of 21 /21
© Arcati Ltd, 2013 1 Arcati Mainframe Yearbook 2013 The Arcati Mainframe Yearbook 2013 The independent annual guide for users of IBM mainframe systems SPONSORED BY: PUBLISHED BY: Arcati Limited 19 Ashbourne Way Thatcham Berks RG19 3SJ UK Phone: +44 (0) 7717 858284 Fax: +44 (0) 1635 881717 Web: http://www.arcati.com/ E-mail: [email protected]

Arcati Mainframe Yearbook 2007Arcati Mainframe … Ltd, 2013 1 Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013 Mainframe strategy The Arcati Mainframe Yearbook 2013 The

  • Upload

  • View

  • Download

Embed Size (px)

Citation preview

© Arcati Ltd, 2013 1

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

The Arcati Mainframe Yearbook 2013

The independent annual guide for users of IBM mainframe systems


Arcati Limited19 Ashbourne WayThatchamBerks RG19 3SJUK

Phone: +44 (0) 7717 858284Fax: +44 (0) 1635 881717Web: http://www.arcati.com/E-mail: [email protected]

© Arcati Ltd, 20132

Arcati Mainframe Yearbook 2013

Mainframe strategy

ContentsWelcome to the Arcati Mainframe Yearbook 2013 ............................................................ 3Where is the COBOL in your SOA? ................................................................................... 4Lost without a trace? .........................................................................................................11Challenging times lead to new tools for z/OS control and network management ...... 17

The 2013 Mainframe User Survey .................................................................................... 22An analysis of the profile, plans, and priorities of mainframe users

Vendor Directory ............................................................................................................... 37Vendors, consultants, and service providers in the z/OS environment

A media guide for IBM mainframers ..............................................................................114Information resources, publications, social media, and user groups for the z/OS


Glossary of Terminology .................................................................................................118Definitions of some mainframe-related terms

Technical information ..................................................................................................... 144Hardware tables – zEC12, z114, and z196; mainframe hardware timeline 1952-2012;

mainframe operating system development

Action Software 40Blenheim Software International 49DataKinetics 63QMSI-Quintessential Mailing Software Incorporated 93

Software AG 4, 97Software Diversified Services 17, 98William Data Systems 11, 111


© Arcati Ltd, 2013 3

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

Welcome to the Arcati Mainframe Yearbook 2013by Mark Lillycrop, Publisher

Welcome to the 2013 edition of the Arcati Mainframe Yearbook!

We are very grateful – as always – to all those who have contributed this year by writing articles, taking part in our annual user survey, or updating their company profiles. In particular I must thank the sponsors and advertisers, without whose support this Yearbook would not be possible.

It would seem that 2012 was not a great year for business with much talk throughout the year of a double dip recession, in the Eurozone and other developed economies. But for IBM, the year started with a new person taking on the role of Chair and CEO – and this time it was a the first woman in the job. Virginia Marie ‘Ginni’ Rometty took over the reins in January.

At the close of play on 29 September, IBM’s closing value was $214 billion. Not a bad figure you might think, but the good news for Ginni and the team at IBM was that, for the first time since 1996, it was greater than Microsoft’s. But it’s still less than half of the value of Apple.

If anyone takes any notice of rankings, then IBM is the second largest US firm in terms of number of employees, according to Fortune. They also estimated that IBM was the fourth largest in terms of market capitalization, the ninth most profitable, but was down at 19th in terms of revenue. On the other hand, Forbes rated IBM the 31st largest company in the world in terms of revenue.

Another sign of the lack of energy in the economy is shown by the number of organizations IBM looked to takeover during 2012. There were eight – the same as 2011 and well down from previous years.

In January IBM took over Green Hat (who offer Cloud-based software testing) and Worklight (a mobile app developer). In April it took over Varicent (who provide compensation and sales performance management software solutions) and Vivisimo (for their enterprise search software. In May it was Tealeaf Technology (for their customer experience analytics software). In August it was Texas Memory Systems (for their solid state storage) and Kenexa Corporation (who have talent management solutions. Finally, in September it was Butterfly Software (for their data analysis and migration software).

Interestingly, 30 September 2012 was the deadline set by US CIO Vivek Kundra for public/external facing servers and services, such as webmail, Domain Name Server (DNS), and Internet Service Provider (ISP) services to use native IPv6.

Internal client enterprise networks were given until the end of the fiscal year 2014. So, although most of the rest of the world has been not worrying too much about 32-bit IPv4 addresses running out any time soon (although, somewhere there must be a Mayan calendar saying that they will!), Federal organizations in the USA have been attempting to be IPv6-compliant – whatever that actually means.

For mainframe sites, does it mean that your mainframe must have a public facing IP address? Can you imagine the security implications if that is the case. Most federal agencies have been fairly quiet about how they are working towards (and it seems the deadline has come and gone, but people are still not quite there in terms of compliance) being complaint.

The reason it is important is that if the USA ‘migrates’ to IPv6 working, it will probably result in the critical mass necessary for the rest of the world to adopt IPv6, which has been waiting in the wings since about 1994. It will be interesting to see how this works out.

