The rapid growth in the storage scale of wide-area distributed file systems (DFS) calls for fast and scalable metadata management. Metadata replication is the widely used technique for improving the performance and scalability of metadata management. Because of the POSIX requirement of file systems, many existing metadata management techniques utilize a costly design for the sake of metadata consistency, leading to unacceptable performance overhead. We propose a new metadata consistency maintenance method (ICCG), which includes an incremental consistency guaranteed directory tree synchronization (ICGDT) and a causal consistency guaranteed replica index synchronization (CCGRI), to ensure system performance without sacrificing metadata consistency. ICGDT uses a flexible consistency scheme based on the state of files and directories maintained through the conflict state tree to provide an incremental consistency for metadata, which satisfies both metadata consistency and performance requirements. CCGRI ensures low latency and consistent access to data by establishing a causal consistency for replica indexes through multi-version extent trees and logical time. Experimental results demonstrate the effectiveness of our methods. Compared with the strong consistency policies widely used in modern DFSes, our methods significantly improve the system performance. For example, in file creation, ICCG can improve the performance of directory tree operations by at least 36.4 times.
The rapid development of ISAs has brought the issue of software compatibility to the forefront in the embedded field. To address this challenge, one of the promising solutions is the adoption of a multiple-ISA processor that supports multiple different ISAs. However, due to constraints in cost and performance, the architecture of a multiple-ISA processor must be carefully optimized to meet the specific requirements of embedded systems. By exploring the RISC-V and ARM Thumb ISAs, this paper proposes RVAM16, which is an optimized multiple-ISA processor microarchitecture for embedded devices based on hardware binary translation technique. The results show that, when running non-native ARM Thumb programs, RVAM16 achieves a significant speedup of over 2.73× with less area and energy consumption compared to using hardware binary translation alone, reaching more than 70% of the performance of native RISC-V programs.
As deep neural networks (DNNs) have been successfully adopted in various domains, the training of these large-scale models becomes increasingly difficult and is often deployed on compute clusters composed of many devices like GPUs. However, as the size of the cluster increases, so does the possibility of failures during training. Currently, faults are mainly handled by recording checkpoints and recovering, but this approach causes large overhead and affects the training efficiency even when no error occurs. The low checkpointing frequency leads to a large loss of training time, while the high recording frequency affects the training efficiency. To solve this contradiction, we propose BAFT, a bubble-aware fault tolerant framework for hybrid parallel distributed training. BAFT can automatically analyze parallel strategies, profile the runtime information, and schedule checkpointing tasks at the granularity of pipeline stage depending on the bubble distribution in the training. It supports higher checkpoint efficiency and only introduces less than 1% time overhead, which allows us to record checkpoints at high frequency, thereby reducing the time loss in error recovery and avoiding the impact of fault tolerance on training.
Blockchain as a decentralized storage technology is widely used in many fields. It has extremely strict requirements for reliability because there are many potentially malicious nodes. Generally, blockchain is a chain storage structure formed by interconnecting blocks A block is consist of a block header and a block body. The metadata is stored in block header and data is stored in block body.
To solve the above problem, we propose Dynamic-EC, which is a Dynamic Erasure Coding method in permissioned blockchain systems. The key idea of Dynamic-EC is to reduce the storage overhead by dynamically adjusting the total number of parities according to the risk level of the whole system, which is determined by the number of perceived malicious nodes, while ensuring the system reliability. To demonstrate the effectiveness of Dynamic-EC, we conduct several experiments on an open source blockchain software Tendermint. The results show that, compared to the state-of-the-art erasure coding methods, Dynamic-EC reduces the storage overhead by up to 42%, and decreases the average write latency of blocks by up to 25%, respectively.
Code review is a critical process in software development, contributing to the overall quality of the product by identifying errors early. A key aspect of this process is the selection of appropriate reviewers to scrutinize changes made to source code. However, in large-scale open-source projects, selecting the most suitable reviewers for a specific change can be a challenging task. To address this, we introduce the Code Context Based Reviewer Recommendation (CCB-RR), a model that leverages information from changesets to recommend the most suitable reviewers. The model takes into consideration the paths of modified files and the context derived from the changesets, including their titles and descriptions. Additionally, CCB-RR employs KeyBERT to extract the most relevant keywords and compare the semantic similarity across changesets. The model integrates the paths of modified files, keyword information, and the context of code changes to form a comprehensive picture of the changeset. We conducted extensive experiments on four open-source projects, demonstrating the effectiveness of CCB-RR. The model achieved a Top-1 accuracy of 60%, 55%, 51%, and 45% on the Android, OpenStack, QT, and LibreOffice projects respectively. For Mean Reciprocal Rank (MRR), CCB achieved 71%, 62%, 52%, and 68% on the same projects respectively, thereby highlighting its potential for practical application in code reviewer recommendation.
