Deep learning tasks (DLT) include training and inference tasks, where training DLTs have requirements on minimizing average job completion time (JCT) and inference tasks need sufficient GPUs to meet real-time performance. Unfortunately, existing work separately deploys multi-tenant training and inference GPU cluster, leading to the high JCT of training DLTs with limited GPUs when the inference cluster is under insufficient GPU utilization due to the periodic inference workload. DeepBoot solves the challenges by utilizing idle GPUs in the inference cluster for the training DLTs. Specifically, 1) DeepBoot designs adaptive task scaling (ATS) algorithm to allocate GPUs in the training and inference clusters for training DLTs and minimize the performance loss when reclaiming inference GPUs. 2) DeepBoot implements auto-fast elastic (AFE) training based on Pollux to reduce the restart overhead by inference GPU reclaiming. Our implementation on the testbed and large-scale simulation in Microsoft deep learning workload shows that DeepBoot can achieve 32% and 38% average JCT reduction respectively compared with the scheduler without utilizing idle GPUs in the inference cluster.

Consumer digital ecosystems include a large volume of different types of applications, and those applications are usually deployed in industrial cloud computing systems. Currently, microservices are one of the most prevailing architectures for industrial clouds. Similar to other architectures, microservices may also produce failures, so failure diagnosis for microservices becomes an inevitable problem in industrial clouds. A majority of existing methods focus on statistical analysis for monitoring data or system topological structure. However, because these methods usually only harness service-level or machine-level metrics, they cannot complete fine-grained failure diagnosis, increasing the running risk of microservice-based industrial clouds. To tackle this issue, in this paper, we design a novel graph structure to represent failure dependencies, especially the heterogeneity, and name it as the heterogeneous failure dependence graph (HFDG). We propose a framework to inform engineers which type of and where failures occur in industrial clouds. The HFDG can be used to mine the propagation of failures between different types of components. We also propose a novel neural network model based on attention mechanism and heterogeneous graph neural network, to fully leverage the metric data and HFDG. We performed experiments on three large-scale public datasets from real-world microservices-based systems. The experimental results demonstrate the superior performance of our model compared to well-known baselines.

Yuntai Bao, Xuhong Zhang, Tianyu Du, Xinkui Zhao, Zhengwen Feng, Hao Peng, Jianwei Yin. Findings of the Association for Computational Linguistics: ACL 2025. 2025.

Web service-oriented transactions have become an integral part of the Internet economy, with the mainstream transaction patterns relying primarily on cloud service markets. However, traditional service transaction methods have deficiencies in terms of both trust and scalability. Distrust between service provider (SP) and consumer (SC), particularly around online payments and data security, impedes the further growth of service transactions. Although blockchain-based transaction mechanisms have made notable progress in addressing trust issues, they still face performance bottlenecks. To tackle these challenges, this paper introduces TrustPay, a service transaction framework that leverages a dual-layer blockchain structure consisting of a parent chain and multiple subchains. The framework partitions the subchain network based on business, with each subchain dedicated to storing service invocation records generated within a specific business unit. The smart contract deployed on the parent chain will settle the invocation records in all subchain networks as transaction records and facilitate automatic transfers among blockchain accounts. This design leverages blockchain's inherent reliability while improving its scalability in large-scale scenarios. Additionally, a novel consensus protocol, REFEREE, is introduced and applied to the subchain network, ensuring efficient recording of invocation data and trusted verification among participants, further enhancing both trust and performance. Comparative experiments and analysis show that TrustPay's dual-layer blockchain structure and REFEREE protocol are not only reliable but also outperform baseline methods in terms of efficiency.

Quantum readout noise turns out to be the most significant source of error, which greatly affects the measurement fidelity. Matrix-based calibration has been demonstrated to be effective in various quantum platforms. However, existing methodologies are fundamentally limited in either scalability or accuracy. Inspired by the classical finite element method (FEM), a formal method to model the complex interaction between elements, we present our calibration framework named QuFEM. First, we apply a divide-and-conquer strategy that formulates the calibration as a series of tensor products with noise matrices. This matrices are iteratively characterized together with the calibrated probability distribution, aiming to capture the inherent locality of qubit interactions. Then, to accelerate the end-to-end calibration, we propose a sparse tensor-product engine to exploit the sparsity in the intermediate values. Our experiments show that QuFEM achieves 2.5×10³× speedup in the 136-qubit calibration compared to the state-of-the-art matrix-based calibration technique [50], and provides 1.2× and 1.4× fidelity improvement on the 18-qubit and 36-qubit real-world quantum devices.