sp/ARCHITECT-BANK originally developed by The Software Partnership (TSP), Runcorn, Cheshire

The Software Partnership Logo

The core electronic banking software product sp/ARCHITECT-BANK was originally developed by The Software Partnership (TSP), a highly specialized British software house co-founded by Nigel Walsh in Runcorn, Cheshire.

Engineered to deliver high-availability, fault-tolerant electronic and desktop home-banking services, it ran natively on Tandem NonStop mainframe computers (now HPE NonStop).

The Software Partnership, Norton House, Crowngate, Runcorn, Cheshire

Over the decades, the product evolved through major corporate acquisitions, eventually being integrated into enterprise-level banking suites like CONNEX Advantage under eFunds and FIS.

The detailed timeline of the product, broken down by corporate era and year, is provided below by Mark Whitfield.

Click the previous link for more sp/ARCHITECT BANK project level detail between 1990 thru 1995.

Also, here is a LinkedIn group for the company Alumni.

🌅 Era 1: The Inception and Independent Software House Era (Mid-1980s–1993)

During this foundational era, The Software Partnership engineered the core product from scratch to meet the emerging demand for “Direct Electronic Banking” before the commercial internet became prevalent.

1985: The Software Partnership (TSP) is co-founded by Nigel Walsh in Runcorn, Cheshire. Development begins on a standard product architecture designed specifically for the transaction processing monitor (PATHWAY) and operating system (Guardian) of Tandem Computers.
1988–1989: The company establishes sp/ARCHITECT (and its core module, sp/ARCHITECT-BANK) as a premier client-server base package for corporate and home-office electronic banking.
1990: The engineering team scales up to build standard product releases written in COBOL85 and utilizing NonStop SQL databases. They develop proprietary testing utilities like sp/TESTBED to simulate PC-to-mainframe interfaces. Mark Whitfield joins the company after graduating in Computing in late 1990.
1991: Major deployment begins for the high-profile Barclays Business Master II (BBM II) desktop corporate banking application, with TSP placing teams (including Mark Whitfield) on-site at Barclays in Knutsford, Cheshire.

Barclays, Radbroke Hall, Knutsford, Cheshire

1992: A batch billing and invoicing suite of modules is engineered over 3-months and appended to the Barclays installation at Poole, Dorset. Mark Whitfield is assigned to this HPE NonStop (Tandem) billing/ invoicing development on the UK south coast. Simultaneously, TSP expands internationally into continental Europe.

1993: TSP develops an automated, touch-tone voice menu system for Girofon (Denmark). The code interfaces phone lines through Periphonics Interactive Voice Response (IVR) hardware directly into the back-end Tandem banking system. Concurrently, the core application handles desktop money transfers and early logic checking for clearing giants TSB and Bank of Scotland. Mark Whitfield is also involved with supporting this IVR technology.

🤝 Era 2: The Deluxe Data International Era (1994–1999)

Recognizing the massive European banking client footprints of sp/ARCHITECT, US-based electronic funds transfer (EFT) specialist Deluxe Data acquired TSP to merge their direct banking and card processing capabilities.

1994: Deluxe Data Corporation acquires The Software Partnership. The Runcorn offices are reorganised as Deluxe Data International Operations.

Deluxe Data International Operations, Wingate House, Northway

1995: The product undergoes heavy code optimization to satisfy customer acceptance loops for international clearers, notably deploying direct electronic banking solutions for major Dutch institutions like Rabobank. Mark Whitfield moves on from Deluxe Data (after 5 years) to Insider Technologies Limited in Salford Quays in late 1995. This to continue HPE NonStop programming work for both monitoring and diagnostic products like Reflex 80:20.
1996: Development transitions toward hybrid enterprise networking. The sp/ARCHITECT system is updated with custom TCP/IP software interfaces to allow newer mid-range UNIX servers (such as IBM RS/6000) to safely communicate with the core Tandem server environment.
1997: Deluxe Data expands the core platform’s messaging logic using Tandem’s Remote Server Call (RSC) facility. This enables early Windows NT operating systems to request live financial data from the sp/ARCHITECT host.

1998: An automated, multi-process file transfer protocol is integrated natively into the bank database, leveraging Connect:Direct transport layers to securely transfer corporate SWIFT financial data files.

🚀 Era 3: The eFunds & Corporate Consolidation Era (2000–2006)

Deluxe Data’s technologies spun off into a new corporate entity called eFunds Corporation, altering the delivery model of the legacy software.

2000: Deluxe Electronic Payment Systems officially merges with other divisions to form eFunds Corporation (EFD). The sp/ARCHITECT package becomes a core pillar of eFunds’ international banking portfolio.
2002–2004: To modernise the transaction handling backbone, components of the sp/ARCHITECT platform are refactored. The system’s underlying communication routing is systematically aligned with CONNEX, a dominant market-leading Electronic Funds Transfer (EFT) processing engine.
2005–2006: eFunds transitions the direct client-server software layers into highly secure corporate portals, providing the foundational logic for what would eventually be rebranded as the CONNEX Advantage banking solution.

