Systems Admin, L4/SME MF system Programmer

Job Title : Systems Admin, L4/SME MF system Programmer
Location : Atlanta, GA (5 Days onsite / Relocation works)
Duration : 12+ months Contract (with possible extension) 

Job Description

System Administrator--the Key skills are Adabas and Natural programming language together as for a MF Sysprog, and skill levels expected to be at L4/SME levels

1.      Mainframe Competencies

• z/OS Job Control Language (JCL): You need to understand how to read, modify, and submit JCL for batch processing, allocating datasets, and executing installation/verification jobs.
• SMP/E (System Modification Program/Extended): Essential for installing and applying maintenance, PTFs, and managing Software AG product sysmods.
• ISPF & TSO: Proficiency in navigating the z/OS environment, editing files, and monitoring spool outputs (via SDSF).
• APF Authorization & Link Lists: Understanding how to mark load libraries (e.g., Natural load modules) as authorized so they can execute privileged system functions.

2. Software AG Specifics

Deploying this specific combination of Software AG stack (Adabas, Natural, Event Replication, Adabas Review, Entire Network, Natural ISPF, Predict, Entire System Server, and Com-plete) on z/OS shifts your role from a standard installer to a full mainframe systems programmer. Because these products closely hook into the operating system kernel, databases, messaging layers, and online TP monitors, a highly specialized skill set is required.

Database & Replication Infrastructure (Adabas & Event Replicator)

• Adabas SVC (Supervisor Call) Installation: Knowledge of defining and installing Software AG-specific SVCs into the z/OS system. You must know how to update the IEASVCxx member in SYS1.PARMLIB
• Cross-Memory Services: Understanding z/OS Cross-Memory (XMEM) communication protocols, as Adabas uses these for highly efficient inter-address-space communication.
• Dual/Protection Log (PLOG/CLOG) Management: Designing and allocating sequential datasets for Adabas logs to ensure the Event Replicator can scrape the PLOG for real-time data streaming.

·  Com-plete Environment Tuning: Com-plete functions as your Teleprocessing (TP) monitor. You must know how to allocate roll data sets, thread sizes, and page pools inside its region.

• Subsystem Definitions: Experience adding Com-plete as a recognized z/OS subsystem via the IEFSSNxx  member in PARMLIB/
• Natural Configuration (NATPARM): Ability to assemble and link-edit a custom Natural Parameter Module (NATPARM). You must explicitly configure it to hook into Com-plete, Adabas Review (REV), and the Entire System Server interface.
• Authorized Program Facility (APF) Management: Both Entire System Server (NPR) and Adabas require execution from APF-authorized libraries. You must manage authorization via PROGxx  members.
• Adabas Review Link Routine Exits: Skills to configure Adabas initialization parameters (REVIEW=DBID) and manage the ADARVU exit module within the Adabas nucleus address space.
• Advanced IBM Utility Proficiency: Master-level experience with IEBCOPY , IDCAMS  (for allocating VSAM and Adabas cluster files), and AMASPZAP  (Superzap).
• Applying Product Zaps: Software AG frequently delivers emergency code fixes as Zaps. You must be skilled at executing AMASPZAP  JCL to alter load modules directly in the target libraries.

Recommended Deployment Order

To ensure proper integration and verify prerequisites along the way, deploy the components in this sequence:

1. Adabas Nucleus (Base database engine).
2. Entire Net-Work (Establishes cross-LPAR/cross-platform communication channels
3. Com-plete (Provides the TP region online environment).
4. Natural (Base language environment hooked into Adabas and Com-plete).
5. Predict ( FDIC ) (Must be installed early so Natural programs can use its dictionary definitions)
6. Adabas Review (Performance monitoring tied directly to the Adabas nucleus).
7. Entire System Server (Enables Natural programs to interact with z/OS system services).
8. Natural ISPF (Relies on Natural, Predict, and Entire System Server being completely active)
9. Event Replicator for Adabas (Configured last, as it relies on a running Adabas hub, Natural administration screens, and messaging configurations).

The essential skills required for a high-performing Adabas DBA are structured below:

1. Database Operations & System Administration

• Utility Execution (ADARUN): Mastery of the standard ADARUN configuration control card. You must know how to configure global database parameters like LBP (Buffer Pool size), LS (Sort Space), and NT (Number of Threads).
• Database Formatting & Management: Proficiency with core file management utilities. This includes ADAFRM (Format datasets), ADALOD (Load/Unload files), ADADEV (Device management), and ADADBS (Database state modifications, like adding fields or changing file lengths).
• Storage Allocation (ASSO, DATA, WORK): Expert understanding of Adabas architecture components. You must know how to allocate, monitor, and extend Associator (ASSO) space, Data (DATA) space, and the vital WORK dataset (used for two-phase commit, transaction logging, and caching).
• Backup and Recovery Processes: Expertise in configuring robust backup jobs using ADABCK. You must know how to perform online backups, recover from system crashes using the Protection Log (PLOG), and regenerate data up to the point of failure.

