GELI encryption - CPU requirements?

[Dan Naumov]

This is great, thanks a lot! Just as I got depressed that if I would
be stuck with 18MB/s throughput if I went with my planned route, you
provide more results that show that Blowfish would be a great choice
for what I want to do, 50 MB/s is going to be plenty for my needs.
Your results also seem to be backed by these benchmark results I found
(Also ran on an Atom):
http://www.mail-archive.com/support@pfsense.com/msg15423.html

Thanks a lot again!

- Dan Naumov



On Thu, May 28, 2009 at 12:22 AM, Volker <volker at vwsoft.com> wrote:
>
> On 05/27/09 22:58, Dan Naumov wrote:
>> Could you also run the test for: Camellia and Blowfish please?
>
> Sure but do not count on the absolute values as the system wasn't idling
> 100% (~88-96%):
>
> type             16 bytes     64 bytes    256 bytes   1024 bytes   8192
> bytes
> camellia-128 cbc    20121.31k    21752.35k    21556.40k    21838.64k
> 22375.53k
> camellia-192 cbc    16804.24k    17087.62k    17283.35k    17494.13k
> 16584.06k
> camellia-256 cbc    15925.81k    17256.89k    17425.49k    17605.67k
> 17092.26k
>
> type             16 bytes     64 bytes    256 bytes   1024 bytes   8192
> bytes
> blowfish cbc     49186.75k    51979.89k    53195.92k    53742.92k
> 52371.59k
>
> Also I found the sha values of interest:
>
> type             16 bytes     64 bytes    256 bytes   1024 bytes   8192
> bytes
> sha1              3303.61k     8459.49k    19921.03k    33585.85k
> 47594.24k
> sha256            3408.83k     8519.20k    15886.04k    20473.32k
> 21481.23k
> sha512            1065.92k     4681.24k     7383.90k    10189.10k
> 11597.66k
>
>
>>
>>> For reference, here're the values taken on a dual core Atom:
>>>
>>> type             16 bytes     64 bytes    256 bytes   1024 bytes   8192
>>> bytes
>>> aes-128 cbc      17947.16k    18502.91k    18703.91k    18271.91k
>>> 18955.39k
>>>
>>> aes-192 cbc      16404.93k    15966.46k    16615.41k    16115.26k
>>> 16466.56k
>>>
>>> aes-256 cbc      13711.70k    14016.79k    14342.35k    14109.98k
>>> 14738.16k
>>>
>>>
>>> FreeBSD dardanos 7.2-STABLE FreeBSD 7.2-STABLE #10 r192673: Sun May 24
>>> 10:22:05 CEST 2009
>>>
>>> CPU: Intel(R) Atom(TM) CPU  330   @ 1.60GHz (1618.44-MHz 686-class CPU)
>>>  Origin = "GenuineIntel"  Id = 0x106c2  Stepping = 2
>>>
>>> Features=0xbfe9fbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
>>>  Features2=0x40e31d<SSE3,DTES64,MON,DS_CPL,TM2,SSSE3,CX16,xTPR,PDCM,<b22>>
>>>  AMD Features=0x20100000<NX,LM>
>>>  AMD Features2=0x1<LAHF>
>>>  Cores per package: 2
>>>  Logical CPUs per core: 2
>>> real memory  = 2137391104 (2038 MB)
>>> avail memory = 2077528064 (1981 MB)
>>> ACPI APIC Table: <INTEL  D945GLF2>
>>> FreeBSD/SMP: Multiprocessor System Detected: 4 CPUs
>>>  cpu0 (BSP): APIC ID:  0
>>>  cpu1 (AP/HT): APIC ID:  1
>>>  cpu2 (AP): APIC ID:  2
>>>  cpu3 (AP/HT): APIC ID:  3
>>> ioapic0: Changing APIC ID to 2
>>> ioapic0 <Version 2.0> irqs 0-23 on motherboard
>>> kbd1 at kbdmux0
>>> cryptosoft0: <software crypto> on motherboard
>
>