🏢 Era 4: The FIS Integration and Legacy Modernisation Era (2007–Present)

The final stage of the product timeline represents its absorption into global banking infrastructure software, where its high-availability DNA remains active in institutional transaction environments.

2007: Financial technology behemoth Fidelity National Information Services (FIS) acquires eFunds Corporation for approximately $1.8 billion. Following industry consolidation, the corporate remnants of the original TSP Runcorn operations are absorbed into Fidelity National Information Services (FIS) and relocated to Aegon House in Daresbury, Warrington.

Fidelity National Information Services (FIS) Aegon House in Daresbury, Warrington 2007 — Fidelity National Information Services (FIS) Aegon House, Warrington (in 2007)

2010: FIS fully absorbs the remaining codebase, utilizing its core Tandem architecture algorithms to fortify transaction processing stability.
2015–2020: The architectural concepts pioneered by sp/ARCHITECT-BANK continue to govern high-volume legacy systems. The logic stays preserved in COBOL85 code bases running on modern HPE Integrity NonStop (Intel Xeon-based) fault-tolerant environments.
2020s–Present: Modern banking infrastructures gradually migrate from the classic database frameworks toward microservice configurations and open-banking APIs. However, the core system layout remains a primary point of historical reference for designing high-throughput, 24/7/365 fault-tolerant banking systems.

sp/ARCHITECT-BANK originally developed by The Software Partnership (TSP), Runcorn, Cheshire

sp/ARCHITECT-BANK Code Evolution Timeline

The timeline below details how the code’s core design, language implementations, and application deployment strategies transformed by era and year.

1. The Monolithic & TAL Foundation Era (1980s – Early 1990s)

During this era, the application focus was strictly high-throughput, fault-tolerant electronic funds transfer (EFT) and point-of-sale (POS) switching systems natively built for Tandem Guardian environments.

Late 1980s: The core design of sp/ARCHITECT is established using TAL (Tandem Application Language). Applications are deployed as single-system monoliths. Code optimization focuses heavily on low-level bit manipulation and message structuring to survive CPU or inter-process failures without losing in-flight transactions.
1991–1993: Structuring of modular execution libraries. Early iterations of the codebase segment transaction processing routes from core database logging routines. The introduction of Tandem’s newer NonStop SQL forces early integration layers to transition from standard unstructured unstructured file systems (Enscribe) to early relational tracking.

2. Distributed Client/Server & pTAL Migration Era (Mid 1990s – Early 2000s)

The architectural demands shifted from single-frame monoliths toward distributed banking systems, giving rise to “Distributed Monoliths” and client/server network structures.

1995–1996: Hardware evolutions transition from the older CISC-based Tandem systems to RISC architectures (MIPS processors). sp/ARCHITECT undergoes a massive compilation shift to pTAL (portable TAL) to preserve legacy code performance across new instruction sets.
1998–1999: Tandem’s acquisition by Compaq pushes the software suite to handle open standard protocols. The application code begins abstracting system calls to prepare for broader networking interfaces.
2001–2003: Deluxe Data / eFunds eras. The code sees the introduction of C/C++ wrappers around the legacy pTAL components. Systems are decoupled into a clear 3-Tier architecture: front-end terminal networks, back-end pTAL transactional engines, and standardized clearing houses.

3. Open Systems, Modern Middleware, & Java Integration Era (Mid 2000s – 2010s)

Following HP’s acquisition of Compaq and subsequent software realignments, the sp/ARCHITECT codebase was re-engineered to prevent vendor lock-in and adopt modern enterprise standards.

2005–2007: Java is introduced into the sp/ARCHITECT ecosystem. New application modules, specifically merchant portal interfaces and settlement reporting tools, are written entirely in Java and run via OSS (Open System Services) environments.
2010–2012: FIS acquisition era integration. Legacy pTAL code blocks are systematically refactored or heavily wrapped in C++ using object-oriented principles to ensure long-term maintenance. The transaction routing engine is altered to support early SOA (Service-Oriented Architecture) paradigms via web-services hooks.
2015–2018: Mainstream deployment of COB (Core Banking) standard formats within the application layer. The system moves away from old proprietary network messaging layouts to ISO 20022 compliance frameworks, utilizing dedicated conversion engines native to the sp/ARCHITECT stack.

4. Modern Cloud-Adjacent & Hybrid Infrastructure Era (2020s)

The current evolutionary footprint centers on maintaining the absolute sub-millisecond reliability of the core architecture while exposing capabilities to dynamic cloud endpoints.

2021–2023: Modernization of the application payload. High-performance micro-frontends handle real-time fraud monitoring and data streaming using asynchronous event-driven pipelines (e.g., Kafka event consumers interfacing directly with the NonStop core runtime environments).
2024–2026: Transition to containerized orchestration and cloud-adjacent infrastructure. The sp/ARCHITECT footprint utilizes x86-based virtualized NonStop systems (NSX), enabling legacy core modules (derived from the original TAL logic) to execute seamlessly on modern virtual environments alongside Linux-based multi-tenant applications.

