Internet Protocol Version 6 Support
z/VM V4.4 provided Internet Protocol Version 6 (IPv6) support, in CP, for OSA-Express guest LANs operating in QDIO mode. (Note that this support does not apply to guest LANs operating in HIPER mode.) z/VM V5.1 enhances its IPv6 support, in TCP/IP, by allowing the stack to be configured for IPv6 networks connected through OSA-Express operating in QDIO mode. The stack can be configured to provide static routing of IPv6 packets and to send IPv6 router advertisements.
This section summarizes measurement results comparing IPv6 to IPv4 and comparing IPv4 to IPv4 over devices defined as IPv6 capable. Measurements were done using guest LAN and using OSA-Express Gigabit Ethernet cards. Additional IPv6 to IPv4 comparisons for the case of communication between Linux systems via z/VM Virtual Switch using the Layer2 transport mode are provided in Virtual Switch Layer 2 Support.
Methodology: An internal version of the Application Workload Modeler (AWM) was used to drive request-response (RR), connect-request-response (CRR) and streaming (STR) workloads. The request-response workload consisted of the client sending 200 bytes to the server and the server responding with 1000 bytes. This interaction was repeated for 200 seconds. The connect-request-response workload had the client connecting, sending 64 bytes to the server, the server responding with 8K and the client then disconnecting. This same sequence was repeated for 200 seconds. The streaming workload consisted of the client sending 20 bytes to the server and the server responding with 20MB. This sequence was repeated for 400 seconds.
A complete set of runs, consisting of 3 trials for each case, for 1, 10, 20 and 50 client-server pairs, was done with the maximum transmission unit (MTU) set to 1492 and 8992.
The measurements were done on a 2064-109 with 3 dedicated processors in each LPAR used. Each LPAR had 1GB of central storage and 2GB expanded storage. CP monitor data was captured for one LPAR (client side) during the measurement and reduced using the Performance Toolkit (Perfkit).
Figure 1. Guest LAN Environment
![]() |
Figure 2. OSA QDIO Environment
![]() |
Figure 1 shows the measurement environment for guest LAN where the client communicates with its stack (tcpip1), the client stack sends the request over a guest LAN to the server stack (tcpip2), which then sends the request to the server. Figure 2 shows the measurement environment for OSA-Express where the client communicates with its stack (tcpip1), the client stack sends the request over the OSA-Express card to the server stack (tcpip2) in another LPAR, and the server stack then sends the request to the server.
Results: The following tables compare the average of 3 trials for each measurement between IPv4 and IPv4 over IPv6 capable devices (noted as v5 in the tables), and between IPv4 and IPv6. The numbers shown are the percent increase (or decrease) relative to IPv4. A positive number for throughput (either MB/sec or trans/sec) is good and a negative number for CPU time is good. Our target was for IPv6 to be within 3% of the throughput and CPU time for IPv4. For guest LAN, this was true for all workloads. For the OSA Express GigaBit Ethernet card (noted as QDIO in the tables) this is not true for the STR and CRR workloads when the MTU size is 1492.
- For streaming, throughput was 10% to 19% less for IPv4 over IPv6 capable devices and 12% to 25% less for IPv6. CPU time was 10% to 40% higher for IPv4 over IPv6 and 11% to 40% higher for IPv6.
- For CRR, throughput for both IPv4 over IPv6 capable devices and for IPv6 met our target overall. However, CPU time was 3% to 5% higher for IPv4 over IPv6 capable and 12% to 13% higher for IPv6.
