Neill Riordan’s Post

Check out this latest post from the Power Automate team #KPMGMBS

New article from the Power Automate Blog

𝐑𝐞𝐚𝐝𝐲 𝐭𝐨 𝐮𝐧𝐥𝐨𝐜𝐤 𝐭𝐡𝐞 𝐟𝐮𝐥𝐥 𝐩𝐨𝐭𝐞𝐧𝐭𝐢𝐚𝐥 𝐨𝐟 𝐚𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐨𝐧? 🚀 In today’s digital landscape, open APIs are the backbone of full automation, enabling seamless data flow and real-time updates. At Oliasoft, we recognize their pivotal role, especially in drilling and well planning. Here's why: • 𝐃𝐢𝐠𝐢𝐭𝐚𝐥𝐢𝐳𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐖𝐨𝐫𝐤𝐟𝐥𝐨𝐰𝐬: Say goodbye to cumbersome workflows! Our APIs empower the transition to digital processes, slashing costs and liberating operations from legacy software limitations. • 𝐑𝐞𝐚𝐥-𝐓𝐢𝐦𝐞 𝐃𝐚𝐭𝐚 𝐔𝐩𝐝𝐚𝐭𝐞𝐬 𝐚𝐧𝐝 𝐂𝐚𝐥𝐜𝐮𝐥𝐚𝐭𝐢𝐨𝐧𝐬: Stay ahead with up-to-the-minute data and calculations, ensuring precision and agility in your operations. • 𝐈𝐧𝐭𝐞𝐠𝐫𝐚𝐭𝐢𝐨𝐧𝐬 𝐀𝐜𝐫𝐨𝐬𝐬 𝐒𝐲𝐬𝐭𝐞𝐦𝐬: Break down silos! Seamlessly integrate diverse systems, fostering collaboration and data exchange across platforms using open industry standards. At Oliasoft, open APIs aren't an add-on – they're standard across all our products. Experience seamless integration into your workflow, whether it's proprietary software or a third-party platform. Our commitment to integration grants you unparalleled flexibility to tailor your digital workflows today and in the future. https://lnkd.in/d6u6s2Zy

Check out this latest post from the Power Automate team #KPMGMBS

🔍 Simplifying Orphaned Flow Management in Power Automate! 🔍 💁♂️Managing orphaned flows in Power Automate just got easier. 🪶New features now allow admins to identify and reassign flows that were left without owners when employees departed. ✳️Through the Power Platform admin center, admins can view orphaned flows alongside key information, like the last owner and the days since it last ran, making it easy to reassign ownership or turn off unnecessary flows in a click. ☘️This proactive tool enhances visibility and reduces operational risks, helping admins actively oversee and streamline their automation landscape. Public Preview available from Jan 2025 #PowerAutomate #RPA #IPA #Automation

.2 Data-First Orchestration Instead of building pipelines by explicitly connecting nodes or tasks, your framework creates data-driven dependencies: Dynamic Relationships: Tasks are tied together based on shared inputs/outputs rather than explicit “edges” or “joins.” Event-Based Triggers: Data changes (not hard-coded schedules) drive execution, enabling true event-driven orchestration. 2.3 Data-First Identity Structures Unlike Data Vault 2.0, your knowledge framework treats hubs, links, and satellites as dynamic, multi-dimensional identity/event models: Hubs and Links: Represent both identity and events—capturing states and relationships in a flexible, non-static way. Satellites: Are treated as arrays or blobs, with identity-based indexing and attributes that are self-defining (e.g., functionally named GUIDs). --- 3. Why YAML and DAG Tools Don’t Work 3.1 YAML YAML is often used for infrastructure-as-code (IaC) and pipeline definitions, but it: Lacks Flexibility: YAML is inherently static—updating configurations or adapting pipelines requires rewriting text-heavy configs. Fails to Generalize: Each YAML script is tied to a specific system, making cross-environment or cross-task orchestration unwieldy. Doesn’t Abstract Logic: Unlike your data-first orchestration, YAML scripts are tied to low-level task definitions. 3.2 DAG Tools (Airflow, Prefect, etc.) DAG-based tools focus on creating task dependencies, but they: Don’t Scale: Adding new tasks or engines creates increasingly complex and brittle workflows. Are Node-Driven, Not Data-Driven: DAGs emphasize node connectivity but fail to abstract the data relationships that should drive orchestration. Static Execution Plans: DAGs lack the dynamic adaptability needed for real-time, event-driven tasks. ---