Insider Technologies Limited (ITL), Company Overview and Timeline by Year

ITL is a specialized UK-based software house & service provider that engineers high-availability monitoring, tracking, and cybersecurity solutions

Insider Technologies Limited is a specialized UK-based software house and service provider that engineers high-availability monitoring, tracking, and cybersecurity solutions for business-critical, 24/7 mission-critical architectures. I worked at ITL in Salford Quays from 1995 through to 2013.

Mark Whitfield attending an EBUG conference
(BASE24 User Group) — Attending an EBUG conference
(European BASE24 User Group)

Foreground, attending a British Isles TANDEM User Group (BITUG) — Foreground, attending a British Isles
TANDEM User Group (BITUG)

Insider Technologies Limited, ITL, website in 2009 — Insider Technologies Limited
(website author in 2009)

Company Overview

🏢 Corporate Identity & Status

Legal Name: Insider Technologies Limited
Founded: 27 February 1989
Headquarters: Manchester, UK (Albert Street, Eccles)
Ownership: Operating as a private independent software company, recently integrated as part of PartnerOne.
Strategic Partnerships: Certified Microsoft Gold Partner for Application Development and long-standing Hewlett Packard Enterprise (HPE) partner.

Insider Technologies Limited (ITL) in 2001, Salford Quays, Chandlers Point

🌐 Core Domain & Industries Served

The company delivers real-time, event-driven diagnostic, tracking, and compliance middleware solutions across three main highly-regulated global verticals:

Banking & Financial Services: Monitoring real-time electronic payments, ATM/POS systems, and transaction workflows (e.g., BASE24, XPNET environments).
Government & Defence: Secure enterprise military messaging tracking and digital evidence handling.
Telecommunications: Tracking data traffic and critical infrastructure middleware.

🛠️ Core Technology Stack & Competencies

Insider Technologies specializes in niche high-availability operating environments—specifically HPE NonStop (historically Tandem Computers and HP NonStop) running Guardian and Open System Services (OSS) environments—alongside enterprise Windows, Linux, and Unix systems. Their expertise spans database transaction management, BASE24 XPNET monitoring, IBM WebSphere MQ tracking, and low-level development (SQL, TAL, TACL, COBOL85, C++, Pathway).

Insider Technologies, the
Core products in 2003 — Insider Technologies –
Core products in 2003

In-Depth Product & Political-Technical Timeline

This timeline breaks down how Insider Technologies evolved its software suite. It demonstrates how their technical development directly responded to shifting geopolitical landscapes—ranging from late-Cold War military messaging security to post-9/11 financial regulations and modern European cloud autonomy initiatives.

🔹 The Foundation & Legacy Tandem Era (1989–1999)

Political Context: The final years of the Cold War and the 1990s globalization boom demanded high-security, fault-tolerant mainframes for NATO-aligned military communications and early global banking clearing networks.
Technical Context: Tandem Computers dominated the un-interruptible 24/7 market. Software was required to monitor these platforms without causing processing overhead.
1989
- Company Incorporation: Insider Technologies is incorporated in the UK to engineer bespoke software for highly robust technical ecosystems.
1990–1995
- Reflex (Core Release): Release of Reflex, a foundational service management and real-time transaction diagnostic tool built specifically for the Tandem Guardian operating system.
1996–1999
- MultiBatch Software: Further Develop and Extend MultiBatch to orchestrate and safely automate complex batch processing on Tandem machines alongside the evolution into HP NonStop computing frameworks.
- Y2K Compliance Focus: Technical adjustments were deployed across Reflex and MultiBatch to assure financial institutions that automated transaction logging would not fail during the millennium rollover.

🔸 The Multi-Platform & Financial Compliance Era (2000–2015)

Political Context: Following the September 11 attacks, global anti-money laundering (AML) and counter-terrorism financing (CTF) frameworks heavily expanded. Financial regulators demanded exact, audible end-to-end payment tracking.
Technical Context: Enterprises began migrating away from single-architecture mainframes toward heterogeneous IT environments, requiring tools that could jump across Windows, Unix, and Linux simultaneously.

2002–2004
- Sentra Development: Launch of Sentra, expanding the firm’s portfolio beyond HPE NonStop into cross-platform environment monitoring for Windows, Linux, and Unix systems.
- Reflex 80:20 & Reflex ONE24: Advanced variations of the Reflex tracking system were built to cater to specialized real-time electronic payment flows like BASE24 with XPNET.
2005–2010
- RTLX Reactor (page 12) and Middleware Monitoring: The release of RTLX Reactor provided message tracking capabilities tailored for IBM WebSphere MQ, allowing institutions to trace various payment and other data through complex middleware chains.

RTLX Reactor, 2013 article -
The Connection Journal — RTLX Reactor, 2013 article –
The Connection Journal

BASE24-eps extraction and RTLX — BASE24-eps extraction
and RTLX (in 2007)

2011–2015
- Corporate Structuring: The creation of Insider Technologies (Holdings) Limited reinforced corporate expansion as the company deepened its footprint in secure military messaging for government defense bodies.

