Re: [RFC] LZO de/compression support - take 3

[Nitin Gupta]

Hi Pavel,

Just did some benchmarking; results below.

On 5/25/07, Pavel Machek <pavel@xxxxxx> wrote:
> What is the performance difference between safe and unsafe version?

File size: 256K
- Following as tests for original test code - not any kernel port of this.
- Test with each block size repeated 5 times - taken avg. of these 5 runs.
- Same file used for each test.
- Used lzotest utility (included with LZO 2.02) for testing.

Blocksize	Comp*		DU*		DS*		Speed%
4		59.356		211.526		195.260		7.689
8		54.623		202.712		188.369		7.075
16		50.342		196.482		183.988		6.358
32		47.499		189.800		177.455		6.504
64		44.148		178.724		167.201		6.447
128		42.125		170.229		159.257		6.445
256		41.830		155.035		146.115		5.753

* All speeds in MB/sec
Comp = LZO1X-1
DU = Decompress (unsafe)
DS = Decompress (safe)
Speed% = ((DU-DS)/DU)*100

I have yet to see how the kernel ports compare against this original version.

Cheers,
Nitin
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/