The Arcati Mainframe Yearbook 2013

Publisher: Mark LillycropEditor: Trevor EddollsContributors: Simon Cooper, Finn Haastrup, Deborah Hodson, Jürgen Lind, Colin van der Ross

© 2013, Arcati Limited.

All company and product names mentioned in this publication remain the property of their respective owners.

This Yearbook is the copyright of Arcati Limited, and may not be reproduced or distributed in whole or in part without the permission of the owner. A licence for internal e-mail or intranet distribution may be obtained from the publisher. Please contact Arcati for details.

© Arcati Ltd, 20134

Arcati Mainframe Yearbook 2013

Mainframe strategy

In terms of hardware, the big story of 2012 was the announcement of the zEnterprise EC12, which promised up to 50% more total system capacity, the firmware-based zAware to identify problems, security built into the architecture, and, of course, the opportunity for hybrid working with Windows blades available for inclusion.

In terms of software, arguably the big news is Version 5.1 of CICS TS and its associated portfolio products. Along with the usual improvements was the concept of identifying and separating ‘applications’ and ‘platforms’ – making it easier to move an application to a different platform.

What can we expect in 2013? It looks like z/OS Version 2.1 will hit the streets. This will require a Sytem z9 or above to run on. We can probably expect a new hybrid mainframe – the zEnterprise BC13, perhaps?. Certainly, IMS Version 13 should become generally available. We do know that from 1 July, the price of Flat Workload License Charges (FWLC) will increase.

As always, in 2013 the mainframe will continue to offer outstanding performance and reliability at the heart of the world’s business-critical applications.

Where is the COBOL in your SOA?Jürgen Lind and Finn Haastrup of Software AG explain how to fully integrate your existing COBOL applications in your enterprise Service-oriented Architecture (SOA) so they can fully participate in business process management, web, mobile and cloud initiatives.

The challengeThe core logic that makes many businesses stand out compared to the competition is actually embedded in their mainframe COBOL, PL1, and Natural programs. Reprogramming those applications in another environment has proved to be a risky and really expensive endeavor, deterring many from leaving their mainframe environments. Luckily the mainframe and associated tools have continued to evolve, securing the mainframe as a rock-solid back-end system for many companies, providing secure, reliable and fast performance. Now business leverage the loosely-coupled Service-Oriented Architecture (SOA) to produce and consume services between the mainframe and other platforms for quick, seamless interoperability.

COBOL programs are also no longer on their way out. For example, COBOL programs still run 90 percent of financial transactions (source: Gary Barnett, Research Director of Ovum, 2005). Thus creating an imperative to ensure COBOL environments can interact with newer non-mainframe systems. Fortunately these newer systems are often an established part of an SOA, providing a unifying foundation for COBOL and many other mainframe assets to fully participate in emerging enterprise initiatives, including business process management, web, social, mobile and cloud.

Software AG’s webMethods EntireXSoftware AG’s webMethods EntireX is a service-enabler designed to provide a cost-effective and efficient path for integrating your mainframe applications into process applications and extending the value of existing common universal platforms. EntireX has proven to be a solid solution for providing bi-directional integration for several hundred customers worldwide.

You can rapidly create new .NET, Java or Web services comprised of business logic and transactions from a wide range of host

© Arcati Ltd, 2013 5

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

environments with EntireX, as shown in Figure 1. Use it to:

• Re-use business logic and transactions from host environments, such as COBOL/CICS, to build services that deliver new business value

• Trigger business processes from your established mainframe applications

• Create services to participate in new web, mobile, social and cloud initiatives

• Reduce costs by easily re-using existing functionality to meet new business needs

• Integrate with the webMethods Integration Server and Business Process Management Suite (BPMS)

You’ve made major investments in the core systems that run your business. Let’s explore how to rapidly assemble new business services

and processes to fast-track integration projects with EntireX.

How it worksEntireX can turn applications running on mainframe, Linux, UNIX or Windows platforms into new services with minimal effort. Programming logic and transactions from a wide variety of host environments can be used to create new services for consumption by .NET and J2EE-based applications. Figure 2 provides a sample list of the many operating systems, environments and code supported by webMethods EntireX.

EntireX is laid out in a typical hub and spoke architecture. The EntireX Broker server is the hub, which is typically placed on the mainframe, and offers a number of services like character

Figure 1: Fast-track SOA integration projects with webMethods EntireX

© Arcati Ltd, 20136

Arcati Mainframe Yearbook 2013

Mainframe strategy

translation, security, persistence, and timeout-handling to the servers and clients that connect to it. The clients and servers that connect to the EntireX Broker just need to make use of the relevant APIs (ie Java®, Microsoft® .NET, C, C++) that are delivered with webMethods EntireX. The very popular generic Web service interface, based on AXIS2, is provided with wizards to help create XML mappings for both incoming and outgoing Web service calls.

Eclipse-based wizardry simplifies service developmentEclipse-based wizards simplify the development

of Web services as shown in Figure 3. To develop interfaces for EntireX, an Eclipse plug-in is provided that generates the clients and servers based on a description of the interface, known as the IDL (Interface Definition Language). This interface, which is just a text-file, is usually extracted by the wizard from an existing COBOL-source or WSDL file.