Number of clients | 1 | 10 | 20 | 50 |
MTU 1492 runid V4 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB |
|
|
|
|
runid V5 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | vg5sn01 77.23 24.78 13.28 11.50 | vg5sn10 57.97 31.14 17.48 13.66 | vg5sn20 48.03 37.20 21.34 15.86 | vg5sn50 37.97 48.31 27.68 20.63 |
runid V6 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | vg6sn01 75.10 24.48 12.98 11.49 | vg6sn10 56.57 31.38 17.45 13.93 | vg6sn20 47.77 37.56 21.35 16.20 | vg6sn50 38.50 47.97 27.53 20.44 |
%diff V4 to V5 |
|
|
|
|
MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | 0% 0% 0% 0% | 3% -1% -1% -1% | 1% -1% -1% 0% | -1% 0% 0% 1% |
%diff V4 to V6 |
|
|
|
|
MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | -3% -2% -3% 0% | 0% 0% -1% 1% | 0% 0% -1% 2% | 1% 0% -1% 0% |
MTU 8992 runid V4 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB |
|
|
|
|
runid V5 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | vg5sj01 103.93 19.35 9.38 9.97 | vg5sj10 111.43 23.33 11.44 11.89 | vg5sj20 93.63 27.11 13.61 13.50 | vg5sj50 74.90 32.00 16.10 15.90 |
runid V6 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | vg6sj01 104.53 18.80 8.91 9.89 | vg6sj10 114.83 22.52 10.82 11.70 | vg6sj20 95.10 26.72 13.21 13.51 | vg6sj50 75.50 31.83 15.89 15.93 |
% diff V4 to V5 |
|
|
|
|
MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | 0% 2% 2% 2% | -1% 2% 2% 2% | 0% 1% 1% 1% | 0% 0% 0% 0% |
% diff V4 to V6 |
|
|
|
|
MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | 1% 1% 0% 3% | 2% -2% -4% 0% | 1% -1% -2% 1% | 1% -1% -1% 0% |
Note: 2064-109; z/VM 5.1.0; TCP/IP 510; V5 denotes IPv4 over IPv6 capable |
Number of clients | 1 | 10 | 20 | 50 |
---|---|---|---|---|
MTU 1492 runid V4 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans |
|
|
|
|
runid V5 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vg5cn01 164.23 6.25 5.08 1.17 | vg5cn10 421.82 4.54 3.49 1.05 | vg5cn20 475.38 4.10 3.09 1.02 | vg5cn50 571.86 3.97 2.86 1.11 |
runid V6 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vg6cn01 208.06 4.91 3.74 1.17 | vg6cn10 433.58 4.48 3.42 1.06 | vg6cn20 485.52 4.08 3.05 1.02 | vg6cn50 582.96 3.95 2.85 1.10 |
% diff V4 to V5 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 1% -1% -1% 0% | 0% 0% 0% 0% | 0% 0% 0% 0% | 0% 0% 0% 1% |
% diff V4 to V6 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 28% -22% -27% 0% | 3% -1% -2% 1% | 2% -1% -1% 1% | 2% -1% -1% 0% |
MTU 8992 runid V4 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans |
|
|
|
|
runid V5 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vg5cj01 168.00 6.09 5.07 1.02 | vg5cj10 442.18 4.72 3.80 0.91 | vg5cj20 501.64 3.95 3.08 0.87 | vg5cj50 574.22 3.83 2.86 0.97 |
runid V6 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vg6cj01 216.95 4.72 3.70 1.02 | vg6cj10 479.68 4.01 3.10 0.91 | vg6cj20 529.28 3.64 2.76 0.87 | vg6cj50 615.28 3.45 2.50 0.95 |
% diff V4 to V5 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 0% 0% 1% -1% | 1% 0% 0% 0% | 0% 0% 0% 0% | 1% 0% -1% 1% |
% diff V4 to V6 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 29% -22% -27% -1% | 9% -15% -18% 0% | 6% -8% -11% 1% | 8% -10% -13% -1% |
Note: 2064-109; z/VM 5.1.0; TCP/IP 510; V5 denotes IPv4 over IPv6 capable |
