Fault injection ensures the resilience of cloud-edge computing systems. System observability for conducting effective and comprehensive fault injection experiments must be enhanced, particularly due to the complex and distributed architectures of cloud-edge environments, where a single failure can trigger cascading effects, creating a dynamic and expanding fault blast radius. Additionally, accurate fault diagnosis becomes even more challenging when multiple teams simultaneously inject faults into interdependent components, highlighting the need for collaborative fault injection rather than confounding concurrent activities. This work addresses these challenges by proposing a collaborative fault injection framework powered by knowledge graph technologies. The framework enhances collaboration by enabling teams to seamlessly synchronize fault injection experiments within a shared environment. By extracting dynamic fault blast radius information from the proposed fault observation knowledge graph after fault injection, the system observability is improved and more efficient fault injection experiments are ensured. To evaluate the effectiveness of the proposed framework, we use an evolutionary game theory model to analyze the dynamic interactions among stakeholders. Utilizing a simulated OpenStack platform as our testbed, we demonstrate that our approach surpasses traditional fault injection techniques, achieving a remarkable 42.76% reduction in the average time required to observe the complete fault blast radius.

Fault injection ensures resilience of cloud-edge computing systems.Recognizing the importance of fault injection underscores a critical challenge: the need to enhance system observability for conducting effective and comprehensive fault injection experiments.This is particularly crucial due to the complex, distributed architectures of cloud-edge environments, where a single failure can trigger cascading failures, creating a dynamic and expanding fault blast radius.To address these challenges, this work advocates for the adoption of knowledge graph technologies to significantly improve fault knowledge representation, system observability, and minimize resource expenditure on monitoring systems.We introduce the Fault Observation Knowledge Graph (FOKG), comprising three detailed sub-knowledge graphs that describe the nuances of application deployment, delineate known fault chains, and outline the monitoring mechanisms in place.By extracting dynamic fault blast radius information and refining monitoring strategies from FOKG after fault injection, we ensure a more targeted observation and efficient fault injection experiment.Our methodology, validated on a custom-developed Open-Stack platform, surpasses traditional fault injection techniques by achieving a remarkable 42.76% reduction in the average time required to observe the complete fault blast radius.

Multi-task learning (MTL) is a common machine learning technique that allows the model to share information across different tasks and improve the accuracy of recommendations for all of them. Many existing MTL implementations suffer from scalability issues as the training and inference performance can degrade with the increasing number of tasks, which can limit production use case scenarios for MTL-based recommender systems. Inspired by the recent advances of large language models, we developed an end-to-end efficient and scalable Generalist Recommender (GRec). GRec takes comprehensive data signals by utilizing NLP heads, parallel Transformers, as well as a wide and deep structure to process multi-modal inputs. These inputs are then combined and fed through a newly proposed task-sentence level routing mechanism to scale the model capabilities on multiple tasks without compromising performance. Offline evaluations and online experiments show that GRec significantly outperforms our previous recommender solutions. GRec has been successfully deployed on one of the largest telecom websites and apps, effectively managing high volumes of online traffic every day.

BACKGROUND: Genome-wide DNA copy number changes are the hallmark events in the initiation and progression of cancers. Quantitative analysis of somatic copy number alterations (CNAs) has broad applications in cancer research. With the increasing capacity of high-throughput sequencing technologies, fast and efficient segmentation algorithms are required when characterizing high density CNAs data. RESULTS: A fast and informative segmentation algorithm, DBS (Deviation Binary Segmentation), is developed and discussed. The DBS method is based on the least absolute error principles and is inspired by the segmentation method rooted in the circular binary segmentation procedure. DBS uses point-by-point model calculation to ensure the accuracy of segmentation and combines a binary search algorithm with heuristics derived from the Central Limit Theorem. The DBS algorithm is very efficient requiring a computational complexity of O(n*log n), and is faster than its predecessors. Moreover, DBS measures the change-point amplitude of mean values of two adjacent segments at a breakpoint, where the significant degree of change-point amplitude is determined by the weighted average deviation at breakpoints. Accordingly, using the constructed binary tree of significant degree, DBS informs whether the results of segmentation are over- or under-segmented. CONCLUSION: DBS is implemented in a platform-independent and open-source Java application (ToolSeg), including a graphical user interface and simulation data generation, as well as various segmentation methods in the native Java language.