As soon as the IDL file is settled, you can generate a large number of different clients/servers such as XML mappings for Web services or COBOL subprograms or CICS programs. After compiling in your own COBOL environment, these bi-directional integration spokes are ready for use.

Operating Systems • Linux• z/Linux• z/OS• z/VSE• z/VM• Windows• BS2000/OSD• AIX• HP-UX• OpenVMS• Solaris• IBM i (OS/400)

Environments • CICS• Complete Batch• .NET• IMS• J2EE• SAP• R/3 and XI/PI• XML• Web Service• CA-IDMS• WebSphere MQ• webMethods Integration Server

Code• Assembler• C• C#• COBOL• Java• Micro Focus• PL/I• Natural• RPG, CL

Figure 2: Sample operating systems, environments and code supported by webMethods EntireX

Figure 3: Three simple steps to Web service development with Eclipse-based wizards

© Arcati Ltd, 2013 7

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

For example, EntireX at one installation handles all the complexities involved in calling a Web service at an external credit-information-company from a CICS-program by providing an auto-generated subprogram that interfaces like any other subprogram from CICS or batch.

Extract the parameters and you’re ready to test! The extraction wizard is part of the Eclipse-based Software AG Designer, which comes with webMethods EntireX. You can extract parameters from the sources and referenced copybooks directly from the mainframe partitioned data set or CA-Librarian datasets or the local file system. Source-control systems like Subversion, CVS, RTC, and Clearstream have Eclipse plug-ins and thus can be used to make the sources available for the extraction wizard.

The fact that the wizard not only extracts parameters from a single copybook (potentially including other copybooks), but can also do a full parsing of the program to determine which data structures are used, makes a huge difference regarding how easy it is to directly reuse existing programs where the linkage section/COMMAREA is not the simple “one-copybook” type. As soon as the parameters are extracted to an IDL file, another wizard can start a test-program for immediate testing of the existing server-program as shown in Figure 4.

Generic servers in CICS or batch do the magicThe magic behind this easy-to-use interface is that along with EntireX comes a generic server. A CICS program that does all the parsing of the incoming data from the client then passes it on to the real CICS program with the right parameters. Behind the scenes, a generated client knows how to “marshal” the client-side data types and pass this metadata on to the generic server, alongside the actual data.

Pointing to LAAAARGE buffersThe COMMAREA limit of 32KB has been a nuisance to CICS developers for a long time. Two workarounds for these have been widely implemented. The oldest one is the use of buffers where pointers in the COMMAREA point to these “getmain’ed” areas, which are then redefined. A somewhat newer workaround is the use of CICS Channel/Containers where the >32KB data can be pushed/passed from one program to another. Both of these large data interfaces are supported out-of-the-box by the EntireX wizard.

These large data interfaces can be used, for instance, for receiving daily grocery orders from supermarkets into the logistic system in CICS. The orders, with pairs of Item-IDs and quantity, vary in size from less than 100 bytes up to 1MB, so the built-in compression option of EntireX comes in handy when it comes to bandwidth. The

Figure 4: Extract parameters and immediately test services

© Arcati Ltd, 20138

Arcati Mainframe Yearbook 2013

Mainframe strategy

theoretical max-limit of one call is currently 2GB - though I dare you to ask your CICS-administrator to increase the thread size to 2GB!

More realistically, a data-area of some hundred MB might actually be worthwhile in a batch environment as an alternative to an FTP now that a single 100MB call only takes a second or two. Definitely quite a different use case to the more than a thousand calls per second that characterize the performance regarding small message sizes!

Flexible XML mapping for open-ended arraysModern Web service interfaces often make use of the flexibility of XML data-structures, but you never know if there are 0 or 1000 occurrences of an array-element. With the flexible XML mapping and usage of (almost) open-ended COBOL arrays using “depending on” structures, it is no problem to fit into the Web service world. This kind of interface can of course be combined with the Large Buffer and Channel/Container interfaces for transmitting large data sizes.

Natural-born polyglot In our increasingly globalized world, handling of many different encodings becomes a standard requirement for most systems. But the translation to/from ASCII and EBCDIC is not a trivial task for customers who have over time made their own here-and-now character translation routines which do NOT correspond to any industry standard.

EntireX handles all standard ICU code-page conversions, including right-to-left and Unicode/multi-byte. The client and server codepages (if not explicitly specified) are automatically passed on from the host-system, but additionally offers the use of customized translation tables that only apply for specific services. Having the ability to “speak multiple languages”, EntireX is the polyglot who can make sense out of the Bablyonic chaos (Figure 5) of the many translation routines.

RACF protected? Naturally!Almost all mainframe environments use RACF or another SAF-compliant security system. Correspondingly, no security-responsible person would, without having his arm painfully twisted around his back, allow access to mainframe resources like programs, dataset or databases without a prior authentication and authorization. EntireX Security plugs directly into RACF and allows for the reuse of existing groups for access rights.

Choose your communication patternDifferent scenarios require synchronous or asynchronous communication patterns. EntireX supports both. Many customers, who previously used asynchronous message-queuing systems, often start off by implementing their first EntireX services asynchronously. But then discover that the quick performance of synchronous calls makes it much easier to implement the logic and error-handling in most scenarios ie “call and wait for answer” compared to “make request, persist message, poll for answer”.

