1. Depending on the type of hardware you have, you may need to refer to the following topics:
v Creating a network server description
v Creating a network interface description
2. You define lines by creating line descriptions. Depending on your hardware, the lines may be attached
to a network server, or a network interface.
Creating a network interface description
Network interface descriptions for asynchronous transfer mode (ATM), frame relay, and integrated services
digital network (ISDN) protocols describe the communications interface.
To create a network interface description, do the following:
1. Type one of these commands on any AS/400 command line for the type of network interface you are
creating and press F4:
v Create Network Interface (ATM) (CRTNWIATM)
v Create Network Interface (Frame Relay Network) (CRTNWIFR)
v Create Network Interface Description for ISDN (CRTNWIISDN)
2. Use the on-line help information to choose the correct parameter values.
3. Press Enter. The network interface description is created.
Creating a network server description
A network server description describes which Integrated PC Server the local area network (LAN) and the
application will be using.
To create a network server description, do the following:
1. Type the Create Network Server Description (CRTNWSD) command on any AS/400 command line and
press F4.
2. Use the on-line help information to choose the parameter settings.
3. Press Enter. The network server description is created.
Creating a line description
You create line descriptions to describe the physical line connection and the data link protocol to be used
between the AS/400 system and the network.
To create line descriptions, do the following:
1. Type one of these commands on any AS/400 command line to define the type of line you are creating
and press F4.
v Create Line Description (Ethernet) (CRTLINETH)
v Create Line Description (Distributed Data Interface (DDI)) (CRTLINDDI)
v Create Line Description (Frame Relay) (CRTLINFR)
v Create Line Description for (IDLC) (CRTLINIDLC)
v Create Line Description (Synchronous Data Link Control (SDLC)) (CRTLINSDLC)
v Create Line Description (Token-ring) (CRTLINTRN)
v Create Line Description (Wireless) (CRTLINWLS)
v Create Line Description (X.25) (CRTLINX25)
2. Use the online help information to choose the correct parameter values.
Many factors can affect the performance of AS/400 application programs. To achieve the best performance
with your particular communications environment, you may want to review these topics:
v Improving wide area network (WAN) performance.
v Improving local area network (LAN) performance.
v Improving data path performance.
Improving wide area network performance
To achieve better performance with your AS/400 when communicating in a wide area network (WAN), you
need to consider the following:
v “Adjusting WAN protocols for optimum AS/400 performance”
v “Adjusting the WAN line speed for optimum AS/400 performance”
v “WAN line speed considerations for IOPs” on page 8
Adjusting WAN protocols for optimum AS/400 performance
Wide area network (WAN) protocols affect the communications performance on AS/400. Let us use X.25
for our example. For each X.25 communications controller, the AS/400 has some processing limitation for
the line, the line speed, and the total number of virtual circuits that can be used. Performance degradation
can be reduced by observing these limitations.
To optimize AS/400 performance for wide area networks, perform these tasks:
v Reduce the total number of frames by using larger frames.
v To take advantage of these large frame sizes, change the MAXFRAME parameter on the line
description (LIND) to reflect the maximum value. For X.25, increase the DFTPKTSIZE and MAXFRAME
parameters to their maximum value.
v Configure a WAN line as full-duplex to provide you with a higher throughput for applications that can
take advantage of this mode. This can also provide higher throughput for multiple users.
v Increase frame relay to capacity.
The data rate for a given protocol may increase as frame size increases. Under these circumstances, the
central processing unit (CPU) and the input/output processor (IOP) do not do as much processing. Fewer
and larger frames also make more efficient use of the communications line (higher effective data rate)
because of fewer overhead bytes and line turn-arounds.
Frame relay has equivalent performance over RS449, X.21, and V.35 assuming equal line speeds and
conditions. Frame relay performance (CPU time) is similar to or slightly better than Synchronous Data Link
Control. For properly tuned large transfer applications, the CPU and IOP have no problem using the line
speed to capacity.
For information about configuring AS/400 communications, see the Communications Configuration
book.
Adjusting the WAN line speed for optimum AS/400 performance
In many cases, the communications line is the largest contributor to overall response time in the wide area
network (WAN). Therefore, you should closely plan and manage its performance. In general, having the
appropriate line speed is the key consideration for gaining the best performance.
To adjust the line speed for your wide area network, perform these tasks:
v Check the difference in performance between half-duplex utilization and full-duplex utilization on the line
v For interactive environments, keep line use below 30% to maintain predictable and consistent response
times. Exceeding 50% line use usually slows down response time. The line use can be measured with
the AS/400 performance tools.
v For large transfer environments, or for environments in which only a small number of users are sharing
a line, increase line use to allow for acceptable response times.
v The CPU usage for fractional T1 support and other high-speed WAN connections is similar to any other
line that runs the same type of work. As the speed of a line increases from a traditional low speed to a
high-speed or full T1/E1/J1 speed, performance characteristics may change as follows:
– With interactive transactions, performance may be slightly faster.
