ABSTRACT. With the transition of applications to Cloud and Serverless, performance engineering must adapt. In this paper, we will compare performance engineering methods, techniques and key findings for a complex, high performance, mission-critical system in public sector during its transition from an on-premises to a Cloud IaaS and Serverless deployment. The transition to serverless happened while the system continued maintaining close to 100% availability, achieving a 10-fold increase in traffic volume, and having to continue maintain compatibility and stability with other legacy systems that lack elasticity. Our work shows that while many traditional concerns are no longer seen in serverless computing, many new challenges arrive in test planning, observability, application design and tuning, that necessitate comprehensive performance engineering and new ways of thinking.

ABSTRACT. This paper provides insight into the need for Front end Performance Testing and suggestions on how to perform it. In today’s competitive world with ever growing web and mobile applications, it is not enough to focus just on functional and load testing. The user today expects a smooth and flawless experience. For this requirement, there is a constant need to not only to test the server side and backend but also focus on the client side.

ABSTRACT. ABAC as an approach that has started too get adopted across IBM and Industry. While the controls and instrumentation have become more available, governance of policies and attributes, along with lifecycle management, have not. This talk will look at how the Global Chief Data Office is providing these capabilities to IBMs most sensitive data.

ABSTRACT. Apart from making voice calls, a smartphone helps you perform a dozen of other tasks including taking photos, surfing web, booking tickets, transferring funds, posting updates to social media, group chats, email, playing games, and using enterprise apps. Undoubtedly, Smartphones offer an enormous potential for digital interactions between businesses and their customers as well.

ABSTRACT. At today’s rapid pace, digital business must stay agile without sacrificing security, availability and performance. Continuous testing is essential for achieving this goal, requiring test tools that are automated and operate at a scale to meet functional, system and performance verification needs. Performance testing is an essential activity in a software development life cycle. From initial planning to production analysis, application performance drives the development of better software iterations and releases. Application stakeholders, programmers and testers must make performance a primary consideration in all iterations of Agile development process.

ABSTRACT. A famous retailer in US has a challenging operation of allocating and re-allocating millions of order lines every day during a short window daily based on its customer priority. The order's inventory, capacity and delivery dates need to be modified reflecting the allocation logic. The customer's expected volume and the time window of this modification posed a challenge in terms of the infrastructure, load against the messaging systems and the various application and database performance tunings that had to be undertaken. IBM Sterling Order Management system took up this challenge as a 2-phased case study by (a) enhancing the product for the functional and performance improvements (b) deploying the application on a scalable, resilient and a reliable Kubernetes based platform such as OpenShift.

This session will delve into the challenges faced by the customer, functional changes that were done to accommodate customer's allocation challenges, performance scalability studies we performed on Order Management System on OCP on IBM Power Virtual Server Cloud which was purposeful to demonstrate the capabilities of IBM Sterling Order Management System (OMS) running on OpenShift Container Platform (OCP) on IBM PowerVS Cloud and exploits the OMS/ Hardware Capabilities to its fullest. After this session, you can design a performant, resilient topology & deployment of IBM Sterling OMS on OCP.

ABSTRACT. The earlier you find out problems in your code, the less impact they have and the less it costs to remediate them. In SW globalization, some tasks are critical in making a localized version, such as to find out hard coded string and make sure translation does not make anything wrong. Traditionally we perform manual or semi-automation testing, which are also known as Translatability and Translation Verification Test. It takes time and error-prone. Therefore, we introduce machine learning to scan hard coded string, check translation and UI issue automatically, which reduce effort and find out defects in earlier in development cycle.

ABSTRACT. Operating 5G edge solutions are complex and the traditional approach for managing networks and applications is changing. This session will discuss what closed loop automation is and how AI Driven closed loop automation can be used to improve the resilience of 5G edge solution by automating their operations to improve network performance, secure the 5G components, provide service assurance and improve availability of the end-to-end solution. Key 5G use edge cases, the core infrastructure components required and the underlying architecture that will improve the solutions resiliency using Closed Loop Automation built on AI will also be discussed. We will conclude by discussing the benefits of integrating closed loop automation into 5G edge solutions and lessons learned working with customers and standards organizations in this journey.

Expected outcomes for the participants is to better understand: - How 5G edge solution can perform better and be more resilient with closed loop automation. - Techniques for AI based operational analytics for anomaly detection, fault localization and root cause analysis. This involves combining large volumes of metrics, logs and topology information at infrastructure layer and core network. The AI-driven automation helps significantly reduce the time to detect and remediate faults in complex telecom deployments for SREs. - How AI capabilities are exposed using the 3GPP standards by incorporating Network Data and Analytics Function (NWDAF) for consumption by clients as well as other network management functions. -How the AI system integrate with Robotic Process Automation and underlying networking tools can automatically correct issues in a 5G cloud native/Kubernetes environment built with multiple network functions (across the Radio Access Network, Core and Transport domains) and applications running at the edge.

We will share our experience in creating resilient 5G edge solutions and the delivery will include a demonstration on how Closed Loop Automation is applied to build resilient 5G edge solutions.

ABSTRACT. Industry experienced the surge of digital traffic during the pandemic times and was able to effectively manage the situation. However, the revenue started to shrink, which will test the capability of the organizations. It’s imperative for the Business to provide always-on services and world class customer experience in order to take competitive advantage. Business models and operations undergo significant changes with Strategic technology roadmaps and application modernization trends. Cloud Adoption is a strategic move by organizations to reduce cost and achieve auto-scaling and auto-healing. But, a mere “lift-and-shift” is not going to yield the benefits of the cloud infrastructure. While following the cloud native principles and leveraging container platform capabilities, the enterprise architecture should include resiliency and Fault Tolerance.

Resiliency Capabilities include: 1) Resiliency is not just about opening the circuit when the ecosystem is going through turbulence. It’s also about building a Fault tolerant system that can carry out the business function through alternate service path following the redundancy principle and failing over to another service provider 2) Failures are bound to happen, but it's important to bring the system back up. Shifting towards zero-touch operation is the key to achieve it.

This presentation describes the key architectural principles and techniques to design fault tolerant system with rationalized APIs and redundancy that can be built to achieve high success rate for a business transaction.

ABSTRACT. In this paper we will walk through the modernization journey of Hardware Management Console (HMC) - the IBM Power System's management interface from a legacy monolith to set of microservices and how it resulted in improved performance, scalability along with becoming a more resilient product.

We will talk about the legacy architecture, challenges and roadblocks we were facing with the legacy product, the need for modernization, approach we took for splitting the product into microservices model and the resulting performance and scalability improvements.

Along with the modernization process, we will also talk about the monitoring tools like Prometheus that we introduced to monitor individual services in the modern architecture.