Figure 5: Handling multi-language encodings may be a Babylonic task without the right tool

© Arcati Ltd, 2013 9

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

One Software AG customer once did a benchmark that showed that EntireX, compared to a standard asynchronous messaging system, used 20 percent of the bandwidth for the same payload with only 10 percent of the lines of code.

How EntireX works in practiceCOBOL programs in both CICS and batch can easily participate in bi-directional integration with other applications in an SOA. Here are a few examples:

• At a major Nordic Supermarket chain, the COBOL mainframe environment not only provides servers to their SOA infrastructure, batch and CICS programs also push mission critical transactions directly into the SOA messaging system using EntireX. The extremely time-critical route-planning information is now provided to the truck drivers as well as packaging information to a meat-packaging partner, all in time for the next morning’s deliveries.

• A major Italian company in Financial Services called Software AG to help them get their fast and stable banking backend system in CICS-COBOL to work with micropayments using secured Web services without exploding the MIPS usage. Today they now have more than 15 Web services in production.

Learn more about how to use webMethods EntireX to integrate COBOL with SOA and BPM, visit: www.softwareag.com/corporate/products/wm/mainframe_integration/entirex/overviewand click on the “Got COBOL?” banner on the right.

Jürgen Lind, Senior Product Manager, Software AG (Darmstadt, Germany)

Jürgen has been responsible for the EntireX mainframe integration product line since 2006. Prior to that, he was a solution architect in research and development and business projects.

Finn Haastrup, Senior Solution Architect, Software AG (Denmark)

Finn has been involved in mainframe integration projects since 1988. He is an expert in gluing the mainframe and the world of Web services, Java and .NET together into one big process landscape.ABOUT SOFTWARE AGSoftware AG helps organizations achieve their business objectives faster. The company’s big data, integration and business process technologies enable customers to drive operational efficiency, modernize their systems and optimize processes for smarter decisions and better service. Building on over 40 years of customer-centric innovation, the company is among the top 10 fastest-growing technology companies in the world and is ranked as a “leader” in fifteen market categories, fueled by core product families Adabas and Natural, ARIS, Terracotta and webMethods.

Software AG has more than 5,500 employees in 70 countries and had revenues of €1.1 billion in 2011. The company is headquartered in Germany and listed on the Frankfurt Stock Exchange (FRA: SOW). Learn more at www.softwareag.com.

Software AG - Get There Faster

Software AG | Uhlandstraße 12 | 64297 Darmstadt | Germany

© Arcati Ltd, 201310

Arcati Mainframe Yearbook 2013

Mainframe strategy

Get There Faster.™


Find out more at www.ITModernization.com

Modernizing isn’t simple, but choosing the right partner is. Here are the four signs you’ve found the right one.

Business Process DrivenGain a competitive advantage when you focus on achieving your business goals.

Continuous AdaptabilityIncrease your organizational agility with your own roadmap to modernization.

Cost EfficientIncrease your ROI by executing a modernization plan customized for you.

Software AG ExpertiseDrive change with best-in-class tools and resources backed by 40 years of experience.

© Arcati Ltd, 2013 11

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

Lost without a trace?Simon Cooper looks at what users might like to see in a network trace – and takes a look at how those features are realized in one such product.

A trace can be a really useful way of debugging problems with networks or applications. In fact sometimes it’s the only way. The kind of low-level information found in a trace often provides a unique clue as to the root cause of a problem. Despite this, many mainframe professionals use tracing solely as a last resort, when all other attempts to resolve a problem have failed. This is due to the perceived difficulty of taking a trace, as well as the complexities of interpreting the resulting output.

Traces can contain vast amounts of data, all of which may potentially be required, but the volume of data can obscure the very problem you’re trying to identify. Interpreting the trace can be equally problematic, requiring knowledge of IPCS to browse and analyze CTRACE or GTF datasets, as well as a familiarity with many IBM manuals, an understanding of trace record layouts, TCP/IP control codes, the format and meanings of VTAM sense bytes, 3270 order codes and so on, just to begin to interpret a trace. Consequently, the job of finding a problem using a trace is usually a long, labour-intensive, and tedious task. Under these circumstances it’s not surprising that trace taking is often the last resort of an exasperated technician.

Keeping it simpleThe introduction in 1994 of EXIGENCE from William Data Systems brought an uncomplicated and straightforward approach to tracing network-related connectivity, response time or performance problems. Now known as ZEN TRACE & SOLVE (ZTS), the product makes it simple to not only take a trace but, perhaps more importantly, interpret the trace data, guiding the user to the problem area

and usually suggesting a solution within minutes. By eliminating all the difficulties associated with trace capture and analysis, organizations can save countless hours and thousands of dollars.

Defining and capturing tracesZTS enables you to easily define and initiate a variety of trace types, including TCP/IP, VTAM, and even NCP line traces, through one of three user interfaces (UI) – Web browser, 3270, and PC client (see Figure 1). Trace definitions are all done from within the UI, requiring no batch jobs to be started and therefore no knowledge of JCL is required.

