I have one @dolphs what do you want me to do/test ?
VPN throughput not sufficient for HD streaming
Started by dolphs, 27 Mar 2016 07:33
66 replies to this topic
Re: VPN throughput not sufficient for HD streaming #61
-
- Senior Member
- 765 posts
+4
Neutral
Posted 10 December 2016 - 00:04
Vu+ Ultimo 4K (OpenPLI-9.0) :: 1x DVB-C FBC :: ZIGGO
Vu+ Ultimo 4K (OpenPLI-9.0) :: 1x DVB-C FBC + 1x DVB-S2 FBC :: ZIGGO
Re: VPN throughput not sufficient for HD streaming #62
-
- Senior Member
- 987 posts
+8
Neutral
Posted 10 December 2016 - 06:12
Please log on to box and show results of
openssl speed -evp aes-128-cbc
openssl speed -evp aes-256-cbc
Edited by dolphs, 10 December 2016 - 06:13.
Re: VPN throughput not sufficient for HD streaming #63
-
- Senior Member
- 765 posts
+4
Neutral
Posted 10 December 2016 - 10:05
root@vuuno4k:~# openssl speed -evp aes-128-cbc Doing aes-128-cbc for 3s on 16 size blocks: 12894632 aes-128-cbc's in 2.97s Doing aes-128-cbc for 3s on 64 size blocks: 3618516 aes-128-cbc's in 2.99s Doing aes-128-cbc for 3s on 256 size blocks: 955382 aes-128-cbc's in 2.98s Doing aes-128-cbc for 3s on 1024 size blocks: 242240 aes-128-cbc's in 2.98s Doing aes-128-cbc for 3s on 8192 size blocks: 30377 aes-128-cbc's in 2.98s OpenSSL 1.0.2a 19 Mar 2015 built on: reproducible build, date unspecified options:bn(64,32) rc4(ptr,char) des(idx,cisc,16,long) aes(partial) idea(int) blowfish(ptr) compiler: arm-oe-linux-gnueabi-gcc -march=armv7-a -mfloat-abi=hard -mfpu=neon --sysroot=/media/build/vti-dorie/build/tmp/sysroots/vuuno4k -I. -I.. -I../include -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -DTERMIO -Os -pipe -g -feliminate-unused-debug-types -Wall -Wa,--noexecstack -DHAVE_CRYPTODEV -DUSE_CRYPTODEV_DIGESTS -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DAES_ASM -DBSAES_ASM -DGHASH_ASM The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 69466.03k 77453.19k 82073.08k 83239.52k 83506.17k
root@vuuno4k:~# openssl speed -evp aes-256-cbc Doing aes-256-cbc for 3s on 16 size blocks: 10030421 aes-256-cbc's in 2.97s Doing aes-256-cbc for 3s on 64 size blocks: 2721730 aes-256-cbc's in 2.98s Doing aes-256-cbc for 3s on 256 size blocks: 707093 aes-256-cbc's in 2.98s Doing aes-256-cbc for 3s on 1024 size blocks: 178380 aes-256-cbc's in 2.98s Doing aes-256-cbc for 3s on 8192 size blocks: 22360 aes-256-cbc's in 2.97s OpenSSL 1.0.2a 19 Mar 2015 built on: reproducible build, date unspecified options:bn(64,32) rc4(ptr,char) des(idx,cisc,16,long) aes(partial) idea(int) blowfish(ptr) compiler: arm-oe-linux-gnueabi-gcc -march=armv7-a -mfloat-abi=hard -mfpu=neon --sysroot=/media/build/vti-dorie/build/tmp/sysroots/vuuno4k -I. -I.. -I../include -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -DTERMIO -Os -pipe -g -feliminate-unused-debug-types -Wall -Wa,--noexecstack -DHAVE_CRYPTODEV -DUSE_CRYPTODEV_DIGESTS -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DAES_ASM -DBSAES_ASM -DGHASH_ASM The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-256-cbc 54035.94k 58453.26k 60743.56k 61295.68k 61674.45k
Vu+ Ultimo 4K (OpenPLI-9.0) :: 1x DVB-C FBC :: ZIGGO
Vu+ Ultimo 4K (OpenPLI-9.0) :: 1x DVB-C FBC + 1x DVB-S2 FBC :: ZIGGO
Re: VPN throughput not sufficient for HD streaming #64
-
- Senior Member
- 765 posts
+4
Neutral
Posted 10 December 2016 - 10:06
Im receiving my VU+ Ultimo 4K in a few days will post results here 
Vu+ Ultimo 4K (OpenPLI-9.0) :: 1x DVB-C FBC :: ZIGGO
Vu+ Ultimo 4K (OpenPLI-9.0) :: 1x DVB-C FBC + 1x DVB-S2 FBC :: ZIGGO
Re: VPN throughput not sufficient for HD streaming #65
-
- Senior Member
- 1,699 posts
+46
Good
Posted 10 December 2016 - 14:15
Dunno if anyone mentioned it...
