1 |
On Wed, May 25, 2011 at 11:58 AM, Volker Armin Hemmann |
2 |
<volkerarmin@××××××××××.com> wrote: |
3 |
> On Wednesday 25 May 2011 10:31:19 Paul Hartman wrote: |
4 |
> |
5 |
>> When I move the swap to a slow SD card instead (2MB/sec transfer |
6 |
>> rate), even in that slow device, swapoff on the eMMC swap partition |
7 |
>> with ~500M in-use takes about 2 or 3 minutes at most with the data |
8 |
>> being swapped slowly into the SD card. |
9 |
>> |
10 |
>> So I think in your case it should be much faster than that! |
11 |
> |
12 |
> you are comparing apples with oranges (harddisks with moving arms with solid |
13 |
> state devices). |
14 |
> |
15 |
> Do yourself a favour. Look up how long a harddisk needs to position its head. |
16 |
|
17 |
Yeah, measuring speed in microseconds versus milliseconds. :) |
18 |
|
19 |
My SD read/write speed is about 0.9MB/sec in benchmark, and about 2 |
20 |
MB/sec in real-life swap usage. Based on Alex's timings it seems his |
21 |
HDD is getting about 0.5 MB/sec in swap usage. HDD benchmark random |
22 |
read/write is usually faster than that. But probably you're right and |
23 |
the seek operations are really killing it. HDD is about 20 times |
24 |
slower than SD card in seeking. |
25 |
|
26 |
If he has a USB flash drive, or memory card reader, he can use it for |
27 |
swap and see if it's any better or worse. I am curious how it will |
28 |
compare. |
29 |
|
30 |
Or, if he has multiple disks, he could create more swap partitions |
31 |
with equal priority and kernel will stripe/load balance automatically, |
32 |
hopefully improving his swap performance. |
33 |
|
34 |
Or better yet, figure out why his system is swapping at all which is |
35 |
what he was going for I think. With 8 GB I think he should be able to |
36 |
disable swap entirely anyway. :) |