For TCP/IP traces, you define only a description and the name of the address space running TCP/IP (and optional filters) for trace capture. Traces can be captured in up to seven address spaces simultaneously, tracing all IP packet exchanges. Enhanced facilities are available for several popular applications, for example TCP, UDP, ICMP, FTP, LPR, RIP, NCPROUT, OSPF, GRE, and Enterprise Extender protocols.

For VTAM traces, the only piece of information you need to provide is a brief description of the trace and the LU name (the terminal or application) that is causing the problem. VTAM traces will capture all exchanges between LUs, LU 6.1, LU 6.2, LU types 0, 1, 2, 3 and so on.

Tracing intermittent problemsOf course it’s not always possible to know in advance when a problem might occur, which may mean you need to leave a trace running for a long period, potentially creating an enormous trace. ZTS has a neat solution to this, allowing you to define a ‘Wrap Mode’ trace, which only uses a finite amount of space for trace recording, looping around and re-sequencing the trace entries into chronological order once the trace is stopped.

© Arcati Ltd, 201312

Arcati Mainframe Yearbook 2013

Mainframe strategy

Automatically taking and merging multiple traces Tracing in a complex multi-LPAR and/or multi-processor environment would normally involve the taking and examination of multiple traces. For example, you may need to trace a single client IP address that has many connections to several systems. The difficulty is compounded if an IP client connects to a service running on multiple systems using dynamic VIPA, because it is impossible to be sure with which system the connection will be made.

In such environments, usually the only option available for problem diagnosis is to initiate traces on each individual system. This is difficult enough. Once taken, all the individual trace records need to be sorted into chronological order before being interpreted to try and diagnose the cause of the problem. Assuming the technician has the knowledge and experience to carry out this task, it is still a time consuming, and therefore costly process.

The ZTS solution is to allow the definition and capture of multiple IP traces at once. Known as “Group Trace”, it’s basically the same as a ‘normal’ trace except that it can be simultaneously started on a group of systems with the resulting trace output viewed as if it had been taken on a single system, with the trace entries interleaved according to their capture time.

New to Version 5 of ZTS, you can now synchronize group trace entries by normal, relative or incremental time, giving you a huge advantage when analysing trace data spanning multiple systems in different time zones. You can also now define Group traces on-the-fly; it is no longer necessary to pre-define the hosts on which a group trace is to be captured. You can even build a group trace from a set of traces that were not originally captured as part of a group trace (eg imported traces).

Figure 1: The evolving EXIGENCE UI


Original 3270 interface circa 1994

PC javaTM client circa 2005

© Arcati Ltd, 2013 13

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

Analysing a captured traceHaving captured a trace, it can then be viewed in a number of different formats, from a summarized view of the session exchanges between socket pairs or LUs known as the Flow display, to an expanded format that shows data structure breakdowns right to the individual bit level, with settings with descriptions provided.

Any 3270 data streams in a trace can be displayed in their entirety either in hex or with all of the control orders interpreted, even when the data stream is encapsulated as in the case of Enterprise Extender. This is particularly useful when an error has been caused by a fault in transmitted data. You can switch between the different display formats with no loss of context.

ZTS analyses the trace data, automatically scanning and highlighting any negative or exception responses and pairing them with the original request. ZTS can even explain what the highlighted problem areas mean. ZTS not only provides an explanation of the data in error, but also pinpoints the most likely cause of the problem and in some cases can even suggest a solution.

ZTS also provides comprehensive search functions, enabling you to scan a trace, either forwards or backwards, for any TCP/IP or VTAM condition, data sequence, or even down to a single bit in a specific byte. This is particularly useful when you already have a good idea of what a problem might be and just want your suspicions confirmed.

Built-in knowledge baseZTS includes a comprehensive ‘Explain’ function. This enables anyone, from Help Desk staff to trace experts to find information on any TCP/IP or VTAM/3270 term, sense code, 3270 order code, and so on. Searching through many IBM manuals to find the meaning of a particular bit setting becomes a thing of the past with ZTS.

ZTS features at a glance• Enables online trace management – no batch

set-up required• Supports simultaneous capture of multiple• IP and SNA traces• Formats and translates all trace data prior to

display• Displays all data flows with meaningful

annotation• Enables display mode switching without loss

of context• Supports powerful filtering techniques enabling

traces to be targeted at: • Defined Applications: (Telnet, FTP, EE etc) • Defined Protocols: UDP, ICMP, TCP, SASP

and specific numbered protocols • Defined Port numbers • Defined IP addresses • LU to LU sessions • LU to Application (eg CICS sessions)• Allows individual trace file sizes to be

minimized via the ‘Wrap Mode’ feature • ‘Peek’ feature allows traces to be browsed “in

flight”• Simplifies import and evaluation of externally

captured traces• Simplifies export, in IBM-recognizable or

libpcap format, of captured traces• Manages all trace types including: • IP Packet and Data Traces • EE Traces • VTAM buffer and I/O Traces • VTAM Extended Trace (XTD) • VTAM Internal Trace (VIT) • NCP 3745 Trace including Scanner,

Generalized PIU and TG Traces • 3746/900 /950 Trace Import• Enhanced analysis facilities for TCP, UDP,


• Incorporates a 3270 and ZEN browser UI. PC client also available

• Provides comprehensive help on any IP, VTAM and/or 3270 terminology, message or sense code

• Full support for IP Version 6• Supports trace printing• Full audit of trace activities.