I have improved home network streaming (3D and UHD files from a NAS) by replacing old CAT5 LAN cables with new CAT7 ones... 
At £2 - 5 a piece it's not a big expense but it prevents needless problems, from my experience...
<span style='font-family: comic sans ms,cursive'>"Enlightenment is man's emergence from his self-incurred immaturity. Immaturity is the inability to use one's own understanding without the guidance of another. This immaturity is self-incurred if its cause is not lack of understanding, but lack of resolution and courage to use it without the guidance of another. The motto of enlightenment is therefore: Sapere aude! Have courage to use your own understanding!</span><br /> <br /><span style='font-family: comic sans ms,cursive'>Laziness and cowardice are the reasons why such a large proportion of men, even when nature has long emancipated them from alien guidance..." I. Kant, "Political writings" (1784)</span><br /> <br /><span style='font-family: comic sans ms,cursive'><a class='bbc_url' href='<a class='bbc_url' href='http://eserver.org/p...lightenment.txt'>http://eserver.org/p...ent.txt</a>'><a class='bbc_url' href='http://www.english.upenn.edu/~mgamer/Etexts/kant.html</a>'>http://www.english.upenn.edu/~mgamer/Etexts/kant.html</a></a> - the jolly text on Enlightenment, at the basis of Modernity...</span>
Re: VPN throughput not sufficient for HD streaming #66
-
- Senior Member
- 103 posts
+2
Neutral
Posted 10 December 2016 - 18:08
Although I see little value in these tests in relation to VPN performance, except for theoretical crypto performance, I will add results of a N3150 CPU with AES-NI support.
They by no means reflect real world performance, for that one needs to test in a real world scenario.
1. Hardware crypto support (AES-NI) disabled:
2. Hardware crypto support (AES-NI) enabled:
3. Hardware crypto support (AES-NI) and hmac-sha1 (Hashing)
In case of OpenVPN, as a minimum one wants data to be encrypted, determined by the cipher directive.
Lowering cipher to AES-128-CBC will help somewhat.
One could also disable the hashing (auth none), determined by the auth directive, for data packets to take away load from CPU.
Even though our set-top-boxes generally do not support AES-NI, we will still somewhat benefit from AES-GCM when OpenVPN 2.4 becomes available end of December...
They by no means reflect real world performance, for that one needs to test in a real world scenario.
1. Hardware crypto support (AES-NI) disabled:
env OPENSSL_ia32cap=0 openssl speed -elapsed -evp aes-256-cbc -multi 8
evp 97659.74k 117140.11k 123383.16k 307494.50k 339199.79k
CPU user 99-100% on all cores
CPU kernel 0-0,3% on all cores
2. Hardware crypto support (AES-NI) enabled:
Compare 1 and 2. Improved performance using hardware crypto support (AES-NI)openssl speed -elapsed -evp aes-256-cbc -multi 8
evp 618920.09k 882027.93k 1003271.00k 1041262.20k 1150214.31k
CPU user 98-100% on all cores
CPU kernel 0-0,3% on all cores
3. Hardware crypto support (AES-NI) and hmac-sha1 (Hashing)
Compare 2 and 3. Decreased performance when including SHA1 HMAC.openssl speed -elapsed -evp aes-256-cbc-hmac-sha1 -multi 8
evp 214362.25k 408669.56k 440537.10k 468988.33k 505701.81k
CPU user 98-100% on all cores
CPU kernel 0-0,3% on all cores
In case of OpenVPN, as a minimum one wants data to be encrypted, determined by the cipher directive.