🔹 The Cyber Autonomy & Sovereignty Era (2016–Present)

Political Context: Escalating nation-state cyber warfare, strict GDPR regulations, and the UK/EU push for technological sovereignty and domestic digital ecosystem resilience heightened the reliance on trusted, onshore technology suppliers.
Technical Context: High-threat environments demand zero-trust visualization, time-accurate logging across digital evidence files, and advanced protection against internal exfiltration vectors.
2019
- 30-Year Milestone & Modernization: The company celebrates its 30th year, accelerating development on modern mobile application extensions to permit real-time, remote secure alerts for operations teams.
2022–2025
- PartnerOne Integration: Insider Technologies aligns its operations under the PartnerOne portfolio, preserving its UK identity while supercharging its enterprise-scale data infrastructure solutions.
- Corporate Integration combines Insider’s expertise with PartnerOne’s portfolio to deliver advanced analytics and secure messaging systems to banking, defence, and telecommunications markets.
- Defense and Public Safety Porting: Technical deployment of specialized capabilities covering digital evidence security, timekeeping tracking, and legacy virtualization modules aimed explicitly at helping the UK Government maintain its historical tech stacks safely.

RTLX Reactor (in 2012) for tracking BASE24-eps and BASE24 XPNET transactions — RTLX Reactor (in 2012) for tracking
BASE24-eps & BASE24 XPNET transactions

Insider Technologies Limited (ITL), Company Overview and Timeline by Year

The Insider RTLX product at ETI-NET is now called C-Deep for Transaction Monitoring;

C-Deep (Transaction Monitoring)

The Insider Reflex product at ETI-NET is now called Sentinel for NonStop Monitoring;

Sentinel (Nonstop Monitoring)

C-Deep for Transaction Monitoring & Sentinel for HPE NonStop Monitoring

Mark Whitfield – Education and Professional Training Timeline Summary

Mark Whitfield is a Greater Manchester-based Senior IT Project and Engagement Manager.

With over 30 years in the IT and software development industry, he has continuously upskilled in project delivery, Agile methodologies, cloud platforms, and cyber security.

Mark Whitfield - Education and Professional Training Timeline Summary

Phase 1: Foundational Education

1985 – 1988: Leigh College, UK
- Focus: Computer Science and Biology (‘A’ Levels)
1988 – 1990: University of Greater Manchester (formerly Bolton Institute of Higher Education, BIHE)
- Focus: Higher National Diploma (HND) in Computer Studies (Graduated with Distinction; First overall in the year)
- Key Modules: System Analysis, Programming Methodology, Database Architecture, and Business Information Systems

Phase 2: Project Management & Professional Training

2000 – 2006: Industry Integration & Early Methodologies
- Focus: Service-Oriented Architecture (SOA), Agile, Sales, and early project management
- Courses/Certifications: Sales and Marketing (In-house Outsource, 2001), Web Services and SOA (Insider Technologies, 2005), PRINCE2 Foundation (2006), Designing Good Marketing Literature (SkillPath Seminars, 2006)
2009: Digital & Communications
- Focus: Digital marketing and content
- Courses: Writing for the Web, and Website Promotion and Visibility by Design (iTrain Education)
2011: Structured Frameworks
- Focus: Formal project frameworks and delivery methodologies
- Courses/Certifications: PRINCE2 Foundation & Practitioner (ILX Group), Agile Scrum (RADTAC)
2012: Operational Management
- Focus: Service management best practices
- Courses/Certifications: ITIL Foundation

Phase 3: Advanced Engagement & Enterprise Training

2017 – 2019: Capgemini Engagement & Compliance
- Focus: High-level engagement management and corporate governance
- Courses/Certifications: Advanced Engagement Management Certification (Level 2), Group Anti-Corruption, and Intellectual Property Rights (IPR) Training
2022: Cloud Modernization
- Focus: Enterprise cloud computing fundamentals
- Courses/Certifications: AZ-900 Microsoft Certified Azure Fundamentals

For more granular details on his certifications and career history, you can check Mark Whitfield Professional Training.

BASE24 Electronic Funds Transfer (EFT) software application developed by ACI Worldwide, Overview

BASE24 is a market-leading, fault-tolerant Electronic Funds Transfer (EFT) software application developed by ACI Worldwide. For decades, it has served as the backbone for global banking, processing billions of ATM, Point of Sale (POS), and smart card transactions.

The product achieves its landmark 24/7/365 uptime by running natively on the HPE NonStop architecture—originally engineered by Tandem Computers.

1. Underlying Technology Stack

BASE24 Classic was built from the ground up to utilize the unique properties of the Tandem/HPE NonStop platform:

Operating System: HPE NonStop Kernel (NSK) / Guardian.
Database: Enscribe, a native hierarchical/flat file database optimized for ultra-fast, unstructured file access. Newer iterations use NonStop SQL/MX.
Programming Languages: Primarily TAL (Tandem Application Language), pTAL, and COBOL/SCOBOL.
Middleware: PATHWAY (PATHCOM), which acts as the transaction processing monitor to dynamically manage and load-balance server processes.

2. High-Level Component Architecture

BASE24 relies on an interconnected network of specialized processes that route and manage messages.

