Berkeley CSUA MOTD:Entry 52933
Berkeley CSUA MOTD
 
WIKI | FAQ | Tech FAQ
http://csua.com/feed/
2025/05/25 [General] UID:1000 Activity:popular
5/25    

2009/5/2-6 [Computer/HW/Laptop, Computer/HW/Drives] UID:52933 Activity:kinda low
5/2     I have a 1Gbps switch. It is connected to a SANS RAID-1 with
        two WD Green 1T that takes 1Gbps connection, as well as a laptop
        that is also 1Gbps. Online benchmarks show *sustained* transfer
        rate of about 1/2 of 1Gbps with the WD Green... cool!!! But
        in practice, I'm only getting 68Mbps (read SANS->PC file transfer),
        far below what I expected. The switch says both devices are using
        1Gbps. BTW, setting the SANS on Jumble Frame/no jumble doesn't
        change the transfer rate at all. What's going on?
        \_ Are you limited by your PC's drive speed? 68 Mbps is 8.5 MB/s
           which may be all your drive can do. What do you get copying
           data on your drive locally?
           \_ When I copy a 796MB file from the same drive, it took about
              a minute. So I figured... 796MB read and 796MB write (gross
              56 sec. So I figured... 796MB read and 796MB write (gross
              assumption that they're about the same, but false).
              That's 796*2 / 60 = 26MB/sec. So I hardly think I'm limited
              by my drive.
              That's 796*2 / 56 = 28MB/sec. So I hardly think I'm limited
              by my drive. I'm almost certain the bottleneck is due to
              something stupid I did, but seriously, I checked that NAS
              is in fact set to use 1000Mbps (or no connection), and I know
              my laptop transfer speed is super fast at work. Reallly baffled...
              my laptop transfer speed is super fast at work. Reallly
              baffled...
              \_ Your drive can't do any better than 224 Mbps. Less over
                 the network given the overhead. Figure maybe 0.66 of 224
                 for 135 Mbps. That's only a factor of two different from
                 what you are seeing. Maybe there is some caching going on
                 to give you that 2x. If the NAS is *really* capable of
                 500 Mbps (which is what my $100K Netapp striped across
                 many disks can do so it seems unlikely) then I'd look to
                 your drive or your network card. Could also be that your
                 switch sucks. Try to connect with a crossover cable to
                 see if that helps.
        \_ What is the Sustained Transfer Rate for the disks in your NAS?
           Odd are it is about 25MBps/per disk. How wide is your stripe?
           http://www.storagereview.com/map/lm.cgi/str
2025/05/25 [General] UID:1000 Activity:popular
5/25    

You may also be interested in these entries...
2013/8/16-10/28 [Computer/HW/Laptop] UID:54728 Activity:nil
8/16    I just left my employer. They didn't ask to get their HW back.
        Is is common for employers to contact me a few months/years later
        and ask for their HW back?
        \_ Is it an iPhone 9 prototype? :-)
           \_ as a matter of fact, it is just a 2 year old laptop
              and they don't seem to keep track of inventor
	...
2013/7/8-8/23 [Computer/HW/Laptop] UID:54706 Activity:nil
7/8     Suppose I just stopped going to work and stopped communicating
        with them. How many days will it take before they stop sending me
        paycheck (and maybe even try to get back the laptop they gave me)?
        \_ They'll probably try to call your emergency contact first to see if
           you're hit a tree or something.
           you've hit a tree or something.
	...
2011/12/21-2012/2/6 [Computer/HW/Laptop, Computer/SW/Editors/Emacs] UID:54269 Activity:nil
12/21   In Emacs, how do I make it so that it will resize its screen
        when the maximum resolution of my monitor changes? When I
        use my laptop, my emacs is too big (and I can't resize it
        because the bottom-right corner is not accessible).
        \_ Which OS?  Can't you drag the top border (not a corner) to resize
           the height to be smaller?
	...
2011/5/9-7/13 [Computer/Companies/Apple, Computer/SW/OS/OsX] UID:54106 Activity:nil
5/4     Any thoughts about apple switching to ARM for all machines?
        \_ Could only have happened if the Mac App Store had been a wild, wild
           success, and they were willing to give up Steam. Games/dual-booting
           is strategically valuable to not driving people away from macs.
           A new ARM iOS based device, or opening the Apple TV to iOS apps
           (same thing, really) would be much more likely than sabotaging their
	...
2011/3/12-4/20 [Consumer/CellPhone, Computer/HW/Laptop] UID:54057 Activity:nil
3/12    I am curious what others think of tablets like iPad. They don't seem
        useful to me, but I use my computer for more than web browsing,
        Facebook, and Twitter. Why would I buy one instead of a laptop?
        They seem like a disabled laptop to me, but at a higher price.
        \_ You are most likely a coder.  iPad is not for coders.  They are
           what you get your non-technical friends.  Or musicians.  Look at
	...