– With a large transfer, performance may be significantly faster.
– With a single job, performance may be too serialized to use the entire bandwidth.
– With high throughput, performance is more sensitive to frame size.
– With high throughput, performance is more sensitive to application efficiency.
– With synchronous data link control (SDLC), the communications controller CPU usage increases
because of polling.
Additional considerations for adjusting the wide area network line speed are the following:
v A common misconception about the line speed of each attached communications line is that central
processing unit (CPU) resource is used in a uniform fashion. Exact statements cannot be made about
the number of lines that any given AS/400 model can support.
v Most communications applications use a lot of CPU resource (to process data, to support disk input and
output) and communications line resource (to send and receive data or display I/O). The amount of line
resource that is used is proportional to the total number of bytes that are sent or received on the line.
Some additional CPU resource is used to process the communications software to support the
individual sends (puts or writes) and receives (gets or reads). Communications input/output processor
resource is also used to support the line activity.
v When a single job is running disk operations or doing non-overlapped CPU processing, the
communications link is idle. If several sessions transfer concurrently, then the jobs are more interleaved
and make better use of the communications link.
v Polling is an important consideration for synchronous data link control (SDLC) environments. All SDLC
polling is handled by the communications controller and is governed by parameters in both the line and
controller descriptions.
v For information about AS/400 configuration, see the Communications Configuration
v For more information about performance tools, see the Performance Tools for AS/400 book.
book.
WAN line speed considerations for IOPs
When configuring a communications controller, you should consider both subsystem storage and
aggregate line speed. Subsystem storage is the amount of storage available on the communications
controller. Aggregate line speed is the sum of individual lines speeds that are attached to the
communications controller.
The following information can help you understand network line speed considerations for input/output
processors (IOPs).
v For interactive environments, you should not exceed 60% use on the communications IOP. Exceeding
this threshold in a large transfer environment or with a small number of concurrent users may still offer
acceptable performance. Use the AS/400 performance tools to get the utilization.
v You can attach multiple IOPs to an AS/400 system. The maximum number of IOPs that can be attached
is determined by the AS/400 model. It is important to distribute the work load across several IOPs if the
performance capabilities of a single IOP is exceeded.
v Even though an IOP can support certain configurations, a given AS/400 model may not have enough
system resource (for example, CPU processing capacity) to support the work load over the lines.
v The use of larger frames generally improves large transfer performance in terms of capacity for the
communications IOP and in terms of system response time. The amount of time that the IOP spends
8Version 5
processing a larger frame is only slightly more than the amount needed to process a smaller frame. If
you use larger frames to transfer a single system message or block of data, decreases the total number
of frames required to complete the transfer.
v The values for IOP use in synchronous data link control (SDLC) environments do not necessarily
increase consistently with the number of work stations or with the workload. An IOP can spend more
time polling when the application is not using the line. It is possible to see a relatively high IOP use at
low throughput levels.
v For information on AS/400 configuration, see the Communications Configuration
v For more information on performance tools, see the Performance Tools for AS/400
book.
book.
Improving local area network performance
To achieve better performance with your AS/400 when communicating in a local area network (LAN), you
need to consider the following.
v “Adjusting LANs for optimum communications performance”
v “Adjusting LAN lines for optimum communications performance”
v “LAN line speed considerations for IOPs” on page 10
Adjusting LANs for optimum communications performance
Local area networks (LAN) affect the communications performance on AS/400. Improvements to LAN
input/output (IOPs) in the areas of increased central processing unit (CPU) time, IOP capacity, and support
of IOP assist make them more efficient. This efficiency allows advanced program-to-program
communications (APPC) to send request units to the IOP, passing the processing cost of processing frame
to the IOP.
The following information can help you understand the protocol considerations for local area networks.
v A Data Link Control (DLC) can achieve a significantly higher data rate than other supported line types.
This is due to the desirable combination of having a high media speed along with large frame sizes.
v When several sessions use a line or LAN concurrently, the aggregate data rate may be higher than
when only one session is used.
v To achieve good performance in a multi-user interactive LAN environment, you should manage the
number of active users so that LAN media use does not exceed 50%. (A 25% utilization is
recommended for Ethernet environments because of media collisions that causes the program to loop).
Operating at higher utilization may decrease response time because of excess queueing time for the
line. In a large transfer environment in which a small number of users contend for the line, a higher line
use may still offer acceptable performance.
For more information about AS/400 configuration, see the Communications Configuration
book.
