A5000643612- Caching and Content Delivery
- Advanced Data Storage Technologies
- Cooperative Communication and Network Coding
- Peer-to-Peer Network Technologies
- Cloud Data Security Solutions
- IoT and Edge/Fog Computing
- Blockchain Technology Applications and Security
- Security and Verification in Computing
- Opportunistic and Delay-Tolerant Networks
- Intracranial Aneurysms: Treatment and Complications
- Network Traffic and Congestion Control
- Acute Ischemic Stroke Management
- Error Correcting Code Techniques
- Data Quality and Management
- Algorithms and Data Compression
- Cellular Automata and Applications
- Scientific Computing and Data Management
- Mobile Ad Hoc Networks
- Cloud Computing and Resource Management
- Time Series Analysis and Forecasting
- Cerebrovascular and Carotid Artery Diseases
Budapest University of Technology and Economics
2013-2021
Aalborg University
2016
Cisco Systems (United States)
2016
Distributed storage is usually considered within a cloud provider to ensure availability and reliability of the data. However, user still directly dependent on quality single system. It also entrusting service with large amounts private data, which may be accessed by successful attack that system or even inspected government agencies in some countries. This paper advocates general framework for network coding enabled distributed over multiple commercial solutions, such as, Dropbox, Box,...
This paper advocates the use of random linear network coding for storage in distributed clouds order to reduce and traffic costs dynamic settings, i.e. when adding removing numerous devices/clouds on-the-fly number reachable is limited. We introduce various approaches that trade-off reliability, costs, system complexity relying on probabilistic recoding cloud regeneration. compare these with other based data replication Reed-Solomon codes. A simulator has been developed carry out a thorough...
Abstract Cloud computing considerably reduces the costs of deploying applications through on-demand, automated and fine-granular allocation resources. Even in private settings, cloud platforms enable agile self-service management, which means that physical resources are shared more efficiently. Nevertheless, using infrastructures also creates opportunities for attacks data breaches. In this paper, we describe SecureCloud approach. The project aims to confidentiality integrity running...
As faster storage devices become commercially viable alternatives to disk drives, the network is increasingly becoming bottleneck in achieving good performance distributed systems. This especially true for erasure coded storage, where reconstruction of lost data can significantly encumber system. Thus, a significant amount research has focused on reducing transferred during this repair process. However, most cases assumed have uniform static structure. One reason behind that many state art...
The amount of data generated worldwide is expected to grow from 33 175 ZB by 2025 in part driven the growth Internet Things (IoT) and cyber-physical systems (CPS). To cope with this enormous data, new cloud storage techniques must be developed. Generalised Data Deduplication (GDD) a paradigm for reducing cost systematically identifying near identical chunks, storing their common component once, compact representation deviation original chunk each chunk. This paper presents system...
A significant amount of research on using erasure coding for distributed storage has focused reducing the data that needs to be transferred replace failed nodes. This continues an active topic as introduction faster devices looks put even greater strain network. However, with a few notable exceptions, most published work assumes flat, static network topology between nodes system. We propose general framework find lowest cost feasible repairs in more realistic, heterogeneous and dynamic...
Although cloud systems provide a reliable and flexible storage solution, the use of single service constitutes point failure, which can compromise data availability, download speed, security. To address these challenges, we advocate for multiple providers simultaneously using network coding as key enabling technology. Our goal is to study two challenges coded systems. First, efficient update number fragments per in system aggregating clouds order boost speed files. We developed novel scheme...
Distributed storage has attracted large interest lately from both industry and researchers as a flexible, cost-efficient, high performance, potentially secure solution for geographically distributed data centers, edge caching or sharing among users. This paper studies the benefits of random linear network coding to exploit multiple commercially available cloud providers simultaneously with possibility constantly adapt changing performance in order optimize retrieval times. The main...