2010/8/23-9/7 [Computer/SW/OS/Windows] UID:53932 Activity:low
9/21    are you guys really all using win7 in some way now?
        \_ Yes.  In my company, we have upgraded all our machines for
           development, QA and other office use (HR, Finance, Receptions, etc.)
           from WinXP to Win7 a few months ago.  Our products now primarily
           support Win7/Win2008 and secondarily support WinXP/Win2003.
           \_ Any weird gotchas support-wise?
	...
2010/5/17-26 [Computer/SW/OS/OsX] UID:53835 Activity:nil
5/13    I am getting a new Macbook Pro. Should I get a glossy or matte screen?
        I always had matte in the past. I use my laptop just about everywhere
        and glare is a concern. Should I stick with matte? I noticed
        glossy is the default now.
        \_ It is weird, but I have the glossy screen that goes to the edge of
           the monitor for my 13" PB and it somehow seems bigger than the one
	...
2010/4/5-15 [Computer/HW/Laptop, Computer/SW/OS/OsX] UID:53770 Activity:nil
4/4     My macbook is dying. After three and a half years, I want to retire it,
        and get a new one, but I live in constant fear* that Apple will pull a fast
        one on me and update them as soon as I do. Anyone know anything about an
        impending Apple laptop refresh?
        * = hyperbole
        \_ http://buyersguide.macrumors.com
	...
2010/1/22-30 [Computer/HW/Laptop, Computer/SW/OS/OsX] UID:53655 Activity:high
1/22    looking to buy a new development laptop
        needs ssdrive, >6 hr possible batt life, and runs linux reasonably
        Anyone have a recommendation? Thx.
        \_ thinkpad t23 w ssdrive and battery inplace of drive bay
        \_ Ever wondered what RICHARD STALLMAN uses for a laptop?  Well,
           wonder no more!
	...
2009/10/24-11/3 [Computer/HW/Laptop] UID:53466 Activity:kinda low
10/24   How well do you see color? I got 8, how about you?
        http://www.xrite.com/custom_page.aspx?PageID=77
        \_ 7
           \_ what monitor did you use?
              \_ LCD on thinkpad x32, under not so great lighting conditions.
        \_ I scored 101, which seems impossible. Then again, I didn't
	...
2012/1/4-2/6 [Computer/HW/Drives] UID:54281 Activity:nil
1/4     I want to test how my servers behave during a disk failure and
        a RAID reconstruction so I want to simulate a hardware failure.
        How can I do this in Linux without having to physically pull
        a drive? These disks are behind a RAID card and run Linux. -ausman
        \_ According to the Linux RAID wiki, you might be able to use mdadm
           to do this with something like the following:
	...
2011/9/14-10/25 [Computer/HW/Drives] UID:54173 Activity:nil
9/13    Thanks to Jordan, our disk server is no longer virtualized. Our long
        nightmare of poor IO performance should hopefully be over. Prepare for
        another long nightmare of poor hardware reliability!
        ...
        Just kidding! (I hope)
        In any case, this means that cooler was taken out back and shot, and
	...
2011/2/14-4/20 [Computer/SW/Unix] UID:54039 Activity:nil
2/14    You sure soda isn't running windows in disguise?  It would explain the
        uptimes.
        \_ hardly, My winbox stays up longer.
        \_ Nobody cares about uptime anymore brother, that's what web2.0 has
           taught us.  Everything is "stateless".
           \_ You;d think gamers would care more about uptime.
	...
2011/2/18-4/20 [Computer/SW/Unix] UID:54044 Activity:nil
2/18    Why does the system seem so sluggish lately?
        \_ Slow NFS is basically always the answer. --toulouse
        \_ Any truth to the rumor that soda will be decommissioned this summer?
           \_ Absolutely none. Soda might go down temporarily while disks are
              reorganized and stuff so soda doesn't suffer from such shitty
              performance nearly as much, but no, we've gotta maintain NFS and
	...
2010/7/22-8/9 [Computer/SW/OS/FreeBSD, Computer/HW/Drives] UID:53893 Activity:nil
7/22    Playing with dd if=/dev/random of=/dev/<disk> on linux and bsd:
        2 questions, on linux when <disk>==hda it always gives me this off
        by one report i.e. Records out == records in-1 and says there is an
        error. Has anyone else seen this?  Second, when trying to repeat this
        on bsd, <disk>==rwd0 now, to my surprise, using the install disk and
        selecting (S)hell, when I try to dd a 40 gig disk it says "409 records
	...
2009/10/27-11/3 [Computer/HW/Drives] UID:53474 Activity:nil
10/27   I just read an article that Facebook had moved their database
        to all SSD to speed throughput, but now I can't find it. Has
        anyone else seen this? Any experience with doing this? -ausman
        \_ I hope you're not running mission critical data:
           http://ask.slashdot.org/story/09/10/27/1559248/Reliability-of-PC-Flash-SSDs?from=rss
        \_ Do you have any idea how much storage space is used by Facebook,
	...