Adjusting LAN lines for optimum communications performance
Several parameters in the line description (LIND) and the controller description (CTLD) play an important
role in system performance that you can change.
The following information can help you to understand the line considerations for local area networks.
v MAXFRAME on the line description (LIND) and the controller description (CTLD): Maximizing the frame
size in a LAN environment supplies the best performance for large transfers. A large frame size does
not negatively affect performance for small transfers. Configure both the AS/400 system and the other
link station for large frames. Otherwise, of the two maximum frame size values, the smaller is used
when you transfer data. Bridges may also limit the maximum frame size. You should change the default
value from 1994 to a larger size.
v LANMAXOUT on the CTLD (for advanced program-to-program communications (APPC) environments):
This parameter governs how often the sending system waits for an acknowledgment. The LANACKFRQ
Chapter 3. Optimizing communications performance9
parameter value on one system should never have a greater value than the LANMAXOUT parameter
value on the other system. The parameter values of the sending system should match the values on the
receiving system.
v Setting appropriate values for the LANMAXOUT parameter along with the LAN acknowledgment
frequency (LANACKFRQ) parameter for both the sending stations and receiving stations is essential for
optimal performance. Other values may decrease throughput by 50% or even more if conditions trigger
time-outs.
v LANWDWSTP for advanced program-to-program communications (APPC) on the controller description
(CTLD): If there are network congestion or overruns to certain target system adapters, then increasing
the value from the default of *NONE to 2 or more may improve performance.
In general, setting the LANMAXOUT parameter value to *CALC or 2 offers the best performance for
interactive environments and adequate performance for larger transfer environments.
v For large transfer environments, changing the LANMAXOUT value may significantly increase
performance. As starting points, use the following guidelines:
– When you are communicating with a recent model personal computer, increase the LANMAXOUT
parameter, but keep the LANACKFRQ parameter set to *CALC. For older models of personal
computers, use *CALC for both values to limit buffer overruns.
– If LANACKFRQ and LANMAXOUT parameter values are changed without noticeable performance
improvements, change the values back to *CALC.
For more information on AS/400 communications, see the Communications Configuration
book.
LAN line speed considerations for IOPs
When configuring an AS/400 system with communications lines and local area networks (LANs), you
should not overload an input/output processor (IOP) to prevent possible system performance bottlenecks.
The following information can help you to understand the line speed considerations for IOPs.
v The integrated PC server performance is similar to the 2619 and the 2617 IOPs for host LAN functions.
For send and receive scenarios, performance is equivalent. For large transfers, the 6506 IOP is slightly
faster than the 2619 TRLAN IOP, but slightly slower than the 2617 Ethernet IOP. These differences are
not significant enough to choose one over the other.
v The 100 Mbps Ethernet support provides the best LAN performance. The IOP can be optimally
configured to have an aggregate transfer rate of 27 Mbps. Multiple concurrent large transfers may be
required to drive at that rate.
v When analyzing communications performance that includes the 2619 TRLAN IOP, you should be aware
that resources other than the IOP use can become the bottleneck.
v You should have the highest capacity IOP available for file serving. You should have the highest
capacity IOP available for environments that use many communications input and output operations for
each transaction. The highest capacity IOP also minimizes the overall response time.
See the following references for more detail:
vFor more information about AS/400 communications, see the Communications Configuration book.
v For more information on IOP performance, see the Performance Tools for AS/400 book.
Improving data path performance
To assess the performance of your data path, you may want to review the following topics:
v Considerations for subsystem configuration for error recovery performance
v Communications performance considerations for interactive jobs
v Consider communications performance for batch jobs
10Version 5
v Mix interactive and batch jobs on a wide area network line
v Performance considerations for AnyNet communications
v Subsystems
Considerations for subsystem configuration for error recovery
performance
Each piece of work that runs on the AS/400 system is called a job. Each job is a single, identifiable
sequence of processing actions that represents a single use of the system. The basic types of jobs
performed are interactive jobs, batch jobs, spooling jobs, autostart jobs, and prestart jobs.
Jobs that run in subsystems do all work that is performed on the AS/400. As the number of users on the
system increases, it becomes important for you to consider how the communications and interactive
subsystems should be configured.
The configuration of subsystems has little impact in normal data path operations. However, multiple
subsystems can provide multiple processes to do cleanup and recovery when error conditions occur. This
can result in improved performance.