𝐒𝐜𝐚𝐥𝐢𝐧𝐠 𝐄𝐯𝐞𝐧𝐭-𝐃𝐫𝐢𝐯𝐞𝐧 𝐒𝐲𝐬𝐭𝐞𝐦𝐬 𝐖𝐡𝐢𝐥𝐞 𝐌𝐚𝐢𝐧𝐭𝐚𝐢𝐧𝐢𝐧𝐠 𝐄𝐯𝐞𝐧𝐭 𝐎𝐫𝐝𝐞𝐫 𝐂𝐡𝐚𝐥𝐥𝐞𝐧𝐠𝐞: Maintaining Event Sequence Integrity in Scalable Systems In scalable event-driven systems, particularly those handling critical sequential operations like financial transactions or order processing, the ability to maintain the order of events becomes a significant challenge. As systems expand and more consumers or services are added to handle increased loads, ensuring that all related events are processed in sequence by the appropriate consumer is crucial but difficult. 𝐒𝐨𝐥𝐮𝐭𝐢𝐨𝐧: 𝐒𝐨𝐥𝐚𝐜𝐞 𝐏𝐚𝐫𝐭𝐢𝐭𝐢𝐨𝐧 𝐐𝐮𝐞𝐮𝐞𝐬 • 𝐂𝐨𝐧𝐭𝐞𝐱𝐭-𝐀𝐰𝐚𝐫𝐞 𝐋𝐨𝐚𝐝 𝐁𝐚𝐥𝐚𝐧𝐜𝐢𝐧𝐠: Solace introduces partition queues that use key-based routing to ensure that all messages sharing the same contextual key (such as a transaction ID or user identifier) are consistently routed to the same consumer. This method ensures the integrity and chronological processing of related events. • 𝐃𝐲𝐧𝐚𝐦𝐢𝐜 𝐒𝐜𝐚𝐥𝐢𝐧𝐠 𝐚𝐧𝐝 𝐈𝐧𝐭𝐞𝐥𝐥𝐢𝐠𝐞𝐧𝐭 𝐑𝐞𝐛𝐚𝐥𝐚𝐧𝐜𝐢𝐧𝐠: Solace’s partition queues are designed to adapt dynamically to fluctuating loads, integrating seamlessly with auto-scaling technologies like Kubernetes Event-Driven Autoscaling (KEDA). This capability allows the system to scale consumers up or down based on real-time demand, while maintaining minimal impact on ongoing processes during rebalancing. • 𝐆𝐮𝐚𝐫𝐚𝐧𝐭𝐞𝐞𝐝 𝐎𝐫𝐝𝐞𝐫 𝐨𝐟 𝐏𝐫𝐨𝐜𝐞𝐬𝐬𝐢𝐧𝐠: By mapping each key to a specific partition and consequently to a unique consumer, Solace guarantees that events are processed in the order they arrive. This is particularly vital in environments where the sequence of events directly impacts business outcomes or data integrity. • 𝐑𝐞𝐝𝐮𝐜𝐞𝐝 𝐂𝐨𝐦𝐩𝐥𝐞𝐱𝐢𝐭𝐲 𝐚𝐧𝐝 𝐎𝐯𝐞𝐫𝐡𝐞𝐚𝐝: Solace’s approach reduces the need for additional layers of logic or infrastructure to manage event distribution and order, simplifying architecture and reducing overhead. Solace’s partition queues provide a robust solution for any high-scale, event-driven system that requires precise control over event sequence and distribution, facilitating both operational efficiency and reliability.

Automation through Power Automate

 Example Integration Workflow Code Generation Phase (Orchestrated) 1. Metadata agent generates schema definitions. 2. Model agent builds relational models. 3. Pipeline agent creates data pipelines. Data Processing Phase (Autonomous) 1. Data ingestion agents independently consume streams. 2. Transformation agents process and align data. 3. API agents expose processed data for downstream applications. Summary • Orchestration for Code: • Ensures structured, error-free generation of workflows and systems. • Suitable for tasks with clear dependencies and sequencing needs. • Autonomy for Data: • Enables real-time, scalable, and adaptive processing of dynamic data. • Ideal for event-driven workflows and decentralized environments. • Unified Model: • Dual-mode agents combine the strengths of orchestration and autonomy, seamlessly switching between structured and reactive behaviors.