© Arcati Ltd, 201314

Arcati Mainframe Yearbook 2013

Mainframe strategy

Special support for Enterprise ExtenderWDS have long been advocates of Enterprise Extender (EE), recognizing its importance as a means of preserving investment in SNA while allowing the exploitation of the latest advances in IP networking technologies. We have developed a number of unique products and features that address shortcomings with the management of EE, including specialist products for monitoring, encrypting, and authenticating EE data streams.

Tracing in an EE environment presents unique challenges, because although the protocol carrying the data is IP-based, the payload data itself is SNA in origin. The Enterprise Extender support in ZTS enables you to expand all the EE headers, stripping away IP-related material to provide an unencumbered view of the underlying SNA (TH, RH, and RU) payload.

A new user-interface for EXIGENCEZTS continues to be at the forefront of network problem determination and is a central component of our ZEN network management suite. The latest version of ZEN TRACE & SOLVE (Version 5), is a no-cost upgrade for customers and brings the proven advantages of our flexible and powerful ZEN User Interface to all EXIGENCE customers.

Innovative Web viewAs a major component of the ZEN suite of Network Management solutions, the ZTS browser interface benefits from the innovative design provided by ZEN. No special desktop software is required and, unlike other browser-based solutions, it has the distinct advantage of not requiring a separate Web server.

A Web server is usually part of a three-tier system comprising a network data collector on the z/OS host, a separate Web server where the Web pages are generated, and a client for data display. A three-tier approach has many drawbacks. It adds cost (additional hardware, software, and configuration), adds complexity to the set-up and maintenance

of the monitor system, adds overhead (additional bandwidth load from the extra connections), and provides an additional point of failure.

The ZTS user interface is part of a two-tier architecture (see Figure 2) that takes advantage of the core strengths of ZEN – namely, efficient, high-speed, and low-overhead data delivery. Proving that less really is more, this innovative two-tier approach has a number of benefits over three-tier systems as it:

• Reduces costs without the need for additional hardware or software

• Simplifies implementation since client software is not required

• Deploys rapidly, providing immediate access to trace data from a standard web browser

• Necessitates little or no processor overhead because all graphical formatting is performed inside the browser

• Provides single interface for all your network management needs













se fu


















ZEN componentsZEN








Figure 2: ZEN’s two-tier architecture

© Arcati Ltd, 2013 15

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

• Is SMP/E maintainable, enabling rapid application of new features and services.

Importing and exporting tracesZTS doesn’t just work with traces taken within the product. It also allows traces to be imported from:

• IBM’s IP packet traces• IBM’s VTAM Buffer traces• OSAENTA traces• libpcap traces (eg Wireshark, Sniffer).

Once a trace has been imported, you have the full power of all the ZTS commands and Expansion functions to help you analyse it.

You may also export ZTS traces to a disk dataset so that they can be sent elsewhere for further analysis. IP traces are exported in CTRACE format, SNA traces in GTF format, and OSAENTA and libpcap traces in their respective formats.

Automatic event tracingThe ZTS command interface can be used to Add, Redefine, Start and Stop traces allowing event-based dynamic trace management activities to be performed, for example, to define and start a trace automatically dependent upon some other event in the system.

A tool of first resortThe traditionally complex task of tracing has usually been undertaken by a diminishing number of old-school skilled technicians. However, the

availability of modern technologies improves accessibility to this extremely useful problem-solving technique for the next generation of technicians, removing complexity and turning tracing into a tool of first resort.

Simon Cooper has worked in the Independent Software Vendor market for nearly twenty-five years and has held a variety of technical, sales, and marketing roles in addition to his current position as head of business development at WDS.

William Data Systems (WDS) is a pioneer of specialized IBM System z network management solutions. Established in 1993, we are an independent global company that provides innovative solutions to run mainframe networks efficiently and securely. ZEN, the WDS network management suite, offers a selection of user-friendly and cost-effective solutions to meet your unique needs.

To help customers overcome both business and technology challenges, WDS provides customers with licensing and pricing terms that are as flexible as our solutions.

WDS supports customers worldwide across all vertical markets and our client list includes Fortune 100 companies and government agencies. WDS is an IBM Business Partner and a member of the IBM PartnerWorld for Developers program. We are committed to the global z/OS networking market and to leading the way with innovative solutions through the latest advances.

© Arcati Ltd, 201316

Arcati Mainframe Yearbook 2013

Mainframe strategy


Reduce time, effort and annual costsSimplify your network management

One ZEN suite, many network solutions

Your gateway to cost saving z/OS Network Management

© Arcati Ltd, 2013 17

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy


Reduce time, effort and annual costsSimplify your network management

One ZEN suite, many network solutions

Your gateway to cost saving z/OS Network Management

Challenging times lead to new tools for z/OS control and network managementColin van der Ross and Deb Hodson at Software Diversified Services describe how struggling with smaller budgets and fewer resources has led some mainframe shops to unexpected improvements: replacement tools that let them do more with less.