2009/8/4-13 [Computer/SW/OS/Windows] UID:53239 Activity:kinda low
8/3     VMWare + Windows XP + Validation question. I need to test stuff with
        Service Pack 3 installed. I have a valid key that I own (yeah yeah I
        actually *bought* a copy, please don't flame me for supporting evil
        M$). Is it possible to register the key once, and then duplicate it
        for testing purposes?  Will Windows or Microsoft detect copies and
        disable the rest the copies?
	...
2009/7/28-8/6 [Computer/HW/Drives] UID:53216 Activity:nil
7/28    Does it make sense to defragment disks on VMWare? My 80GB disk
        on VMWare isn't really using 80GB, it just uses what it needs.
        Will defragment do anything to it?
        \_ If you want to speed up disk operation in your VM, it's best to
           defragment the disks in your VM, then defragment the disk on your
           host machine where the VM files are.
	...
Cache (4864 bytes)
www.storagereview.com/map/lm.cgi/str
For real-world transfers of average files, what we are concerned with is the rate at which the drive can transfer data sequentially from multiple tracks and cylinders on the disk. This specification is the drive's sustained transfer rate (sometimes the sequential transfer rate), abbreviated STR. Sustained transfer rates are most relevant for reflecting the drive's performance when dealing with largish files. Also, STR is normally measured in bytes, not bits like the media transfer rate, and includes only data, not the overhead portions of each sector or track. An example: let's say we want to read a 4 MB file from a hard disk that has 300 sectors per track in the zone where the file is located; If the drive has three platters and six surfaces, this means that if this file is stored sequentially, it will on average occupy 26 tracks over some portion of 5 cylinders. Reading this file in its entirety would require (at least) 25 head switches and 4 cylinder switches. STR can be calculated from various characteristics of a disk, but this isn't nearly as conceptually simple as calculating a media transfer rate on a single track. Rather than just provide a lengthy formula, I'll try to explain how the calculation is done. A transfer rate is of course data transferred per unit of time. So our equation will be a ratio of data transferred to the time taken to transfer it. Now, to represent a sustained transfer we need to cover an entire cylinder, so we include all the head switches while reading the cylinder, and one cylinder switch time as well (to get us to the next cylinder). The data that is transferred for an entire cylinder read is as follows: Data transferred per cylinder = Number of surfaces * Sectors per track * 512 bytes where "number of surfaces" is identical to the number of tracks per cylinder, of course. First, we of course have to wait for the disk to make one complete revolution for each track read, as the data is read. Then we need to add a number of head switches equal to the number of surfaces less one, and finally, one cylinder switch. The final equation then looks like this: STR = (Number of surfaces * Sectors per track * 512) / ( 2 * Number of surfaces * Latency + (Number of surfaces - 1) * Head Switch Time + Cylinder Switch Time) The result is in bytes per second. This drive has 452 sectors in its outermost zone, and a 7200 RPM spin speed (for latency of 417 ms). This family's head switch time is 15 ms and cylinder switch time is 20 ms. Since the drive with more platters performs a higher ratio of (faster) head switches compared to (slower) cylinder switches, its STR is a bit higher. Still, it's only a difference of less than 1% between the biggest and smallest members of the family. An important question to consider is how meaningful the STR numbers really are: if you have the drive above, will it really let you read at a rate of about 23 MB/second? First, since STR is derived directly from the media transfer rate, its value also depends on what part of the disk is being read; larger outer cylinders have the highest STR, smaller inner cylinders have the lowest. Second, there's the matter of whether the access is really sequential. Once you fragment the file, you aren't doing a consecutive data transfer any more. Each fragment of the file introduces the need for an additional positioning step to the location where the next piece starts, which slows the transfer and introduces other factors into the performance measurement. Finally, real-world transfers incur all sorts of penalties due to operating system overhead and other considerations. A good rule of thumb in the computer world is that you never get the theoretical maximum of anything. STR has in the last few years started to get more attention than it traditionally has--some would say too much. It is important to those who do a lot of work with large files, but not as critical to those who work with a large number of smaller files, which includes many, if not most, Windows users. The number of platters influences it by changing the mix of head and cylinder switches; actuator design and controller circuitry affect the switch times; media issues and spindle speed influence the all-important underlying media transfer rates. There's probably no other performance specification that is affected by so many different design factors. A final point about internal sustained transfer rates vs. In order to get the most from the hard disk, the interface must be fast enough to be able to handle the maximum STR of the drive. This is usually not a problem because most disk interfaces have sufficient "headroom" to handle drives that run on them. It will work, but clearly you will not get STR of 23 MB/s over that interface. The converse is that putting a drive on an interface much faster than it won't improve performance much;