Since the advent of network function virtualization (NFV), cloud service providers (CSPs) can implement traditional dedicated network devices as software and flexibly instantiate network functions (NFs) on common off-the-shelf servers. NFV technology enables CSPs to deploy their NFs to a cloud data center in the form of virtual network functions (VNFs) without costly capital expenditures and operating expenses. However, it is an essential but intractable issue for CSPs to devise a suitable VNF placement scheme to optimize network resource consumption and improve network performance. In this article, we focus on the VNF placement problem for mapping users’ service function chain requests (SFCRs) in cloud networks. To enhance network resource utilization, we consider the fundamental resource overheads and implementation method of VNFs. The VNF placement problem is formulated as an integer linear programming model with the aim of minimizing the total network resource consumption while guaranteeing the delay requirements of SFCRs. We devise a two-phase optimization solution (TPOS) to solve the problem. TPOS contains a mapping phase to map SFCRs on servers and an adjustment phase to optimize the placement of VNFs and VNF requests. Evaluation results demonstrate that TPOS can derive near-optimal server resource consumption and significantly enhance network resource utilization. TPOS can guarantee the delay requirements of SFCRs and outperform contrastive schemes in terms of activated servers, SFCR acceptance ratio, and average VNF utilization.

Network Function Virtualization (NFV) is a promising paradigm that separates network functions from proprietary devices. Network service in NFV-enabled networks is achieved as a Service Function Chain (SFC), consisting of a series of ordered Virtual Network Functions (VNFs). However, migration of VNFs for more flexible services within dynamic networks is a key challenge. Current VNF migration studies mainly focus on single VNF migration decisions without considering the sharing and concurrent migration of VNF Instance (VNFI). In this paper, we assume that each deployed VNFI is used by multiple SFCs and deal with the optimal location allocation for the concurrent migration of VNFIs based on the actual network situation. We first formalize the VNF migration and SFC reconfiguration problem as a mathematical model, which aims to minimize the end-to-end delay for all affected services and to guarantee network load balancing after the migration simultaneously. To this end, we prove the NP-hardness of this problem and propose the Improved Hybrid Genetic Evolution (IHGE) algorithm to address it. Besides, to reduce the computation overhead of IHGE for large-scale networks, a multi-stage heuristic algorithm based on optimal order (MSH-OR) is designed. Finally, we perform a side-by-side comparison with prior algorithms. Extensive evaluation shows that the proposed approaches can effectively reduce the average delay for different scale networks while ensuring network load balancing.

Nowadays, network softwareization is an emerging techno-economic transformation trend that significantly impacts how enterprises deploy their network services. As an essential technology in this trend, Network Function Virtualization (NFV) enables scalable and inexpensive network services by flexibly instantiating Virtualized Network Functions (VNFs) on commercial-off-the-shelf devices. In this paper, we focus on the VNF placement problem in NFV-enabled networks, aiming to maximize the number of accepted Service Function Chain Requests (SFCRs) while guaranteeing their delay requirements. To improve resource utilization efficiency, we take account of Fundamental Resource Overheads (FROs) and the shareability of VNF instances. We mathematically formulate the VNF placement problem and propose the Throughput Optimization and Delay Guarantee (TO-DG) heuristic solution, consisting of an affinity-based SFCR mapping algorithm and a VNF request adjustment algorithm. The evaluation results show that the performance of TO-DG is near to results derived by ILP solver for small scale problems. Moreover, TO-DG obtains higher network throughput than contrasting schemes in different scenarios and significantly improves network resource utilization.

With greater numbers of cloud applications being deployed in virtual clusters (VCs) and running in datacenters, the energy consumption of datacenters is experiencing rapidly cumulative growth. Thus, it is a critical issue that how to allocate virtual machines (VMs) of the VC in physical machines (PMs) for energy savings as much as possible. Considering each PM/switch has a certain failure rate, VMs may not be executed when they meet with any PM/switch fault. So, a compact allocation scheme may benefit in low energy consumption but increase the risk of violating the availability of the VC. In this paper, we consider four optimization objectives about the VC and the datacenter, i.e., availability, energy consumption, average resource utilization, and resource load balance. Then we propose a multi-objective optimization model and raise an evolution algorithm to trade-off among these four optimization objectives. Finally, experimental results show the effectiveness and efficiency of our algorithm.

Multi-modal knowledge graphs (MMKGs) enrich the semantic expression capabilities of traditional knowledge graphs by incorporating diverse modal information, showcasing immense potential in various knowledge reasoning tasks. However, existing MMKGs encounter numerous challenges in the link prediction task (i.e., knowledge graph completion reasoning), primarily due to the complexity and diversity of modal information and the imbalance in quality. These challenges make the efficient fusion and enhancement of multi-modal information difficult to achieve. Most existing methods adopt simple concatenation or weighted fusion of modal features, but such approaches fail to fully capture the deep semantic interactions between modalities and perform poorly when confronted with modal noise or missing information. To address these issues, this paper proposes a novel framework model-Adaptive Fusion and Modality Information Enhancement(AFME). This framework consists of two parts: the Modal Information Fusion module (MoIFu) and the Modal Information Enhancement module (MoIEn). By introducing a relationship-driven denoising mechanism and a dynamic weight allocation mechanism, the framework achieves efficient adaptive fusion of multi-modal information. It employs a generative adversarial network (GAN) structure to enable global guidance of structural modalities over feature modalities and adopts a multi-layer self-attention mechanism to optimize both intra- and inter-modal features. Finally, it jointly optimizes the losses of the triple prediction task and the adversarial generation task. Experimental results demonstrate that the AFME framework significantly improves multi-modal feature utilization and knowledge reasoning capabilities on multiple benchmark datasets, validating its efficiency and superiority in complex multi-modal scenarios.

As multilingual artificial intelligence systems proliferate, achieving robust cross-lingual understanding remains an open challenge. Recent works have made progress on visual question answering (VQA) models by pretraining on large English image-text datasets. However, there is a language gap as most models are English-centric. Existing attempts at multilingual VQA rely on machine translation or multilingual model pretraining, but cannot effectively transfer rich cross-modal knowledge from English models. In this work, we propose the cross-lingual multimodal knowledge transfer (CMKT) framework to efficiently extend English VQA models to non-English languages via knowledge distillation. Specifically, we introduce a code-mixed cross-lingual mask modeling (CCM) method to establish representations for new languages using small image-text data. We also design a multimodal knowledge distillation (MMKD) method to transfer modal understanding from English models by imitating their sequence processing. Experiments on the xGQA benchmark demonstrate that CMKT can effectively improve zero-shot learning and few-shot learning in non-English languages. Our method reduces the data and computation needed to train multilingual VQA models from scratch. The knowledge transfer paradigm enables non-English languages to inherit and generalize the intricate visual-semantic relationships learned from English. The results also show that the proposed method outperforms previous state-of-the-art methods in the zero-shot setting on the xGQA dataset.