Lowering cipher to AES-128-CBC will help somewhat.
One could also disable the hashing (auth none), determined by the auth directive, for data packets to take away load from CPU.
Even though our set-top-boxes generally do not support AES-NI, we will still somewhat benefit from AES-GCM when OpenVPN 2.4 becomes available end of December...
Today's scientists have substituted mathematics for experiments, and they wander off through equation after equation, and eventually build a structure which has no relation to reality. Nikola Tesla
Re: VPN throughput not sufficient for HD streaming #67
-
- Senior Member
- 765 posts
+4
Neutral
Posted 14 December 2016 - 20:44
root@vuultimo4k:~# openssl speed -evp aes-128-cbc Doing aes-128-cbc for 3s on 16 size blocks: 11188077 aes-128-cbc's in 2.98s Doing aes-128-cbc for 3s on 64 size blocks: 3173450 aes-128-cbc's in 2.97s Doing aes-128-cbc for 3s on 256 size blocks: 838157 aes-128-cbc's in 2.98s Doing aes-128-cbc for 3s on 1024 size blocks: 212530 aes-128-cbc's in 2.98s Doing aes-128-cbc for 3s on 8192 size blocks: 26646 aes-128-cbc's in 2.98s OpenSSL 1.0.2a 19 Mar 2015 built on: reproducible build, date unspecified options:bn(64,32) rc4(ptr,char) des(idx,cisc,16,long) aes(partial) idea(int) blowfish(ptr) compiler: arm-oe-linux-gnueabi-gcc -march=armv7-a -mfloat-abi=hard -mfpu=neon --sysroot=/media/build/vti-dorie/build/tmp/sysroots/vuultimo4k -I. -I.. -I../include -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -DTERMIO -Os -pipe -g -feliminate-unused-debug-types -Wall -Wa,--noexecstack -DHAVE_CRYPTODEV -DUSE_CRYPTODEV_DIGESTS -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DAES_ASM -DBSAES_ASM -DGHASH_ASM The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 60070.21k 68384.11k 72002.75k 73030.44k 73249.68k
root@vuultimo4k:~# openssl speed -evp aes-256-cbcDoing aes-256-cbc for 3s on 16 size blocks: 8769350 aes-256-cbc's in 2.99s Doing aes-256-cbc for 3s on 64 size blocks: 2388097 aes-256-cbc's in 2.97s Doing aes-256-cbc for 3s on 256 size blocks: 620325 aes-256-cbc's in 2.99s Doing aes-256-cbc for 3s on 1024 size blocks: 156607 aes-256-cbc's in 2.97s Doing aes-256-cbc for 3s on 8192 size blocks: 19621 aes-256-cbc's in 2.98s OpenSSL 1.0.2a 19 Mar 2015 built on: reproducible build, date unspecified options:bn(64,32) rc4(ptr,char) des(idx,cisc,16,long) aes(partial) idea(int) blowfish(ptr)compiler: arm-oe-linux-gnueabi-gcc -march=armv7-a -mfloat-abi=hard -mfpu=neon --sysroot=/media/build/vti-dorie/build/tmp/sysroots/vuultimo4k -I. -I.. -I../include -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -DTERMIO -Os -pipe -g -feliminate-unused-debug-types -Wall -Wa,--noexecstack -DHAVE_CRYPTODEV -DUSE_CRYPTODEV_DIGESTS -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DAES_ASM -DBSAES_ASM -DGHASH_ASM The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-256-cbc 46926.29k 51460.68k 53111.44k 53995.14k 53938.00k
Edited by WanWizard, 14 December 2016 - 21:51.
fixed formatting
Vu+ Ultimo 4K (OpenPLI-9.0) :: 1x DVB-C FBC :: ZIGGO
Vu+ Ultimo 4K (OpenPLI-9.0) :: 1x DVB-C FBC + 1x DVB-S2 FBC :: ZIGGO
8 user(s) are reading this topic
0 members, 8 guests, 0 anonymous users