SaaS (Software-as-a-Service) is a service model provided by cloud computing. It has a high requirement for QoS (Quality of Software) due to its method of providing software service. However, manual identification and diagnosis for performance issues is typically expensive and laborious because of the complexity of the application software and the dynamic nature of the deployment environment. Recently, substantial research efforts have been devoted to automatically identifying and diagnosing performance issues of SaaS software. In this survey, we comprehensively review the different methods about automatically identifying and diagnosing performance issues of SaaS software. We divide them into three steps according to their function: performance log generation, performance issue identification and performance issue diagnosis. We then comprehensively review these methods by their development history. Meanwhile, we give our proposed solution for each step. Finally, the effectiveness of our proposed methods is shown by experiments.
Classic Graph Convolutional Networks (GCNs) often learn node representation holistically, which ignores the distinct impacts from different neighbors when aggregating their features to update a node’s representation. Disentangled GCNs have been proposed to divide each node’s representation into several feature units. However, current disentangling methods do not try to figure out how many inherent factors the model should assign to help extract the best representation of each node. This paper then proposes D2-GCN to provide dynamic disentanglement in GCNs and present the most appropriate factorization of each node’s mixed features. The convergence of the proposed method is proved both theoretically and experimentally. Experiments on real-world datasets show that D2-GCN outperforms the baseline models concerning node classification results in both single- and multi-label tasks.
Real-world objects exhibit intricate semantic properties that can be characterized from a multitude of perspectives, which necessitates the development of a model capable of discerning multiple patterns within data, while concurrently predicting several Labeling Dimensions (LDs) — a task known as Multi-dimensional Classification (MDC). While the class imbalance issue has been extensively investigated within the multi-class paradigm, its study in the MDC context has been limited due to the imbalance shift phenomenon. A sample’s classification as a minor or major class instance becomes ambiguous when it belongs to a minor class in one LD and a major class in another. Previous MDC methodologies predominantly emphasized instance-wise criteria, neglecting prediction capabilities from a dimension aspect, i.e., the average classification performance across LDs. We assert the significance of dimension-wise metrics in real-world MDC applications and introduce two such metrics. Furthermore, we observe imbalanced class distributions within each LD and propose a novel Imbalance-Aware fusion Model (IMAM) for addressing the MDC problem. Specifically, we first decompose the task into multiple multi-class classification problems, creating imbalance-aware deep models for each LD separately. This straightforward method performs well across LDs without sacrificing performance in instance-wise criteria. Subsequently, we employ LD-wise models as multiple teachers and transfer their knowledge across all LDs to a unified student model. Experimental results on several real-world datasets demonstrate that our IMAM approach excels in both instance-wise evaluations and the proposed dimension-wise metrics.
Association in-between features has been demonstrated to improve the representation ability of data. However, the original association data reconstruction method may face two issues: the dimension of reconstructed data is undoubtedly higher than that of original data, and adopted association measure method does not well balance effectiveness and efficiency. To address above two issues, this paper proposes a novel association-based representation improvement method, named as AssoRep. AssoRep first obtains the association between features via distance correlation method that has some advantages than Pearson’s correlation coefficient. Then an improved matrix is formed via stacking the association value of any two features. Next, an improved feature representation is obtained by aggregating the original feature with the enhancement matrix. Finally, the improved feature representation is mapped to a low-dimensional space via principal component analysis. The effectiveness of AssoRep is validated on 120 datasets and the fruits further prefect our previous work on the association data reconstruction.
Recently, crowdsourcing has established itself as an efficient labeling solution by distributing tasks to crowd workers. As the workers can make mistakes with diverse expertise, one core learning task is to estimate each worker’s expertise, and aggregate over them to infer the latent true labels. In this paper, we show that as one of the major research directions, the noise transition matrix based worker expertise modeling methods commonly overfit the annotation noise, either due to the oversimplified noise assumption or inaccurate estimation. To solve this problem, we propose a knowledge distillation framework (KD-Crowd) by combining the complementary strength of noise-model-free robust learning techniques and transition matrix based worker expertise modeling. The framework consists of two stages: in Stage 1, a noise-model-free robust student model is trained by treating the prediction of a transition matrix based crowdsourcing teacher model as noisy labels, aiming at correcting the teacher’s mistakes and obtaining better true label predictions; in Stage 2, we switch their roles, retraining a better crowdsourcing model using the crowds’ annotations supervised by the refined true label predictions given by Stage 1. Additionally, we propose one f-mutual information gain (
In this paper, we propose a novel warm restart technique using a new logarithmic step size for the stochastic gradient descent (SGD) approach. For smooth and non-convex functions, we establish an