A. XPNET (The Networking Engine)

XPNET is a critical, proprietary communication subsystem. It provides the messaging infrastructure where applications interface with network communication lines. XPNET acts as the buffer layer, monitoring physical lines, enforcing transaction timing checks, and distributing data loads uniformly across CPUs.

B. Device Handlers (DH)

Device Handlers act as the translators for peripheral devices.

Function: They intercept hardware-specific protocol messages (e.g., Diebold or NCR formats from ATMs) and normalize them into BASE24’s internal standard message format.
Security: DH processes handle terminal-level PIN encryption, coordinate MAC (Message Authentication Code) keys, and initiate terminal downline loads.

C. Authorization Process (AUTH)

AUTH is the core decision engine of the application.

Function: It validates card restrictions, tracks card usage accumulations, and performs transaction risk checks.
Fallback Management: If a bank’s core system goes offline, AUTH drops into “Stand-Alone” or “Negative/Parametric Authorization” mode, approving transactions locally up to safe, pre-defined limits.

D. Host Interfaces (HI)

The Host Interface connects BASE24 to the financial institution’s primary backend core banking systems. It handles “On-Us” transactions—meaning the card used belongs to the bank owning the terminal.

E. Interchange Interfaces (II)

The Interchange Interface formats, translates, and routes transactions to global credit/debit networks (such as Visa, Mastercard, AMEX) or regional switches. It transforms internal BASE24 data formats into compliance standard formatting, such as ISO 8583. It handles “Not-On-Us” transactions.

3. Core Database & File Structure

BASE24 captures system activities across specialized transactional and tracking files, mostly utilizing Enscribe:

TLF (Transaction Log File): The primary log capturing every ATM event, amount, response code, and terminal ID in real-time.
PTLF (POS Transaction Log File): Mirrors the utility of the TLF, but optimizes records strictly for merchant POS transactions.
LCONF (Logical Network Configuration File): Dictates how network configurations, devices, institutions, and communication paths map into XPNET.
CAF (Cardholder Authorization File): Stores specific card numbers, limits, and statuses used for stand-alone authorization if host links break down.

4. Daily Operational Processes

Beyond live message switching, BASE24 executes several critical back-office operations:

Extract: Periodically filters transaction data from live TLF/PTLF logs to move to external billing arrays.
Refresh: Downloads updated data dumps (such as blacklisted cards or updated balances) from core hosts into local BASE24 database files.
Settlement Initiator: Aggregates transaction volumes at specified cutoff times to reconcile balanced records between ATMs, POS terminals, and clearing networks.

5. Why Tandem/HPE NonStop is Essential to BASE24

BASE24 relies on the hardware/software synergy provided by HPE NonStop to achieve near-zero downtime:

Shared-Nothing Architecture: Processors operate independently with their own memory stacks. If a physical CPU suffers hardware failure, it cannot corrupt the rest of the application.
Process Pairs: BASE24 components operate via a primary process in one CPU and a backup process in an alternate CPU. The primary constantly syncs checkpoint data with its backup. If the primary drops, the backup assumes processing instantly without interrupting transaction flights.
Active/Active Configuration: Utilizing replication software like HPE Shadowbase or DRNet, financial firms link distinct geographic NonStop locations. Both processing sites operate concurrently, managing localized transactions and replicating states reciprocally.

6. Product Evolution: BASE24 Classic vs. BASE24-eps

ACI Worldwide evolved the platform from BASE24 Classic into BASE24-eps (Enterprise Payment System):

BASE24 Electronic Funds Transfer (EFT) software application developed by ACI Worldwide, Overview

HPE NonStop Tandem Architecture Walkthrough

The HPE NonStop architecture (originally engineered by Tandem Computers in 1976) is a specialized, 100% fault-tolerant computing platform designed to achieve continuous application availability and absolute data integrity. Unlike traditional mainframes or high-availability clusters that rely on rapid rebooting or switching resources upon a crash, NonStop prevents downtime entirely by masking failures through a hardware-software co-designed shared-nothing architecture.

1. Hardware Architecture: Massively Parallel & Shared-Nothing

At the physical tier, a NonStop system is built as a Loosely Coupled Multiprocessing (LCM) environment.

Independent Processor Modules: A single system consists of 2 to 16 independent CPUs (expandable via clustering up to 4,000+ CPUs). Each processor module contains its own dedicated Intel Xeon cores, memory, and I/O logic. Processors share no main memory, buses, or execution states. This isolation guarantees that a memory corruption or hardware crash in one CPU cannot physically propagate to another.
The Interconnect Fabric (ServerNet / RoCE): Because CPUs share nothing, they cooperate entirely by passing high-speed messages. Historically, this handled via a proprietary dual-bus named Dynabus, which evolved into ServerNet (the foundational grandfather of InfiniBand). Modern HPE NonStop X systems leverage RDMA over Converged Ethernet (RoCE) as the multi-gigabit interconnect fabric, providing dual-path, point-to-point messaging with sub-microsecond latency.
Dual-Ported, Redundant I/O Controllers: Every storage device, network interface, and controller card is physically dual-ported and cross-connected to two separate processor modules. If Processor A fails, Processor B seamlessly accesses the disk or network line using the alternate hardware path.
No-Spare, Active-Active Components: Every active element operates under a “no-spare” philosophy. Power supplies, cooling fans, and storage arrays are fully redundant and hot-swappable, ensuring the system can be repaired or upgraded while fully operational.