In every discussion we have with mainframers these days, we hear about cost pressures like we’ve never heard before. z/OS shops are having to get more done with less budget and an accelerating brain drain. Seasoned, irreplaceable staff have gone into retirement. Making up for the

loss by spending more on new software seems out of the question.

Some hard-pressed shops try to negotiate cheaper licenses with vendors, and sometimes that works.

But over the past two or three years we’ve also been able to help many z/OS shops get ahead in hard times by replacing old, legacy software tools with more productive, less costly alternatives. Shops ranging from 300 to over 30,000 MIPs have saved significant money by replacing old tools with new ones that cost 40% to 60% as much.

For z/OS network management and system automation, those shops installed VitalSigns software from Software Diversified Services (SDS). They report back that VitalSigns for Network Automation and Control – VNAC for short

Figure 1: With a single log-on to a secure server, a VNAC user at a web browser sees z/OS system logs, VTAM session data (SAW), command lines, JES spools, automation scripts, datasets, and SNA and IP network alerts from any number of z/OS systems, LPARs, and sysplexes.

© Arcati Ltd, 201318

Arcati Mainframe Yearbook 2013

Mainframe strategy

– is simple to use, provides some new functionality, and drastically lowers expenses.

“Getting VNAC was a no-brainer,” one z/OS system programmer told us.

Simplifying z/OS management by centralizing accessTake Session Awareness (SAW) for example.

You can log onto a single web-browser application and VNAC will show you all the SAW data from all the z/OS LPARS (see Figure 1). All the session information shows up on a single, searchable screen. The filter tool selects for sessions that ended in error this morning, for example. Or for sessions originating with a specified LU. Users can filter and re-arrange the display to suit their networks.

VNAC does the same for z/OS system and console messages, including data from the SNA network and the TCP/IP stack. They can all be filtered and read from a single browser window.

Let’s say you’re looking for errors in the same application running on a dozen different LPARs, across three different sysplexes. And you want to see those errors now, immediately, the moment they show up in system logs.

Not a problem. Tell VNAC to look at all those logs and bring up all those errors. That kind of consolidate-and-search power dramatically decreases the time it takes to diagnose problems.

To make life simpler for technicians who have a hard time finding details in IBM bookshelves (and who doesn’t?), when VNAC presents system and network logs, it also provides direct hyper-links to the corresponding IBM documentation. In other words, when SAW data tells you that a session ended for “reason #00, sense code 087D0001,” mouse-click on those numbers in the VNAC display. VNAC calls up the IBM documentation, and you learn that an unbind command disconnected two LUs because a VTAM system control point (SSCP) tried and failed to re-route a request. Further details explain possible causes of the error (see Figure 2).

Figure 2: VNAC presentation of z/OS messages includes hyperlinks to the corresponding IBM bookshelf documentation.

© Arcati Ltd, 2013 19

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

Automating responses to messages, AND to alertsNetView’s Message Automation Tables (MAT) define how REXX execs automatically respond to messages at system consoles. Those NetView execs readily migrate to VNAC.

You can tell VNAC to recognize specific system messages, SNA network messages, and TCP/IP stack messages, then VNAC execs can respond in specific ways. Later, when you need to know what automation is in place, a single VNAC command will report on all the responses defined for all systems, LPARs, and sysplexes.

VNAC can also respond to alerts regarding performance of the IP network.

“VNAC reduced our annual costs,” said a system

VNACVitalSigns for Network Automation and Control (VNAC) provides all the basic, commonly-used functions of NetView®: - Network management commands

(NetView’s NCCF) - System and network automation via REXX

execs - Session Awareness data, or SAW

(NetView’s NLDM) - Browsing of system logs, including SNA

and IP network data - Easy management of global variables and

global stem variables - Panel applications, and simple import of

NetView panels to VNAC- Central management of jobs, logs, and output

on the JES spool (SDSF).

Other products of the VitalSigns suite provide - Monitoring and diagnosis for the IP network,

including HPR, Enterprise Extender, OSA, and IPv6.

- Seamless FTP – SFTP solutions.

programmer at a large insurance company, “plus we got an IP monitor we’d wanted for years.”Automation regarding IP network performance is a collaboration of two VitalSigns products: VitalSigns for Network Automation and Control and VitalSigns for IP (VIP).

A VIP Agent on a z/OS LPAR monitors an IP network and alerts network engineers immediately when performance or events exceed specified thresholds. Thus they learn immediately about slow response times, failing network devices, overtaxed interfaces, etc.

VIP sends those alerts to e-mail addresses, to SNMP receivers, to system consoles – or directly to its partner VNAC. Then VNAC can invoke a REXX exec to fix problems, redirect traffic, or otherwise deal with the issue.

SDS provides VNAC and VIP as a combined solution. A VNAC license includes a free VIP license. Along with sending alerts to VNAC, VIP provides monitoring and control for z/OS HPR, Enterprise Extender, OSA, telnet, FTP, Sysplex Distributor, and TCP/IP traffic.

Saving money by offloading z/OS CPU cyclesThe mainframe shops we’re talking about implemented VitalSigns because of low license fees. There are other savings as well.

