[TriLUG] macbook question

Cristóbal Palmer cmp at cmpalmer.org
Mon May 16 16:25:14 EDT 2011


On Mon, May 16, 2011 at 7:24 AM, Magnus <magnus at trilug.org> wrote:
>
> The naysayers forget what it was like to run VMware on a Pentium 4 with 1GB of RAM and have been spoiled by their i5's and i7's.

Don't get me wrong: it definitely works. The point is is that he can
get waaay better hardware bang for the buck buying something other
than the Air.

More numbers for my Natty VM on the Air:

34 seconds to boot.
6 seconds to start Firefox.

tiobench output:

cmpalmer at zomg:~$ sudo tiobench
No size specified, using 2000 MB
Run #1: /usr/bin/tiotest -t 8 -f 250 -r 500 -b 4096 -d . -TTT

Unit information
================
File size = megabytes
Blk Size  = bytes
Rate      = megabytes per second
CPU%      = percentage of CPU used during the test
Latency   = milliseconds
Lat%      = percent of requests that took longer than X seconds
CPU Eff   = Rate divided by CPU% - throughput per cpu load

Sequential Reads
                              File  Blk   Num                   Avg
  Maximum      Lat%     Lat%    CPU
Identifier                    Size  Size  Thr   Rate  (CPU%)  Latency
  Latency      >2s      >10s    Eff
---------------------------- ------ ----- ---  ------ ------ ---------
-----------  -------- -------- -----
2.6.38-8-generic              2000  4096    1   54.52 11.15%     0.071
      63.65   0.00000  0.00000   489
2.6.38-8-generic              2000  4096    2   51.76 12.75%     0.150
     163.68   0.00000  0.00000   406
2.6.38-8-generic              2000  4096    4   51.61 28.77%     0.301
     402.29   0.00000  0.00000   179
2.6.38-8-generic              2000  4096    8   51.83 45.29%     0.598
     393.71   0.00000  0.00000   114

Random Reads
                              File  Blk   Num                   Avg
  Maximum      Lat%     Lat%    CPU
Identifier                    Size  Size  Thr   Rate  (CPU%)  Latency
  Latency      >2s      >10s    Eff
---------------------------- ------ ----- ---  ------ ------ ---------
-----------  -------- -------- -----
2.6.38-8-generic              2000  4096    1    7.90 12.64%     0.493
      15.60   0.00000  0.00000    63
2.6.38-8-generic              2000  4096    2   11.60 27.45%     0.655
      50.49   0.00000  0.00000    42
2.6.38-8-generic              2000  4096    4   15.89 53.88%     0.942
      76.75   0.00000  0.00000    29
2.6.38-8-generic              2000  4096    8   15.42 116.4%     1.960
      64.40   0.00000  0.00000    13

Sequential Writes
                              File  Blk   Num                   Avg
  Maximum      Lat%     Lat%    CPU
Identifier                    Size  Size  Thr   Rate  (CPU%)  Latency
  Latency      >2s      >10s    Eff
---------------------------- ------ ----- ---  ------ ------ ---------
-----------  -------- -------- -----
2.6.38-8-generic              2000  4096    1   57.02 18.18%     0.065
    1265.17   0.00000  0.00000   313
2.6.38-8-generic              2000  4096    2   58.46 46.39%     0.125
    1128.73   0.00000  0.00000   126
2.6.38-8-generic              2000  4096    4   62.58 80.91%     0.235
    1139.66   0.00000  0.00000    77
2.6.38-8-generic              2000  4096    8   61.09 164.4%     0.476
    1479.20   0.00000  0.00000    37

Random Writes
                              File  Blk   Num                   Avg
  Maximum      Lat%     Lat%    CPU
Identifier                    Size  Size  Thr   Rate  (CPU%)  Latency
  Latency      >2s      >10s    Eff
---------------------------- ------ ----- ---  ------ ------ ---------
-----------  -------- -------- -----
2.6.38-8-generic              2000  4096    1   12.40 45.21%     0.030
      12.59   0.00000  0.00000    27
2.6.38-8-generic              2000  4096    2   15.43 119.4%     0.041
      16.69   0.00000  0.00000    13
2.6.38-8-generic              2000  4096    4   14.97 268.2%     0.043
      25.15   0.00000  0.00000     6
2.6.38-8-generic              2000  4096    8   15.04 336.8%     0.095
      72.80   0.00000  0.00000     4
cmpalmer at zomg:~$


Cheers,
-- 
Cristóbal Palmer
Systems Administrator
University of North Carolina at Chapel Hill



More information about the TriLUG mailing list