As the number of users on the system increases, you must consider the importance of how subsystems
are configured:
v Consider limiting the number of devices that are serviced by a single subsystem. Between 200 and 300
devices for each subsystem are recommended. Use the following recommendations to divide these
users:
– The number of users in any given subsystem
– The connectivity used to access the system
– The type of work the users do
– The geographic location of the users
v Create additional communications and interactive subsystems to split the work into multiple subsystems.
v The work that is performed in the QCMN subsystem is for connecting and disconnecting from the
system. Error recovery considerations are important in the configuration of the communications
subsystem.
v To prevent a subsystem from ever allocating a device, ensure that there are no workstation or type
entries for the devices that you do what allocated by that subsystem.
v Only use the AT(*ENTER) option if you must allow jobs to transfer into that subsystem.
v For each subsystem you have defined, you need to identify which users will run in which subsystems.
Use the Add Work Station Entry (ADDWSE) command and the Remove Work Station Entry (RMVWSE)
command. You can set up work stations entries that identify which devices that subsystem should
allocate, as well as which devices a subsystem should not allocate.
Note: You can use the ADDWSE commands while the subsystem is active. However, subsystems do not
reallocate device locks dynamically. Eventually, it may be necessary to end and restart the
subsystems to have the device locks allocated to the desired subsystem.
To specify the devices a communications subsystem should allocate:
3. If desired, update your system startup program to start your new subsystems automatically.
Communications performance considerations for interactive jobs
An interactive job is one that uses a keyboard and character-type display. If a job needs the user to type
on the keyboard and display character results, that job is probably considered interactive. Interactive in this
sense means that the job and the user depend on each other to get the work done.
To optimize communications performance for interactive jobs, consider the following:
v Attach work stations through communications requires more CPU overhead than 5250 local
workstations.
v Use a twinaxial controller to provide better performance than an American National Standard Code for
Information Interchange (ASCII) controller.
v Keep the line utilization below 30 percent for best performance when interactive users are attached.
This will maintain predictable and consistent response times. Exceeding 50 to 60 percent line utilization
will usually cause unacceptable response times.
If your system has interactive users who are connected many different ways, you should consider
configuring your interactive subsystems to separate the users. Local workstation, remote workstations,
5250 display station pass-through, or Telnet are some examples of these types of connections that should
be separated. When you configure interactive subsystems, identify how you want the interactive users to
be separated and create the appropriate subsystem descriptions.
During error recovery, when many users risk losing their sessions at one time, an interactive subsystem
can be very busy performing device recovery. This device recovery can adversely affect the work of other
users in the subsystem who would otherwise be unaffected by the failure. Therefore, you may need to
change how the interactive subsystems are configured. However, multiple subsystems can provide multiple
processes to do cleanup and recovery when error conditions occur. This can result in improved
performance.
The example below shows how to configure an interactive subsystem to allocate devices that begin with
devname* and present a signon display on those display devices:
Use the following example to configure an interactive subsystem so that the device name devname* is not
allocated and a signon display does not appear.
Adding workstation entries with AT(*ENTER) allows you to use the Transfer Job (TFRJOB) function into
that subsystem. If the TFRJOB function is not required or necessary, there is no need to add the
workstation entries with AT(*ENTER).
To specify the devices an interactive subsystem should allocate when the subsystem is started:
6. If desired, update your system startup program to start your new subsystems automatically.
Communications performance considerations for batch jobs
Each piece of work run on the AS/400 system is called a job. Each job is a single, identifiable sequence of
processing actions that represents a single use of the system. The basic types of jobs that are performed
are interactive jobs, batch jobs, spooling jobs, autostart jobs, and prestart jobs.
Batch jobs are predefined groups of processing actions that are submitted to the system to be performed
with little or no interaction between the user and the system. Batch jobs can be tuned for optimized
performance.
To optimize batch jobs for communications, consider the following:
v Break the application into pieces and having multiple batch threads (jobs) operate concurrently.
v Reduce the number of open and close operations, input and output operations.
v If you have a considerable amount of main storage available, consider using the Set Object Access
(SETOBJACC) command. This command preloads the complete database file, database index, or
program into the assigned main storage pool if sufficient storage is available. The objective is to
improve performance by eliminating disk-read/write operations.
v Try to limit the number of communications input and output operations by doing fewer (and perhaps
larger) application sends and receives when communications lines are used.
v Block the data in the application. Try to place the application on the same system as the frequently
accessed data.
For more information about batch job performance, see the Communications Management
book.
Mixing interactive and batch jobs on a WAN line
When interactive users and large transfers are running on a communications line concurrently, you may
need to change configuration parameters. You should be able to configure AS/400 communications to work
with interactive and batch jobs.
To mix interactive and batch jobs on a wide area network (WAN) line, consider the following to keep
interactive performance acceptable:
v Use Advanced Peer-to-Peer Networking (APPN) transmission priority to prioritize the interactive user’s
transfer over that of the large transfer. This is the preferred method to transfer batch and interactive
jobs.
v Change the request/response unit size to a lower value for the large transfer. This parameter setting
optimizes response time at the expense of large transfer performance.
v Reduce the pacing values for the large transfer to slow it down, which allows the interactive users more
windows for getting on the line.