VitalSigns can reduce the bill for using z/OS CPU because it gives customers the option of moving some processing loads to zIIP or zAAP processors, or to z/Linux, or to separate Linux or Windows boxes.

Here’s how it works:

VitalSigns software is built in an Agent-Server-Browser architecture.

VNAC Agents run as jobs or started tasks on any number of z/OS LPARs. They provide the system automation, command execution, console

© Arcati Ltd, 201320

Arcati Mainframe Yearbook 2013

Mainframe strategy

listeners, z/OS log and dataset access, and security. Agents can also run as z/OS subsystems, meaning they can start before the JES spool and so help automate system IPL.

The Agents typically communicate with two VNAC Servers – one primary, one hot standby – running on z/OS USS (UNIX System Services) on two different LPARs. Both Servers record console messages from all Agents to the Servers’ databases and pass user-input commands to Agents for execution.

VitalSigns Servers don’t have to run on z/OS USS, however. If customers so choose, the Servers run equally well on zIIP, zAAP, or IFL processors, or on z/Linux, or on remote Linux or Windows boxes. When CPU is billed by the MIP, running VitalSigns servers outside of z/OS can lower the bills. It can also delay the cost of CPU upgrades.

VitalSigns servers communicate with users at common desktop web browsers, typically Internet Explorer or Firefox. So users see a graphic, easily navigated interface for issuing z/OS commands and reading z/OS data from any LPAR on a network. Individual users can customize the interface to their individual needs. And any number of users, almost anywhere on Earth, can access VNAC.

Migrating and saving in 60 daysVitalSigns customers report that the process of installing and configuring VitalSigns surprisingly painless. A 60-day project is usually all it takes. “It’s quick and easy to implement,” reported a publishing company’s network engineering manager.

The migration typically starts with a free trial. SDS tech support shows customers how to install and configure VitalSigns before anyone makes a commitment.

SDS technical support will show you how to identify automation currently in place, and how to make VNAC fill the same roles. You can run VNAC alongside other tools, risk free, until everyone is satisfied.

Existing REXX execs and panel applications migrate to VNAC with little or no change. VNAC’s graphic interface on your web browser is intuitive to navigate. When you have questions or need help, SDS provides immediate assistance.

With VitalSigns for Network Automation and Control, SDS provides a companion product free-of-charge: Either VitalSigns for IP or VitalSigns for FTP.

Software Diversified Services has been providing the mainframe industry with software tools for 30 years. Technical support and software developers work side-by-side in the Minneapolis, Minnesota, headquarters. SDS tests VitalSigns products on an in-house z/114 processor that includes zIIP and zAAP engines. SDS customers include many Global 500 companies in banking, finance, insurance, and retail, as well as local, state, and national governments.

Colin van der Ross is Senior Systems Engineer at SDS. He was formerly a network systems programmer for a large bank, where he led one of the first roll-outs of IBM Enterprise Extender and made extensive use of NetView.

Deborah Hodson is Sales Manager at SDS. She has 30 years’ experience in IT, having worked in sales management and consulting roles in IBM in large systems, in Candle’s OMEGAMON and Automated Operations teams, and in HP’s OpenView software organization.

Hodson and van der Ross welcome your questions: [email protected], 763-571-9000.

© Arcati Ltd, 2013 21

Arcati Mainframe Yearbook 2007Arcati Mainframe Yearbook 2013

Mainframe strategy

How do I convert FTP to SFTP on my z/OS main-frame with no batch JCL changes? There’s only one way. VitalSigns for FTP and SSH/Tectia.

If your organization is facing new security and compliance requirements, you’re probably bracing for impact. Thousandsof FTP jobs to be edited and changed to SFTP command syntax. Hundreds of hours of tedious work and additional training.

Or… you can leave everything to VitalSigns for FTP (VFTP) and SSH/Tectia, a solution that makes the migration from FTP to SFTP seamless and automatically converts FTP to SFTP.

This integrated SFTP solution enriches your system with transparent monitoring and auditing capabilities that meet today’s toughest compliance standards, including SOX,

(Sarbanes-Oxley Act), GLBA (Gramm-Leach-Bliley Act), PCI DSS (Payment Card Industry Data Security Standard), HIPAA (Health Insurance Portability and Accountability Act).

From an intuitive browser interface, users can automate and control batch processes, drop in security, encryption and authentication protocols, and closely monitor FTP traffic on your TCP/IP network.

Save time. Save money. And save your sanity! Take a closer look at VitalSigns for FTP and SSH/Tectia.

VitalSigns for FTP is brought to you by SDS. Since 1982, Software Diversified Services (SDS) has provided several hundred enterprise clients worldwide with the highest quality mainframe software, documentation, and technical support. We are an IBM “Partner in Development” and an advanced member of IBM PartnerWorld®.

See VitalSigns for FTP in action with a free trial. Contact [email protected] today,

or visit www.sdsusa.com/VFTP

system manager

large insurance company

VitalSigns™ for FTP and SSH are saving us a million dollars or more. There’s nothing like this combined solution on the market—period.

SDS-VFTP-Tectia-ad-mec-5-HR.indd 1 06/10/2012 17:17