2. Operating System Architecture: NonStop OS (Guardian)

The foundational operating system is NonStop OS, which embeds the Guardian Kernel.

Distributed Copy Model: Every individual processor module loads and runs its own separate copy of the Guardian kernel. Rather than a monolithic OS orchestrating all chips, the system runs as a highly cooperative, message-driven distributed microkernel OS.
The Message System: The core of Guardian is its message router. Every operational request—whether writing a line to a database, opening a network socket, or checking a disk—is written as an inter-process message sent across the RoCE fabric. If a local resource is occupied, the message router redirects the request transparently across the fabric, making the entire cluster appear to applications as a single system image (SSI).
Continuous Heartbeats: All components and processors continually broadcast periodic “alive” heartbeat messages to one another. If a processor fails to respond to a heartbeat within a few milliseconds, the remaining CPUs immediately sever ties with it, declare it dead, and safely re-route pending workloads.

3. Software Fault Tolerance: Process Pairing

Hardware isolation is only half the battle. To tolerate software failures without dropping transactions, NonStop utilizes Process Pairs.

Primary and Backup Processes: When a critical application or system service starts, it creates two instances: a Primary Process executing on Processor 1, and a Hot-Standby Backup Process residing on Processor 2.
Real-Time Checkpointing: As the primary process performs work (e.g., executing a financial transaction step), it sends regular checkpoint messages to the backup process. These checkpoints copy vital state changes, register values, and memory updates.
Instant Takeover: If Processor 1 crashes, the Guardian OS instantly promotes the backup process to Primary. Because the backup contains the mirror state of the last transaction checkpoint, it picks up execution precisely where the failed process stopped. No state is lost, no connections drop, and the end-user experiences zero interruption.

4. Database & Storage Architecture: Enscribe, NonStop SQL, and TMF

Data integrity is paramount in NonStop’s design. It enforces strict ACID compliance at massive scale through layered data management software.

Enscribe & NonStop SQL/MX: NonStop supports Enscribe (a highly resilient structured file system) and NonStop SQL/MX (an ANSI-compliant relational database management system). Both are entirely decentralized, natively distributing table partitions across different physical disk drives managed by separate CPUs.
Mirrored Disks: Storage volumes are configured via volume-level mirroring (Disk 1 and Disk 2 track identical data blocks). Disk writes are executed in parallel across distinct I/O paths. If a drive fails or a sector corrupts, reads are immediately diverted to the mirror disc.
Transaction Monitoring Facility (TMF): TMF is the protected transaction manager. It acts as a distributed two-phase commit coordinator. If an application crashes mid-transaction, or an entire processing module loses power, TMF uses audit logs to back out incomplete transactions cleanly, guaranteeing that the database is never left in an inconsistent or corrupt state.

HPE NonStop Tandem Programming Languages and Development Timeline

The development of programming languages on the HPE NonStop platform (originally founded as Tandem Computers) is tightly bound to its architectural hardware transitions: from custom CISC stack machines to MIPS RISC, Intel Itanium, and eventually standard Intel x86-64 infrastructures.

Detailed List of NonStop Programming Languages

1. Core Proprietary & System Languages

TAL (Transaction Application Language): The foundational system programming language for Tandem. It is a block-structured, machine-dependent procedural language designed to compile directly into highly efficient machine instructions. It features ALGOL/Pascal-like syntax but implements C-like semantics, structural pointers, and weak data typing.
pTAL & epTAL: Specialized evolutions of TAL. Rather than rewriting legacy codebases from scratch during architecture shifts, pTAL was introduced to compile existing TAL code natively into MIPS RISC architectures. Later, epTAL was developed to target Intel Itanium microprocessors.
TACL (Tandem Advanced Command Language): A built-in command interpreter and interpreted scripting language. It functions like a Unix Bash shell but features highly complex macro capacities used to orchestrate system configurations, monitor processes, and automate failover procedures.

2. Enterprise & Enterprise Legacy Languages

COBOL85 (and older COBOL74): The undisputed workhorse of NonStop commercial workloads. HPE’s tailored implementation of the COBOL85 standard natively interfaces with the Guardian OS. It allows programmers to embed SQL/MP statements and program fault-tolerant Process Pairs through HPE NonStop Pathway (TS/MP).
SCOBOL (Screen COBOL): A specialized, high-level structural derivative of COBOL utilized exclusively to build blocks for character-cell terminal interfaces (such as the 6530 terminal environments) running within Pathway architectures.
NonStop SQL (SQL/MP and SQL/MX): While technically a database system, its embedded syntax acts as a declarative language integrated into C and COBOL. SQL/MP works with the legacy Guardian file system, while SQL/MX brings ANSI-compliant SQL closely bound with the Open System Services (OSS) environment.