Note: The overall central processing unit time increases for the large transfer.
For more information about AS/400 communications, see the Communications Configuration
book.
Performance considerations for AnyNet communications
AnyNet communications is a good performance factor for you to consider. It is more expensive to use than
any of the OS/400 protocols because you spend twice as much to run two protocols.
To optimize AnyNet performance, consider the following:
v For send and receive pairs, the most efficient use of an interface is with its own protocol stack. That is,
intersystem communications function (ICF) and common programming interface communications (CPI
Communications) perform the best with advanced program-to-program communications (APPC). There
is additional CPU time when the crossover between the protocols processes.
v Each communications interface performs differently depending on the scenario. ICF and CPI
Communications perform the best with APPC.
Note: An alternative to AnyNet communications is to have SNA and TCP/IP running parallel or over the
same lines in your network. Hence, performance implications can be surpassed by not using
AnyNet.
For more information about AnyNet/400 sockets, see the book Sockets Programming
Setting up the AnyNet environment
AnyNet/400 is an AnyNet family product. These products allow you to use application programs that are
written for a certain communications protocol but also run over non-native communications protocols
without changing (or even re-compiling) the application program. The choice of the destination address
controls whether the request is sent over the native protocols or through the AnyNet code and on to a
non-native protocol.
.
To configure Transmission Control Protocol/Internet Protocol (TCP/IP) over advanced program-to-program
communications (APPC), you need to take two basic actions:
1. Identify the set of IP addresses to route over the SNA network.
2. Tell the system how to convert the IP address to the SNA format.
For more information about APPC Over TCP/IP Configuration, see the APPC Programming book.
For more information about IPX Support, see the Internetwork Packet Exchange (IPX) Support.
For related information about AnyNet, see:
“AnyNet communications for the AS/400 system”
“Performance considerations for AnyNet communications”
AnyNet communications for the AS/400 system
AnyNet is an IBM implementation of the Multiprotocol Transport Networking (MPTN) architecture, such as
AnyNet/2 and AnyNet/Multiple Virtual Storage (MVS). AnyNet capability allows applications and associated
services that use application programming interfaces, such as sockets, intersystem communications
function (ICF), or CPI Communications, the flexibility to use alternative network protocols, such as
Systems Network Architecture (SNA) or TCP/IP. AnyNet is a family of products that allow applications that
14Version 5
are written for one type of network protocol to run over a different type of network protocol. For example,
without AnyNet, your choice of application program interface (API) dictates your choice of network
protocol, or your choice of network protocol dictates your choice of APIs.
AnyNet allows you to mix and match applications with network protocols. In fact, you can do this without
changing your application programs. Your destination address (such as a remote location) determines the
type of network protocol to use.
v AnyNet/400 Sockets
This support converts TCP/IP addresses to SNA addresses that are based on tables that are configured
by the network administrator. Programs supported include File Transfer Protocol (FTP), Simple Mail
Transfer Protocol (SMTP), Simple Network Management Protocol (SNMP), PING, and user-written
sockets programs.
– TCP/IP over SNA
– TCP/IP over IPX
v AnyNet/400 APPC (advanced program-to-program communications)
This support allows programs that are written to traditional APPC APIs (such as ICF,
CPI-Communications, and CICS/400) to be run over non-APPC networks. The application program uses
Location names to specify the source and destination address. A TCP/IP domain name server converts
these location names to IP addresses. Programs supported include distributed data management
(DDM), Distributed Relational Database Architecture (DRDA), SNA distribution services (SNADS),
display station pass-through, Client Access, user-written CPI-Communications programs, and
user-written ICF programs .
– APPC over TCP/IP
– APPC over IPX
For more information about using both AnyNet and nonAnyNet sockets, see the Sockets Programming
book.
Subsystems
A subsystem is a single, predefined operating environment through which the system coordinates work
flow and resource usage. OS/400 can contain several that are independent operating subsystems. The
run-time characteristics of a subsystem are defined in an object that is called a subsystem description.
IBM supplies several subsystem descriptions that can be used with or without modification:
QINTER
Used for interactive jobs
QBATCH
Used for batch jobs
QBASE
Used for both interactive and communications batch jobs
QCMN
Used for communications batch jobs
QSERVER
File server system
QSYSWRK
Used for general system work
QUSRWRK
This is the default subsystem used for TCP/IP Client Access Host Servers that used to run in
QSYSWRK.
A new subsystem can also be defined with the Create Subsystem Description (CRTSBSD) command.
For more information about creating subsystems, see the Work Managementbook.
