Benchmarking and Performance testing

The performance of an IPsec system depends on CPU, RAM, NICs, switches, kernel and configuration.

The Alteeve Niche's Anvil RN2-M2 platform

We ran some tests on this very fast hardware. The platform is based on a set of Fujitsu RX300 S8 servers (specification) The machine has a number of Intel Corporation 82599ES 10-Gigabit cards that are bonded. All NICs are connected to a set of Brocade ICX6610-24 switches. We picked one bonded pair of 10Gbps on interface bond1 for our IPsec tests. The Anvil comes with an 8 core Intel(R) Xeon(R) CPU E5-2637 v2 @ 3.50GHz with AES-NI support. The MTU was left at the default 9k setting. The kernel used was 2.6.32-504.1.3.el6.x86_64.

IPsec performance measured with iperf

iperf used with default settings

• 9.78 Gbits/sec unencrypted

• 1.27 Gbits/sec AES256-SHA1

• 1.39 Gbits/sec AES128-SHA1

• 197 Mbits/sec 3DES-SHA1

We did some additional tests, but those are less accurate. using protoport= we could use multiple IPsec SA's (in the hope that it would distribute better) or have encrypted and unencrypted streams going.

• two streams, one plaintext 8.64 Gbits/sec plaintext plus 1.24 Gbits/sec AES256-SHA1

• two streams AES256-SHA1: 819 Mbits/sec plus 615 Mbits/sec (possibly was aes128)

CPU/crypto performance measured with openssl

(AES-NI disabling done via export OPENSSL_ia32cap=~0x200000200000000)

Without AES-NI, no multi: openssl speed -evp aes-256-cbc

type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-256-cbc 241508.56k 266220.03k 273663.06k 276314.11k 275479.81k

With AES-NI, no multi: openssl speed -evp aes-256-cbc

type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-256-cbc 502470.66k 528580.69k 532890.45k 535901.87k 536368.47k

Without AES-NI, no multi: openssl speed -evp aes-128-cbc

type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 320425.43k 366515.97k 377561.00k 383643.99k 383777.51k

With AES-NI, no multi: openssl speed -evp aes-128-cbc

type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 688604.26k 732936.83k 742459.28k 748241.92k 748756.99k

With AES-NI, using all cores : openssl speed -multi 8 -evp aes-256-cbc

evp 3729202.24k 4009617.79k 4053305.43k 4065434.97k 4068764.33k

With AES-NI, using all cores : openssl speed -multi 8 -evp aes-128-cbc

evp 5033772.55k 5494390.59k 5632183.30k 5668856.15k 5679707.48k

NIC settigs

#ethtool eth1
Settings for eth1:
	Supported ports: [ FIBRE ]
	Supported link modes:   10000baseT/Full 
	Supported pause frame use: No
	Supports auto-negotiation: No
	Advertised link modes:  10000baseT/Full 
	Advertised pause frame use: No
	Advertised auto-negotiation: No
	Speed: 10000Mb/s
	Duplex: Full
	Port: Other
	PHYAD: 0
	Transceiver: external
	Auto-negotiation: off
	Supports Wake-on: umbg
	Wake-on: g
	Current message level: 0x00000007 (7)
			       drv probe link
	Link detected: yes

# ethtool -k eth1
Features for eth1:
rx-checksumming: on
tx-checksumming: on
	tx-checksum-ipv4: on
	tx-checksum-unneeded: off
	tx-checksum-ip-generic: off
	tx-checksum-ipv6: on
	tx-checksum-fcoe-crc: on [fixed]
	tx-checksum-sctp: on [fixed]
scatter-gather: on
	tx-scatter-gather: on
	tx-scatter-gather-fraglist: off [fixed]
tcp-segmentation-offload: on
	tx-tcp-segmentation: on
	tx-tcp-ecn-segmentation: off
	tx-tcp6-segmentation: on
udp-fragmentation-offload: off [fixed]
generic-segmentation-offload: on
generic-receive-offload: on
large-receive-offload: on
rx-vlan-offload: on
tx-vlan-offload: on
ntuple-filters: on
receive-hashing: on
highdma: on [fixed]
rx-vlan-filter: on [fixed]
vlan-challenged: off [fixed]
tx-lockless: off [fixed]
netns-local: off [fixed]
tx-gso-robust: off [fixed]
tx-fcoe-segmentation: on [fixed]
tx-gre-segmentation: off [fixed]
tx-udp_tnl-segmentation: off [fixed]
fcoe-mtu: off [fixed]
loopback: off [fixed]