little convoluted "communicate at a rate of 3 Gbit/s. 
Taking 8b/10b encoding overhead into account, they have an actual 
uncoded transfer rate of 2.4 Gbit/s (3000000000*8/10/1024/1024/8 ? 
286.10 MiB/s). The theoretical burst throughput of SATA 3.0 Gbit/s is 
roughly double that of PATA/133. Also, SATA devices offer enhancements 
such as NCQ that improve performance in a multitasking environment." 
(i.e. SATA is gradually getting some of the advantages of the SCSI 
spec.) And later "However, high-performance flash drives are 
approaching SATA 3 Gbit/s transfer rate, and this is being addressed 
with the SATA 6Gb/s interoperability standard."

[Sorry, the SATA quotes above is not linear with the article.]

So what does it all mean to you and I, the home consumer, or to business?

- SCSI will become even less popular, given the higher cost than SATA, 
as SATA performance improves, in (large?) RAID arrays / shelves?

- even fewer devices than I would wish, like laptops, will come with 
eSata ports in favour of USB3?

- OpenWRT devices won't get eSata (typically?) or USB3, the processors 
not (typically?) being able to keep up to the drive? Vis a vis NAS?

- the benefits of USB3 or SATA3 will largely only be felt by multiple 
drives feeding multiple servers that feed multiple clients (e.g. SQL). 
Even if the clients are all connected Gigabit, the net will still be 
the limiting factor?

   - even on a home connected web server, the limitation will be ISP 
speed, not disk speed? IIRC, typically systems are disk bound, but 
there's a whole lot of 'stuff' between the CPU and disk, preventing an 
increase in disk speed from providing an equivalent increase in system 
speed? (And this is even presuming sufficient requests are coming in 
sufficiently often to "make 'it' worthwhile"?

- systems, even more so, will stop coming with moving parts hard 
drives, in favour of much faster (and more reasonably smaller) flash 
drives for the OS. Particularly blade centers?

- if drives are SATA, and the bus is USB3, why USB3 instead of just 
staying with (e)SATA?

- why USB3? HD video streaming? (Limited, by ISP speeds?)

- NAS vs more drives in a current system, the question still goes 
begging. Especially for the home, if it's not a GUI I don't understand 
it, crowd.

More information about the kwlug-disc mailing list