16Version 5
Chapter 4. Communications applications
Communications applications that are used in an APPC (advanced program-to-program) environment are
also available to be used in an APPN and HPR environment; only the method by which data is transported
is changed. APPC delivers the data from applications higher in the SNA layers down to APPN for
transportation through the network. User-written APPC applications and distributed data management
(DDM) are fully supported in an APPN and HPR environment. The topic, Application programming
interface (API) performance considerations gives a more complete discussion of APPC applications.
When you encounter problems that indicate that the route to the remote location cannot be found, you can
attempt to make the connection again with the Start Pass-Through (STRPASTHR) command. See the
topic, Solving remote communication problems using STRPASTHR for more information.
For information on Connecting Windows 95/NT Clients to your AS/400, see Client Access.
User written APPC applications
APPN performs many functions in a communications environment. Therefore, it is important to consider
time-out parameters in APPC programs which use ICF. In particular, it may be important to increase the
WAITFILE parameter for these applications so that they do not time-out while waiting for APPN functions
to be performed.
APPN function is transparent to APPC programs using APPN take advantage of the following routing
functions:
v Non-adjacent nodes appear adjacent and so APPC programs may communicate directly to programs in
non-adjacent nodes (without any APPC programs on the intermediate nodes).
v Performance is improved for APPC programs with session endpoints that are not physically adjacent in
the network.
v APPC programs may communicate directly to programs in nodes in an adjacent APPN network through
network nodes.
Distributed data management (DDM)
DDM is a function of the operating system that allows an application program or user on one system to
use database files stored on remote systems. The systems must be connected by a communications
network, and the remote systems must also be using DDM.
DDM on the AS/400 allows application programs or users to:
v Access data files that reside on remote systems (target systems). The remote systems can also access
data files on the local AS/400 system.
v An application can add, change, and delete data records in a file that exist on a target system.
v Create, delete, or rename files on a remote system.
v Copy a file from one system to another.
When DDM is in use, neither the application program nor the program user needs to know if the file that is
needed exists locally or on a remote system. Remote and local file processing are essentially handled the
same way.
Application program interface (API) performance considerations
To achieve better performance with your AS/400, you need to consider the application programming
interface (API) available on the AS/400. To optimize APPC performance, consider the following:
v Using larger sends for a given large transfer (record sizes) provides a higher application data rate and
decreases CPU time. With the larger record size, the CPU has less processing to do because there are
fewer application reads and writes to transfer the same amount of data.
v If a value of *CALC is selected for maximum Systems Network Architecture (SNA) request/response unit
(RU), the system selects an efficient size compatible with the frame size. The frame size is on the line
description that you choose. Changing the RU size to a value other than *CALC may negate the
performance feature.
v Compression with APPC should be used with caution and only for slower speed wide area network
(WAN) environments. Many suggest that compression should be used with speeds 19.2 kbps and
slower.
v If you are doing tasks that include repetitive, small puts; better performance is achieved if you use ICF
or CPI Communications.
See the following topics for a more complete discussion of APPC applications:
v Performance considerations for Intersystem Communications Function
v Performance considerations for Common Programming Interface communications
For information about AS/400 communications, see the Communications Configuration
For more information about CICS/400, see the CICS/400 Administration and Operations Guide
book.
.
Performance considerations for intersystem communications function
You can use intersystem communications function (ICF) to write application programs that you want to
communicate with advanced program-to-program communications (APPC). ICF also provides
program-to-device communications between the AS/400 system and hardware devices. You must
determine which system is to send data first before you write the program. ICF data management handles
the communication functions and the data for your program. In particular, ICF should be used to do tasks
that include repetitive, small inputs.
To optimize ICF performance, consider the following:
v Eliminate unused record formats.
v Use separate record formats instead of multipurpose record formats with option indicators.
v Code to use the same record format for repeated operations.
v Set the maximum program devices equal to 1.
v Use a nonshared file.
v Use a separate indicator area.
v The use of the ICF keywords force data and confirm should be minimized.
v Use the Request to Send keyword only when necessary.
v Use the Invite Only keyword when soliciting input from multiple devices, otherwise use the Read
keyword instead.
v If using the Invite keyword to solicit from multiple program devices, follow it with a Read-from-invited
operation, not a Read operation.
To create device descriptions to get your system set up for ICF, do the following:
1. Type the appropriate Create Device Description commands on the AS/400 command line and press
F4.
2. Use the online help information to choose the parameter values.
3. Press Enter. The device description is created.
For more information about ICF, see
v “Application program interface (API) performance considerations” on page 17
v ICF Programming
18Version 5
Performance considerations for Common Programming Interface
Communications
You can use Common Programming Interface Communications (CPI Communications) to write application
programs that you want to communicate with advanced program-to-communications (APPC). The interface
makes use of the System Network Architecture (SNA) LU (logical unit) 6.2 architecture to do the following:
v Establish a conversation
v Send and receive data
v Exchange control information
v End a conversation
v Notify a partner program of errors.