2. Mainframe Ecosystem Integration (z/OS)

• Storage and File Allocation (IDCAMS): Advanced skills using IBM IDCAMS to define and allocate massive Linear Data Sets (LDS) or Sequential Data Sets required for Adabas clusters.
• System Definitions & SVCs: Collaboration with system programmers to implement the Adabas Supervisor Call (SVC). You must understand how the ADASVC acts as the bridge between user address spaces (like Com-plete or CICS) and the database engine.
• Job Control Language (JCL): Expert-level capability in writing batch JCL, utilizing cataloged procedures, managing tape setups for archival logs, and optimizing step conditional execution.

3. Data Architecture & Modeling (Predict & Natural)

• Predict Data Dictionary Administration: Mastery of Software AG's Predict. You must know how to define files, fields, superdescriptors, and subdescriptors in Predict, and then physically instantiate them in Adabas via generation schemas.
• Adabas-Specific Data Modeling: Unlike relational DBAs (SQL), an Adabas DBA must master non-relational modeling concepts. This includes Periodic Groups (PE) and Multiple-Value Fields (MU) to map repeating data structures inside a single record efficiently.
• Natural Language Basics: While not an application developer, an Adabas DBA must be able to read and write basic Natural code. You need this to troubleshoot application access patterns, run database diagnostic scripts, and analyze performance bottlenecks.

4. Performance Monitoring & Optimization

• Adabas Review Configuration: Expertise in setting up and analyzing reports via Adabas Review.  You must know how to identify "heavy hitters"—long-running Natural programs or poorly indexed queries causing high I/O or CPU consumption.
• Command Queue Tuning: Skill in monitoring the Command Queue (NQ) and User Queue (NU) to prevent transaction bottlenecks and keep response times minimal during peak processing windows.

5. Advanced Features & Modernization (Optional but High Value)

• Event Replication for Adabas: Knowing how to configure the Replicator Server to capture database updates directly from the PLOG and stream them to Kafka, IBM MQ, or relational data lakes.
• Adabas Encryption & Security: Implementing encryption at rest and in transit, alongside tying database access permissions to mainframe security managers like RACF, ACF2, or Top Secret.

The essential skills required for a Natural Adminstrator are structured below:

1. Configuration & Parameter Management (NATPARM)

• Customizing the Natural Parameter Module (NATPARM): This is the core skill. A NAT Admin must know how to modify, assemble, and link-edit the macro-based NATPARM module to control environment variables, memory limits, and product hooks.
• Session Parameter Optimization: Expert understanding of critical parameters like LS (Line Size), PS (Page Size), MADIO (Maximum Adabas I/O operations allowed per program), and MAXROP (Maximum number of nested loops).
• Buffer Pool Management: Skill in calculating, initializing, and managing the Natural Global Buffer Pool (GBP). You must monitor buffer pool efficiency using the SYSBPM utility to ensure frequently used programs reside in memory, reducing disk I/O.

2. Environment Integration & Interfaces

• TP Monitor Configuration (Com-plete): You must know how to configure the Natural interface for your specific Teleprocessing (TP) monitor. For Com-plete, this includes defining thread sizes, roll slots, and transaction codes.
• Database Interface Configuration: Ensuring Natural can securely communicate with Adabas, Db2, or VSAM. You must be able to manage database IDs (DBID) and file numbers (FNR) via the NTDB macro in NATPARM.
• Security Subsystem Integration (Natural Security / NSC): Deep knowledge of implementing Natural Security. You must know how to define users, groups, application libraries, and RPC services, and how to restrict access to command prompts, specific system files, or sensitive utilities.

3. Object Management & Deployment Utilities

• System File Administration (FNAT and FUSER): Natural isolates system objects (FNAT) from user/business applications (FUSER). You must manage space allocation on these files in tandem with the Adabas DBA.
• The INPL and SYSTRANS Utilities: Mastery of migrating Natural applications across lifecycles (Development → Test → Production). You must know how to extract components into transfer files and load them cleanly using batch or online utilities.
• Natural Studio / Single Point of Development (SPoD): Configuring and troubleshooting the remote development environment interface, allowing programmers to use a modern PC-based GUI to write code that executes directly on the z/OS mainframe.