3. Standard Mainstream Languages

C & C++: Heavily introduced during the RISC transition to allow software portability. Mainstream development on modern NonStop systems uses standard C/C++ cross-compilers. They run in either the native fault-tolerant Guardian personality or the standard POSIX-compliant Open System Services (OSS) environment.
Java: A first-class language layer deployed natively on NonStop. HPE optimizes the Java Virtual Machine (JVM) to scale across multi-CPU shared-nothing frameworks, allowing modern enterprise web apps to run with out-of-the-box system availability.

4. Modern Open-Source Options

Python, Go, & JavaScript (Node.js): Modern procedural and script utilities provided by HPE. These environments leverage the OSS POSIX platform layer, running modern DevOps orchestration, microservices, and hybrid-cloud pipelines alongside the native database engines.

Detailed Timeline Breakdown by Era and Year

The evolution of NonStop languages maps directly across distinct engineering ownership eras.

The Proprietary Foundation Era (Tandem Computers: 1974–1989)

1976: Tandem ships the original Tandem/16 (NonStop I). TAL is the only available language on the platform. The entire Guardian Operating System is written completely in TAL.
1981: The NonStop II hardware is introduced. Tandem expands language support to include COBOL74, FORTRAN, and BASIC to attract mainstream banking clients.
1983: Tandem releases the Transaction Monitoring Facility (TMF) and Pathway application management software. SCOBOL is introduced alongside them to program secure terminal entry interfaces.
1985: TACL is deployed, completely modernizing the command line shell ecosystem with scalable macros and structured operational control.
1986: Tandem launches NonStop SQL, the first linearly scalable, fault-tolerant relational database engine. Embedded SQL syntax is integrated directly into TAL and COBOL compilers.
1988: Compilers undergo a major update to natively support the newly established COBOL85 standard, which quickly replaces COBOL74 for all mission-critical banking transactions.

The Open Systems & Hardware Transition Era (Compaq: 1990–2001)

1991: Hardware migrates from CISC stacks to MIPS RISC architectures with systems like the Cyclone/R. To protect client software assets, Tandem delivers the pTAL compiler to translate TAL source code into native RISC binaries.
1995: Tandem introduces Open System Services (OSS), a POSIX-compliant UNIX subsystem running over the Guardian kernel. This brings full-scale, native native compliance for standard ANSI C and C++ programming.
1997: Compaq acquires Tandem Computers. Engineering shifts heavily toward implementing Java on NonStop, targeting cross-platform, enterprise internet-banking codebases.
2000: NonStop SQL/MX is released. It allows developers to use embedded SQL statements within standard C, C++, and emerging Java applications inside the OSS runtime environment.

The Corporate Alignment & Itanium Era (Hewlett-Packard: 2002–2014)

2002: HP merges with Compaq. Java is designated as a first-class citizen on the platform, receiving deeper optimization to tie into native clusters seamlessly.
2005: HP releases the Integrity “NonStop i” servers, moving processors away from MIPS onto Intel Itanium architectures. The epTAL compiler is rolled out alongside standard C/C++ updates to seamlessly compile older environments onto Itanium.
2011: Open-source scripting engines, including early ports of modern Python, are introduced to the OSS environment, easing the system-management burden for engineers unfamiliar with legacy TACL.

The Modern Enterprise Era (Hewlett Packard Enterprise: 2015–2026)

2015: HP splits, and the platform transitions to HPE. Standard Intel x86-64 hardware dominates with the NonStop X architecture. Compilers utilize an standard GCC/LLVM-based back end, allowing normal Linux/Unix C++ programs to build on NonStop with minimal alteration.
2020: Sales of Itanium systems officially terminate. Legacy languages like TAL are deprecated for new software creation but are preserved to support older, foundational logic.
2023–2024: HPE rolls out modern cloud-ready DevOps Starter Kits. Full, native support is added for modern languages such as Go, modern Python 3.x, and Node.js, allowing them to integrate into modern automated CI/CD build environments.

HPE NonStop Tandem Programming Languages and Development Timeline

HPE NonStop Pathway is a transaction processing & application server environment (TS/MP)

HPE NonStop Pathway is a premier transaction processing and application server environment (TS/MP) that powers mission-critical Online Transaction Processing (OLTP). It handles critical application services—such as fault tolerance, load balancing, memory management, and process scheduling—automatically, allowing developers to focus strictly on business logic.

Detailed Timeline Breakdown

The history and evolution of the Tandem NonStop platform and its Pathway environment span decades of architectural transformations and corporate ownership, categorized by distinct hardware and software eras:

1. The Tandem Era (1974–1997)

1974: Tandem Computers Inc. is founded by Jimmy Treybig to build the first fault-tolerant commercial hardware.
1976: The first Tandem NonStop system (NSI) is launched. Early apps had to be manually coded for fault tolerance.
1981: NonStop II is released, bringing 32-bit addressing.
1983: The Transaction Monitoring Facility (TMF) is introduced. Together with the launch of the Pathway transaction management software, the need for programmers to write manual fault-tolerance logic into their code is officially eliminated.
1986: Tandem releases the EXT as an entry-level system, followed by the VLX.
1991: Tandem introduces the Cyclone/R and initiates a massive architectural shift away from proprietary stack machines towards MIPS RISC processors.
1997: Compaq acquires Tandem Computers, placing the NonStop product line under its umbrella.