Intersystem communications feature (ICF) and CPI Communications programs have similar performances
for small data transfers.
To optimize CPI Communications application programs, do the following:
v Minimize the use of flush and confirm.
v Receive a compile record and parse it in your buffer.
v Do not use multiple receive calls to receive a single record.
v Use Request-to-Send only when necessary.
To add or change communications entries to get the system set up for CPI Communications, do the
following:
1. Type appropriate command on the AS/400 command line and press F4.
v Add Communications Entry (ADDCMNE)
v Remove Communications Entry (RMVCMNE)
v Change Communications Entry (CHGCMNE)
2. Use the online help information to change, add, or remove parameter values.
3. Press Enter. The communications entries are added, changed or removed.
For more information about configuring CPI Communications, see:
v “Application program interface (API) performance considerations” on page 17
v CICS/400 Administration and Operations Guide
Chapter 4. Communications applications19
20Version 5
Chapter 5. Communicating with host systems
You can configure the AS/400 system to communicate with a host system by matching AS/400
parameters.
Another option for AS/400 users is Dependent LU Requester Support (DLUR). DLUR allows dependent
secondary logical units (LU 0, 1, 2, and 3) an entry point into the APPN network. DLUR support gives the
appearance of having an adjacent connection to VTAM, but allows traversing the APPN network through
intermediate nodes. To configure DLUR, see the page Configuring Dependent LU Requester (DLUR).
Matching AS/400 parameters for a host system
You can configure the AS/400 system to communicate with a host system. This configuration requires the
coordination of parameters and values. The list contains only those configuration prompts and parameters
that require coordination on both the AS/400 and the host system. In addition, some of the parameters that
are listed may not apply to your particular configuration.
For examples of connecting an AS/400 to a host system, see “Examples: Connecting AS/400 to a host
system” on page 26.
For information about configuring host systems, see the manuals VTAM Installation and ResourceDefinition, SC23-0111, and Network Control Program Resource Definition Reference, SC30-3254.
v “Matching AS/400 line description parameters for a host system”
v “Matching AS/400 controller description parameters for a host system” on page 23
v “Matching AS/400 device description parameters for a host system” on page 24
v “Matching AS/400 mode and class-of-service description parameters for a host system” on page 25
v For more information on AS/400 parameters, see Communications Configuration
.
Matching AS/400 line description parameters for a host system
You must match host system communications configuration parameters with AS/400 values. A description
of these AS/400 values are in the following table. For information about configuring host systems, see the
manuals VTAM Installation and Resource Definition, SC23-0111, and Network Control Program ResourceDefinition Reference, SC30-3254.
You can specify some host system parameters on multiple definition statements, such as the GROUP,
LINE, PU, and LU. The following table lists only the lowest level definition statement that is used by the
host system.
To configure an AS/400 to a host system:
v See “Examples: Connecting AS/400 to a host system” on page 26 for an example of connecting an
AS/400 to a host system.
v Use the following table for the line description parameter.
Host DIALNO parameter is a concatenation of:
SSAP/DSAP/remote-adapter-address.
AS/400 CRTLINTRN command ADPTADR value must
match the remote-adapter-address portion of the host
DIALNO parameter. The DSAP portion of the DIALNO
parameter must correspond to the SSAP value
specified on the AS/400 controller description.
PUMACADDR
For 9370/LAN only, the AS/400 line description
ADPTADR must match the host MACADDR parameter.
MACADDR can be coded as an 8- or 12-digit
hexadecimal number; the 8-digit variation assumes
4000 in the first four positions (4000xxxxxxxx).
If the AS/400 line description CNN parameter is
*SWTPP or *SHM, DIAL=YES must be specified for the
host system; if CNN is *MP or *NONSWTPP, DIAL=NO
must be specified.
If CNN(*MP) is specified, the SERVICE
macroinstruction must be used to specify the sequence
in which stations are served.
Exchange identifierEXCHIDPUIDBLK, IDNUM
The AS/400 block number (digits 1-3 of the EXCHID) is
always 056. The remaining 5 digits (based on the
system serial number if *SYSGEN is used) are
specified in the IDNUM parameter.
Line speedLINESPEEDLINESPEED
Line speeds specified for each system must match.
Maximum frame sizeMAXFRAMEPUMAXDATA
Values specified for each system must match.
NRZI data encodingNRZILINENRZI
Values specified for each system must match.
Station addressSTNADRPUADDR
AS/400 system station address must be unique within
host PU definitions. (Ignored within 9370/LAN
environment.)