We introduce Titchy, which is a compression method for time-series data generated by the Internet of Things. Our proposed flexible and has several advantages when applied in IoT ecosystem: 1) it able to compress even only small amount memory can be allocated it; 2) compresses tiny chunks, so introduces very little latency during online operation thus, enables frequent timely updates with compressed data; 3) facilitates efficient storage retrieval enabling low-cost random access data,...
We are witnessing a wave of emerging cloud computing technologies and services that empower advanced applications from different vertical sectors, with diverse requirements. These trends give rise to number fundamental challenges relate the application deployment, support heterogeneous infrastructures provided security. In this setting, SERRANO project steps in define an intent-based paradigm operating federated consisting edge, HPC resources, which will be realized through platform....
Current video streaming services use a conventional, client-server network topology that puts heavy load on content servers. Previous work has shown Peer-to-Peer (P2P) assisted solutions can potentially reduce this load. However, implementing P2P-assisted poses several challenges in modern networks. Users tend to stream videos the go, using their mobile devices. This mobility makes difficult orchestrate. Furthermore, peers have contribute storage network, which is challenging, since devices...
Distributed storage solutions have become widespread due to their ability store large amounts of data reliably across a network unreliable nodes, by employing repair mechanisms prevent loss. Conventional systems rely on static designs with central control entity oversee and the process. Given costs for maintaining cooling centers, our work proposes studies feasibility fully decentralized that can even and, sometimes, unavailable mobile devices. This imposes new challenges design as number...
High capacity storage systems distribute files across several devices (nodes) and apply an erasure code to meet availability reliability requirements. Since can lose network connectivity or fail permanently, a dynamic repair mechanism must be put in place. In such cases new recovery node gets connected given subset of the operating nodes receives part stored data. The objective this paper is investigate data survival for Random Linear Network Coding (RLNC) as function topology communication...
Video on Demand (VoD) streaming makes up an increasing portion of global IP traffic. Traditionally, data is served by content providers directly, putting a huge network load the servers. We propose using our WebRTC-based protocols WebPeer and CodedWebPeer to create P2P-assisted VoD system with coding. introduce two metrics: health measure overall saturation stability show if peers in are able serve each other without help server. have implemented testbed emulate real-life scenarios that...
Cloud computing considerably reduces the costs of deploying applications through on-demand, automated, and fine-granular allocation resources. Even in private settings, cloud platforms enable agile self-service management, which means that physical resources are shared more efficiently. Nevertheless, using infrastructures also creates opportunities for attacks data breaches. In this paper, we describe SecureCloud approach. The project aims to confidentiality integrity running potentially...
Cervical artery dissection is a common cause of stroke in young adults. It might occur shortly after forceful neck trauma or minor injury. However, spontaneous also common, which associated with genetic, anatomical environmental risk factors. can produce broad spectrum clinical presentation varying from local symptoms to focal neurological deficits determined by the arterial territory involved. Early recognition important since immediate initiation treatment significantly improve patient...
Conventional cloud storage services offer relatively good reliability and performance in a cost-effective manner. However, they are typically structured centralized highly controlled fashion. In more dynamic scenarios, these approaches unfeasible developing decentralized becomes critical. The novelty of this paper is the introduction distributed mobile cloud, which uses free resources on user devices to move edges network. At core our approach, lies use random linear network coding provide...
The need for real-time analytics and faster decision-making mechanisms has led to the adoption of hardware accelerators such as GPUs FPGAs within edge cloud computing continuum. However, their programmability lack orchestration seamless deployment make them difficult use efficiently. We address these challenges by presenting SERRANO, a project transparent application in secure, accelerated, cognitive In this work, we introduce SERRANO platform its software, orchestration, services, focusing...
Peer-to-peer (P2P) applications for mobile devices are becoming more and popular because of increasing bandwidth, computational performance storage capabilities. Such a based distributed architecture offers significant advantages in several scenarios from the perspective both users network operators. In these situations, certain redundancy must be built into system availability nodes can vary greatly. The goal this paper is to present methodology minimize necessary by predicting number...