4. Advanced Tooling & Maintenance

• Natural Remote Procedure Calls (RPC): Setting up and managing the Natural RPC environment (using Entire Net-Work). This allows Natural client programs on z/OS to execute subprograms on open systems, or vice-versa.
• Natural ISPF (NII) Administration: Customizing the user workspace shell, linking it to the Entire System Server (NPR) node tables, and ensuring proper mainframe dataset access permissions are enforced.
• System Utilities Mastery: Daily utilization of built-in Natural command utilities, including:
• SYSPROF: Profile parameter management.
• SYSMAIN: Cross-library object copying, moving, and deleting.
• SYSERR: Configuring centralized application error messages and codes.
• SYSBPM: Live monitoring and tuning of the Natural Buffer Pool.

The Natural Admin's Lifecycle Workflow

A Natural Administrator acts as the gatekeeper of the development ecosystem. Their structured workflow typically follows these operational pillars:

1. Initialize & Optimize: Build NATPARM, allocate the FNAT/FUSER libraries, and boot up the Global Buffer Pool.
2. Secure: Apply permissions in Natural Security (NSC) to ensure developers only touch their designated application profiles.
3. Connect: Configure interfaces to Com-plete (for online users), Adabas (for data), and Predict (for data definitions).
4. Deploy: Execute INPL batch jobs to roll out code updates to production cleanly without disrupting active online sessions.

The essential skills required for an Event Replication Adminstrator are structured below: 

To effectively manage real-time data streaming out of a z/OS mainframe, an Event Replication Administrator (often called a Replicator Admin) must have a specialized skill set. This role sits at the intersection of mainframe database administration, enterprise messaging infrastructure, and modern data-streaming architecture.

The essential skills required for an Event Replication Administrator include:

1. Replicator Server Architecture & Control

• Replication Configuration (Replication Definitions): Mastery of defining Global Options, Source Definitions (which Adabas files and fields to capture), and Destination Definitions (where to send the data).
• Subscription Management: Skill in creating and managing subscriptions that map specific source database files to designated target destinations.
• Initial State Processing: Expertise in performing data synchronization (Initial States). You must know how to extract bulk legacy data using utilities like ADALOD or internal replication functions to prime target databases without disrupting production performance.
• Filter and Transformation Logic: Ability to write and embed custom filter rules or event-driven user exits to clean, format, or drop specific data payloads before they leave the z/OS host.

2. Mainframe Infrastructure & Log Scraping

• Adabas PLOG/CLOG Integration: Deep understanding of how the Event Replicator interacts with the Adabas Protection Log (PLOG). You must monitor PLOG capture speeds to ensure the replication subsystem does not lag behind active online transactions.
• Replicator Engine Subsystem (ADARUN): Proficiency in configuring the dedicated Event Replicator engine address space using specific parameter cards (e.g., RPLSYSID, LRPL).
• Cross-Memory Operations: Knowledge of how data is passed between the source Adabas nucleus and the Replication Server via z/OS cross-memory communication services.

3. Messaging Middleware & Network Connectivity

• IBM MQ (WebSphere MQ) Configuration: Knowledge of mainframe MQ queues, message channels, and execution modules (ADAMQS). You must ensure message persistence, manage queue depths, and handle transaction commit structures.
• Entire Net-Work Linkage: Understanding how to route replicated packets across LPARs or open systems using Software AG's Entire Net-Work TCP/IP protocols.

4. Monitoring, Diagnostics, and Recovery

• RPLKEY and Queue Management: Monitoring internal replication queues to prevent memory exhaustion on the mainframe. You must know how to handle target-side outages where destinations stop responding.
• Data Recovery & Resubmission: Expert knowledge on how to re-stream missed or corrupted transaction intervals from archival PLOG tapes without causing duplicate key errors at the destination.
• Event Replicator Online (RPO): Daily utilization of the screen-driven RPO utility (executed via Natural) to monitor active message counts, modify destination states, and view system statistics in real time.

The Replicator Admin's Daily Routine

An Event Replication Admin continuously manages a critical loop of low-latency data:

1. Configure: Define exactly which fields in Adabas trigger an event.
2. Capture: Monitor the Adabas nucleus to ensure updates are instantly copied to the Replicator pool.
3. Transport: Route the payload cleanly through IBM MQ channels or Entire Net-Work drivers.
4. Audit: Use RPO to verify that 100% of mainframe updates successfully commit to the target streaming platform.