For more information on AS/400 parameters, see Communications Configuration.
For steps on how to create a line description, see “Creating a line description” on page 5.
22Version 5
Matching AS/400 controller description parameters for a host system
You must match host system communications configuration parameters with AS/400 values. A description
of the AS/400 values are in the following table. For information about configuring host systems, see the
manuals VTAM Installation and Resource Definition, SC23-0111, and Network Control Program ResourceDefinition Reference, SC30-3254.
You can specify some host system parameters on multiple definition statements, such as the GROUP,
LINE, PU, and LU. The following table lists only the lowest level definition statement that is used by the
host system.
To configure an AS/400 to a host system:
v See “Examples: Connecting AS/400 to a host system” on page 26 for an example of connecting an
AS/400 to a host system.
v Use the following table for the controller description parameter.
AS/400
AS/400 Prompt
Adjacent link stationADJLNKSTNPUname
LAN remote adapter
address
Destination service
access point
Parameter
ADPTADRLINELOCADD
DSAPPORTSAPADDR
Host Definition
StatementHost Parameter
AS/400 adjacent link station name must match the
name assigned to the PU macro instruction in the host
system switched major node definition. This match is
required if AS/400 host controller description specifies
RMTCPNAME(*ANY), SWITCHED(*YES) or
SNBU(*YES), and LINKTYPE is *SDLC or *IDLC.
This parameter should be specified only if the host
system is running VTAM Version 4 Release 1 or later
and NCP Version 6 Release 2 or later.
Values specified for each system must match. If
LOCADD is specified, ECLTYPE=PHYSICAL must also
be specified on the GROUP definition statement.
PORTMACADDR
For 9370/LAN only, the AS/400 controller description
ADPTADR must match the host MACADDR parameter.
MACADDR can be coded as an 8- or 12-digit
hexadecimal number; the 8-digit variation assumes 4000
in the first four positions (4000xxxxxxxx).
For 9370/LAN only, the AS/400 controller description
DSAP must match the host SAPADDR parameter.
Local exchange
identifier
The SAPADDR parameter is a decimal value (4-252);
the AS/400 value is specified as a 2-digit hexadecimal
number.
LCLEXCHIDPUIDBLK, IDNUM
For parallel connections only. Required if the AS/400
system specifies RMTCPNAME(*ANY),
SWITCHED(*YES), and LINKTYPE is *SDLC or *IDLC.
The LCLEXCHID specified must match the values
specified in the switched major node definition PU
macro instruction.
Chapter 5. Communicating with host systems23
AS/400
AS/400 Prompt
Maximum frame sizeMAXFRAMEGROUPMAXDATA
Remote control point
name
Remote network
identifier
Source service access
point
SSCP identifierSSCPIDVTAMLSTSSCPID
Parameter
RMTCPNAMEVTAMLSTSSCPNAME
RMTNETIDVTAMLSTNETID
SSAPPUSAPADDR
Host Definition
StatementHost Parameter
Values specified for each system must match.
Required only if APPN(*YES). AS/400 controller
description value must match SSCPNAME specified in
the Virtual Telecommunications Access Method (VTAM)
start options list (ATCSTRyy).
Required only if APPN(*YES). AS/400 controller
description value must match NETID specified in the
VTAM start options list (ATCSTRyy).
For 9370/LAN only, the AS/400 controller description
DSAP must match the host SAPADDR parameter.
The SAPADDR parameter is a decimal value (4-252);
the AS/400 value is specified as a 2-digit hexadecimal
number.
Required if APPN(*YES) or if RMTCPNAME is not
specified. AS/400 controller description value must
match SSCPID specified in the VTAM start options list
(ATCSTRyy).
The SSCPID parameter is a decimal value (0-65535);
the AS/400 value is specified as a 12-digit hexadecimal
number, of which the first 2 digits are 05.
Station addressSTNADRPUADDR
AS/400 system station address must be unique within
host PU definitions. Controller description STNADR
must match the value specified in the line description.
For more information on AS/400 parameters, see Communications Configuration.
Matching AS/400 device description parameters for a host system
You must match host system communications configuration parameters with AS/400 values. A description
of the AS/400 values are in the following table. For information about configuring host systems, see the
manuals VTAM Installation and Resource Definition, SC23-0111, and Network Control Program ResourceDefinition Reference, SC30-3254.
You can specify some host system parameters on multiple definition statements, such as the GROUP,
LINE, PU, and LU. The following table lists only the lowest level definition statement that is used by the
host system.
To configure an AS/400 to a host system:
v See “Examples: Connecting AS/400 to a host system” on page 26 for an example of connecting an
AS/400 to a host system.
24Version 5
Loading...
+ 74 hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.