Migration from VTAM to Telnet
This section explores the performance implications of migrating end-user 3270 connectivity from VTAM to TCP/IP Telnet. It should be used in conjunction with a similar 9121-480 comparison that is summarized in "Migration from VTAM to Telnet" in the VM/ESA 2.3.0 Performance Report. The measurements shown here were obtained on a larger processor that supports many more users.
Measurements were obtained by running the FS8F0R workload on a 9121-742 processor with the end users simulated by TPNS running on a separate system. VM/ESA 2.3.0 was used for all measurements. For the base measurement, connectivity was provided by VTAM 3.4.1 through a CTCA connection with the TPNS system. Table 1 compares this VTAM base measurement to a measurement using TCP/IP 310 Telnet through a 3172-3 Interconnect Controller and a 16Mbit IBM Token Ring.
Processor model: 9121-742 Processors used: 4 Storage: Real: 1024MB (default MDC) Expanded: 1024MB (MDC BIAS 0.1) Tape: 3480 (Monitor) DASD:
Type of DASD | Control Unit | Number of Paths |
|
| - Number of Volumes - TDSK |
|
|
|
---|---|---|---|---|---|---|---|---|
3390-3 | RAMAC 2 | 4 | | 6 | | 32 | | |
3390-2 | 3990-3 | 4 | 6 | | 2 | | | 2 |
3390-2 | 3990-2 | 4 | 6 | | 2 | | | |
3390-2 | 3990-2 | 4 | 10 | | | | |
Note: RAMAC 2 refers to the RAMAC 2 Array Subsystem with 256MB cache and drawers in 3390-3 format.
Communications (CTCA):
|
| Lines per Control Unit |
|
---|---|---|---|
3088 | 1 | NA | 4.5MB |
Communications (Token Ring): 16 Mbit IBM Token Ring 3172-3 Interconnect Controller
Driver: TPNS Think time distribution: Bactrian CMS block size: 4KB Virtual Machines:
Virtual Machine |
|
| Machine Size/Mode |
|
|
|
---|---|---|---|---|---|---|
VSCSn | 3 | VSCS | 64MB/XA | 10000 | 1200 | QUICKDSP ON |
VTAMXA or | 1 | VTAM/VSCS | 64MB/XA | 10000 | 550 | QUICKDSP ON |
| | | | | | |
TCPIP | 1 | TCP/IP | 256MB/XA | 10000 | 2700 | QUICKDSP ON |
| | | | | | |
SMART | 1 | RTM | 32MB/XA | 3% | 500 | QUICKDSP ON |
WRITER | 1 | CP monitor | 2MB/XA | 100 | | QUICKDSP ON |
Unnnn | 5100 | Users | 3MB/XC | 100 | |
The results demonstrate that TCP/IP VM Telnet connectivity can support large numbers of users (5100) with good response time (0.36 seconds).
VTAM supports the 3270 interface through the *CCS CP system service (accessed using IUCV requests), while Telnet provides this function through use of the Diagnose X'7C' logical device support facility. This difference is reflected in the results as a large decrease in PRIVOP/CMD and a large increase in DIAG/CMD. The fact that diagnose X'7C' has a longer pathlength than *CCS accounts for much of the CPU usage increase observed in the TCP/IP measurement relative to the VTAM base measurement. Another contributing factor is that TCP/IP does more communication I/Os than VTAM, as shown by the increase in DIAG 98/CMD.
The 3.9% increase in total processing requirements (PBT/CMD (H)) is much less than the 8.2% increase observed for the 9121-480 configuration (see "Migration from VTAM to Telnet" in the VM/ESA 2.3.0 Performance Report). This is because the larger 9121-742 configuration was configured with VSCS in 3 separate virtual machines, whereas the 9121-480 configuration was small enough that VSCS could be configured within the VTAM virtual machine. The external VSCS configuration is less efficient, raising the total VTAM processing requirements and reducing the difference between the VTAM and TCP/IP results.
Note the increase in master processor utilization (MASTER TOTAL (H)) that occurred when going from VTAM to TCP/IP. This is due to an increase in CP master processor utilization (MASTER CP (H)). This occured because most of the CP modules that implement Diagnose X'7C' obtain their MP serialization by running on the master processor. This increase in master processor contention caused the response time increase relative to the VTAM base case to be larger than it otherwise would have been.
Table 1. Migration from VTAM to TCP/IP
Communications Interconnection Run ID | VTAM 3.4.1 CTCA S4AE5101 | TCP/IP 310 3172-3/TR S4AE5100 |
|
|
---|---|---|---|---|
Response Time TRIV INT NONTRIV INT TOT INT TOT INT ADJ AVG FIRST (T) AVG LAST (T) |
|
|
|
|
Throughput AVG THINK (T) ETR ETR (T) ETR RATIO ITR (H) ITR EMUL ITR ITRR (H) ITRR |
|
|
|
|
Proc. Usage PBT/CMD (H) PBT/CMD CP/CMD (H) CP/CMD EMUL/CMD (H) EMUL/CMD |
|
|
|
|
Processor Util. TOTAL (H) TOTAL UTIL/PROC (H) UTIL/PROC TOTAL EMUL (H) TOTAL EMUL MASTER TOTAL (H) MASTER TOTAL MASTER EMUL (H) MASTER CP (H) MASTER EMUL TVR(H) TVR |
|
|
|
|
Storage NUCLEUS SIZE (V) TRACE TABLE (V) WKSET (V) PGBLPGS PGBLPGS/USER TOT PAGES/USER (V) FREEPGS FREE UTIL SHRPGS |
|
|
|
|
Paging READS/SEC WRITES/SEC PAGE/CMD PAGE IO RATE (V) PAGE IO/CMD (V) XSTOR IN/SEC XSTOR OUT/SEC XSTOR/CMD FAST CLR/CMD |
|
|
|
|
Queues DISPATCH LIST ELIGIBLE LIST |
|
|
|
|
I/O VIO RATE VIO/CMD RIO RATE (V) RIO/CMD (V) NONPAGE RIO/CMD (V) DASD RESP TIME (V) MDC REAL SIZE (MB) MDC XSTOR SIZE (MB) MDC READS (I/Os) MDC WRITES (I/Os) MDC AVOID MDC HIT RATIO |
|
|
|
|
PRIVOPs PRIVOP/CMD DIAG/CMD DIAG 04/CMD DIAG 08/CMD DIAG 0C/CMD DIAG 14/CMD DIAG 58/CMD DIAG 7C/CMD DIAG 98/CMD DIAG A4/CMD DIAG A8/CMD DIAG 214/CMD DIAG 270/CMD SIE/CMD SIE INTCPT/CMD FREE TOTL/CMD |
|
|
|
|
VTAM or TCPIP Machines WKSET (V) TOT CPU/CMD (V) CP CPU/CMD (V) VIRT CPU/CMD (V) DIAG 98/CMD (V) |
|
|
|
|
Note: 9121-742, 4 processors; 1024 MB real storage, 1024 MB expanded storage; VM/ESA 2.3.0; 5100 users; FS8F0R workload; T=TPNS, V=VMPRF, H=Hardware Monitor, Unmarked=RTM |