Number of clients | 1 | 10 | 20 | 50 |
MTU 1492 runid V4 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans |
|
|
|
|
runid V5 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vg5rn01 1341.23 1.08 0.69 0.39 | vg5rn10 2798.36 0.98 0.65 0.33 | vg5rn20 2797.05 0.97 0.65 0.31 | vg5rn50 2730.91 1.01 0.70 0.32 |
runid V6 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vg6rn01 1340.28 1.10 0.71 0.40 | vg6rn10 2770.68 0.99 0.66 0.33 | vg6rn20 2774.36 0.98 0.66 0.32 | vg6rn50 2715.20 1.02 0.70 0.32 |
% diff V4 to V5 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 0% -2% -1% -2% | 0% 0% 0% 0% | 0% 0% 0% 0% | 0% 0% 0% 0% |
% diff V4 to V6 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 0% 1% 1% -1% | -1% 1% 1% 0% | -1% 1% 1% 0% | -1% 0% 1% 0% |
MTU 8992 runid V4 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans |
|
|
|
|
runid V5 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vg5rj01 1340.51 1.09 0.69 0.39 | vg5rj10 2796.73 0.98 0.65 0.32 | vg5rj20 2789.85 0.97 0.66 0.32 | vg5rj50 2717.61 1.02 0.70 0.32 |
runid V6 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vg6rj01 1340.78 1.09 0.69 0.39 | vg6rj10 2793.82 0.98 0.65 0.33 | vg6rj20 2789.51 0.97 0.65 0.32 | vg6rj50 2726.55 1.01 0.70 0.32 |
% diff V4 to V5 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 0% 0% 0% -1% | 0% 0% 0% 0% | 0% 0% 0% 0% | 1% -1% -1% 0% |
% diff V4 to V6 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 0% 0% 0% -1% | 0% 0% 0% 0% | 0% 0% 0% 0% | 1% -1% -1% -1% |
Note: 2064-109; z/VM 5.1.0; TCP/IP 510; V5 denotes IPv4 over IPv6 capable |
Number of clients | 1 | 10 | 20 | 50 |
MTU 1492 runid V4 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB |
|
|
|
|
runid V5 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | vq5sn01 71.33 12.85 8.68 4.17 | vq5sn10 68.17 18.15 12.16 5.99 | vq5sn20 69.07 21.21 14.22 6.99 | vq5sn50 69.07 24.91 16.54 8.37 |
runid V6 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | vq6sn01 59.30 13.01 8.84 4.17 | vq6sn10 71.07 17.91 12.02 5.90 | vq6sn20 69.63 21.16 14.21 6.96 | vq6sn50 70.80 24.56 16.45 8.11 |
% diff V4 to V5 |
|
|
|
|
MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | -10% 10% 14% 3% | -19% 39% 45% 28% | -14% 40% 46% 30% | -14% 37% 41% 30% |
% diff V4 to V6 |
|
|
|
|
MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | -25% 11% 16% 3% | -16% 37% 43% 26% | -14% 40% 46% 30% | -12% 35% 40% 26% |
MTU 8992 runid V4 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB |
|
|
|
|
runid V5 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | vq5sj01 115.10 9.42 5.75 3.67 | vq5sj10 117.90 12.60 7.79 4.80 | vq5sj20 117.83 13.00 8.05 4.96 | vq5sj50 104.53 13.98 8.32 5.65 |
runid V6 MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | vq6sj01 115.03 9.43 5.77 3.66 | vq6sj10 117.70 12.29 7.58 4.72 | vq6sj20 117.57 12.54 7.74 4.80 | vq6sj50 106.07 13.51 7.92 5.59 |
% diff V4 to V5 |
|
|
|
|
MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | 0% 1% 2% 0% | 0% 1% 2% 0% | 0% 2% 3% 1% | -2% 4% 6% 1% |
% diff V4 to V6 |
|
|
|
|
MB/sec Total CPU msec/MB Emul CPU msec/MB CP CPU msec/MB | 0% 1% 2% 0% | 0% -1% -1% -2% | 0% -1% -1% -2% | 0% 1% 1% 0% |
Note: 2064-109; z/VM 5.1.0; TCP/IP 510; V5 denotes IPv4 over IPv6 capable |
Number of clients | 1 | 10 | 20 | 50 |
MTU 1492 runid V4 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans |
|
|
|
|
runid V5 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vq5cn01 141.88 2.04 1.45 0.58 | vq5cn10 434.73 2.00 1.43 0.58 | vq5cn20 522.55 1.98 1.41 0.57 | vq5cn50 675.16 2.04 1.47 0.57 |
runid V6 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vq6cn01 151.83 2.19 1.56 0.63 | vq6cn10 446.06 2.16 1.54 0.62 | vq6cn20 537.74 2.15 1.54 0.62 | vq6cn50 661.57 2.24 1.58 0.66 |
% diff V4 to V5 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | -4% 4% 3% 5% | -1% 4% 3% 6% | 1% 5% 4% 6% | 3% 3% 3% 2% |
% diff V4 to V6 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 3% 12% 11% 14% | 1% 12% 11% 14% | 4% 13% 13% 14% | 1% 13% 11% 18% |
MTU 8992 runid V4 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans |
|
|
|
|
runid V5 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vq5cj01 158.15 1.57 1.13 0.44 | vq5cj10 435.97 2.03 1.60 0.43 | vq5cj20 514.56 2.03 1.61 0.43 | vq5cj50 648.80 2.05 1.58 0.47 |
runid V6 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vq6cj01 179.05 1.71 1.22 0.50 | vq6cj10 460.01 1.65 1.25 0.40 | vq6cj20 539.64 1.60 1.19 0.41 | vq6cj50 725.19 1.90 1.45 0.44 |
% diff V4 to V5 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | -2% 1% 2% 0% | 0% 0% 0% 0% | -1% 0% 0% 0% | -5% 7% 7% 6% |
% diff V4 to V6 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 11% 10% 9% 12% | 5% -18% -22% -6% | 4% -21% -26% -5% | 6% -2% -2% 0% |
Note: 2064-109; z/VM 5.1.0; TCP/IP 510; V5 denotes IPv4 over IPv6 capable |
Number of clients | 1 | 10 | 20 | 50 |
MTU 1492 runid V4 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans |
|
|
|
|
runid V5 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vq5rn01 1011.29 0.51 0.35 0.16 | vq5rn10 2612.15 0.47 0.33 0.15 | vq5rn20 3098.32 0.48 0.33 0.15 | vq5rn50 3835.60 0.51 0.36 0.15 |
runid V6 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vq6rn01 1008.28 0.50 0.35 0.16 | vq6rn10 2604.14 0.47 0.33 0.15 | vq6rn20 2588.53 0.48 0.33 0.15 | vq6rn50 3886.89 0.51 0.36 0.15 |
% diff V4 to V5 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 2% -2% -1% -2% | 0% -1% -1% 0% | 0% 0% 0% 1% | 0% 0% 0% 1% |
% diff V4 to V6 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 2% -2% -1% -4% | 0% -1% -1% -2% | -16% 0% 0% 0% | 2% 0% 0% 0% |
MTU 8992 runid V4 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans |
|
|
|
|
runid V5 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vq5rj01 1013.60 0.50 0.34 0.15 | vq5rj10 2627.53 0.47 0.32 0.15 | vq5rj20 3113.63 0.48 0.33 0.15 | vq5rj50 3771.75 0.50 0.35 0.15 |
runid V6 trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | vq6rj01 1006.37 0.50 0.35 0.15 | vq6rj10 2544.41 0.50 0.33 0.15 | vq6rj20 2553.34 0.50 0.33 0.15 | vq6rj50 3733.04 0.50 0.36 0.15 |
% diff V4 to V5 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | 0% -1% -1% -1% | 0% 0% -1% 0% | 0% -1% -1% 0% | -2% 0% -1% 1% |
% diff V4 to V6 |
|
|
|
|
trans/sec Total CPU msec/trans Emul CPU msec/trans CP CPU msec/trans | -1% 0% 1% -1% | -3% 0% 0% 0% | -18% 0% -1% 0% | -3% 0% 0% 0% |
Note: 2064-109; z/VM 5.1.0; TCP/IP 510; V5 denotes IPv4 over IPv6 capable |