2. The Compaq & Early HP Era (1997–2014)

2001–2002: Hewlett-Packard (HP) merges with Compaq. The platform is rebranded as HP NonStop.
2005: The HP Integrity NonStop (TNS/E) series is introduced, migrating the fault-tolerant platform to Intel Itanium microprocessors. Pathway continues to be the main driver for high-volume banking and telecom applications.
2011: Further hardware advancements lead to the release of HP Integrity NonStop BladeSystems.

3. The Modern HPE Era (2015–Present)

2015: Hewlett-Packard splits, and the NonStop environment transitions to Hewlett Packard Enterprise (HPE).
2015/2016: Introduction of NonStop X (TNS/X) systems, marking the platform’s migration to standard Intel x86-64 processors and adopting InfiniBand interconnects. Pathway capabilities are updated to span dynamic server classes across multiple systems (Pathway Domains).
Present: HPE continues to modernize the NonStop architecture, integrating the platform with HPE GreenLake for consumption-based models and providing native support for modern DevOps tools and hybrid cloud deployments.

HPE NonStop Tandem EMS Subsystem Overview and Chronological Timeline

In the HPE NonStop ecosystem, EMS (Event Management Service) is the core software subsystem responsible for collecting, formatting, filtering, logging, and routing system and application event messages. It provides fault-tolerant monitoring by gathering data from EMS collectors and selectively delivering alerts to consoles, log files, or automated management applications.

EMS events as viewed in Console, Reflex 80:20 event viewer (ITL) — EMS event detail as viewed in Console, **Reflex 80:20 event viewer (ITL)**

EMS event view configuration window in Console — EMS event view configuration window in **Console**

The evolution and detailed historical timeline of NonStop EMS events and architecture is broken down by era below:

1. The Tandem Guardian Era (Late 1970s – 1980s)

Focus: Foundation of Fault-Tolerant Event Logging

1976: Tandem releases the original Tandem/16 (NonStop I) system. Early event handling was primarily a rudimentary terminal console logging process.
1978: System administrators struggled with message scaling as clusters and terminal networks expanded. Tandem began developing structured event tracking, paving the way for standardized subsystem messages.
1980s: Introduction of early message formatting. Event messages 1 through 511 were reserved for unformatted, raw console events. The Event Management Service (EMS) was gradually formalized to centralize scattered terminal messages.

2. The D-Series & TMF Era (1990s)

Focus: Distributed Management & The Birth of Modern EMS

1991: Tandem releases the Cyclone/R (CLX/R) and later the Himalaya K-series using MIPS processors.
1993: The publication of the seminal EMS Reference Summaries standardized EMS APIs and SPI (System Programming Interface). Event IDs were structured into standardized subsystems (e.g., negative-numbered kernel messages).
1995: The NonStop Kernel introduced Open System Services (OSS), natively integrating Unix-like event logs into the Guardian architecture.
1997: Compaq acquires Tandem Computers. EMS underwent significant rewrites to interface with remote servers and client networks. The S-Series was launched utilizing ServerNet fabric, introducing advanced, distributed event collectors and distributors.

3. The Compaq Transition & HP Integration (2000s)

Focus: Web-Based Management and Automation

2000 – 2003: Legacy ViewPoint tools were expanded. The emergence of GUI interfaces and DSM/PM (Distributed Systems Management/Performance Monitor) allowed operators to browse and filter EMS logs on alternate/primary event files.
2003 – 2005: The transition to Web ViewPoint commenced, turning text-based EMS event logs into interactive, web-based graphical operations interfaces.
2006 – 2009: With HP fully in charge after merging with Compaq, EMS event viewing was modernized through TSM (Tandem/HP Systems Management) and the Open System Management (OSM) Event Viewer.

4. The Modern HPE Era (2010s – 2026)

Focus: Cloud Integration, Virtualization, & Real-Time Analytics

2015 – 2017: The platform is rebranded as HPE NonStop as the architecture migrates to x86 processors. EMS systems are upgraded to handle large datasets, feeding complex event processing (CEP) and SNMP trap frameworks for modern data centers.
2018 – 2023: HPE integrates NonStop systems with HPE GreenLake. EMS event logging is modernized with API-driven integrations, allowing system events to be consumed by off-platform enterprise loggers, Splunk, and cloud-management consoles.
2024 – 2026: EMS events operate in highly virtualized and hybrid cloud (x86 and Virtual NonStop) environments. Event management heavily relies on modern distributed systems where EMS distributors push logs seamlessly into centralized IT monitoring suites and continuous availability dashboards.

To look up specific system-generated event codes or statuses from any era, consult the legacy ⁠HP NonStop Operator Messages Manual or the broader ⁠HPE Nonstop Compute documentation portal.

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this:

Share this: