Elasticity is a key feature in the cloud computing context, and perhaps what distinguishes this paradigm of other ones, such as cluster grid computing. Considering importance elasticity objective paper to present comprehensive study about mechanisms available today. Initially, we propose classification for mechanisms, based on main features found analysed commercial academic solutions. In second moment, diverse related works are reviewed order define state art clouds. We also discuss some...

The IoT is pervading our daily activities and lives with devices scattered all over cities, transport systems, buildings, homes bodies. This invasion of sensors communication capabilities brings big concerns, mainly about the privacy confidentiality collected information. These concerns hinder wide adoption IoT. To overcome them, in this work, we present an Blockchain-based architecture for access authorizations. Following tendency requirements, user transparent, friendly, fully...

Network devices have always been considered as configurable black boxes until the emergence of software-defined networking (SDN). SDN enables networks to be programmed according user requirements; furthermore, it allows network easily modified suit transient demands. However, how do we program network? SDN-compliant switches offer a low-level interface that makes programming error-prone. The controllers provide application interfaces, but still low-level, limited, and inflexible. High-level...

Elasticity is undoubtedly one of the most striking characteristics cloud computing. Especially in area high performance computing (HPC), elasticity can be used to execute irregular and CPU-intensive applications. However, on- the-fly increase/decrease resources more widespread Web systems, which have their own IaaS-level load balancer. Considering HPC area, current approaches usually focus on batch jobs or assumptions such as previous knowledge application phases, source code rewriting...

Network neutrality is becoming increasingly important as the global debate intensifies and governments worldwide implement withdraw regulations. According to this principle, all Internet traffic must be processed without differentiation, regardless of origin, destination, and/or content. Neutrality supporters claim that differentiation can compromise innovation, fair competition, freedom choice. However, detecting an Service Provider (ISP) not employing practices still a challenge. This...

Summary The IoT is changing the way we interact with world. Very soon, almost all of our daily tasks will be made through self intelligent systems embedded in devices scattered around us. Their mission to turn cities, transportation systems, buildings, homes, and bodies smart environments. These environments bring us more comfort, improve performance, increase profits, take away time‐consuming tasks. However, besides its great benefits, also a big source concerns, mainly because good part...

VCube is a distributed diagnosis algorithm for virtually interconnecting network nodes. presents several logarithmic properties, and logical hypercube when all nodes are fault-free. dynamic in the sense that can leave rejoin system as they become faulty repaired. The topology re-organizes itself keeps its properties even if an arbitrary number of faulty. Fault based on tests. All fault-free with N detect event latency at most log <sub xmlns:mml="http://www.w3.org/1998/Math/MathML"...

The principle of Network Neutrality (NN) has been debated around the world for nearly two decades. NN states that all traffic in Internet must be treated equally, regardless content, origin and/or destination. main motivation this is to protect fair competition, innovation, and ensure freedom choice consumers. global debate revolves whether should enforced through regulations or not, as well potential impact such – lack thereof on telecommunications market. In context, multiple governments...

This paper describes the architecture and design of Cubrick, a distributed multidimensional in-memory DBMS suited for interactive analytics over highly dynamic datasets. Cubrick has strictly data model composed cubes, dimensions metrics, supporting sub-second OLAP operations such as slice dice, roll-up drill-down terabytes data. All stored in is range partitioned by every dimension within containers called bricks an unordered sparse fashion, providing high ingestion rates indexed access...

This paper presents Virtual-Machines-MIB, a MIB (Management Information Base) directed to virtual machines management through SNMP (Simple Network Management Protocol). Virtual-Machines-MIB aims define standard interface of management, allowing the several monitors, like Xen, KVM and VMWare, with common tool. Different from previous MIBs, which allows manager perform only monitoring operations, control create, delete, restart, turn on, pause shut down machines. It is also possible use...

Open Educational Resources (OER) are educational resources openly available to be used by educators and students an important tool support education. A considerable effort has been made build repositories that allow the sharing reuse of these OERs. However, many offer unsatisfactory search engines, resulting in a frustrating experience for users. The problem content is partly explained lack appropriate metadata on ranking mechanisms, instance based user interaction over objects. In this...

Users of large scale network testbeds often execute experiments that require a set nodes behave and communicate among themselves in reasonably stable pattern. In this work we call such clique, introduce monitoring strategy allows their detection PlanetLab, non-trivial task for dynamic network. Nodes monitor each other by sampling the RTT (Round-Trip-Time) computing its variation. Based on data threshold, pairs are classified as or unstable. A graphs is generated, which maximum sized cliques...

Disk schedulers should provide QoS guarantees to applications, thus sharing proportionally the storage resource and enforcing performance isolation. must execute requests in an efficient order though, preventing poor disk usage. Non-work-conserving help increase throughput by predicting future requests' arrival therefore exploiting spatial locality. Previous work are limited either or exploit In this paper, we propose a new non-work-conserving scheduler called High-throughput Token Bucket...

This work presents a high performance heterogeneous parallel method for encryption using GPUs. Our design executes most of the processes on GPU and partially CPU. Aside from AES 16 Byte block size, our CTR algorithm divides in small logical data blocks. We have devised novel latency hiding scheme that suits mode by speculatively encrypting caching blocks ahead time GPU. The also allows execution all process memory. show that, this only mode, outperforms current published implementations...

Elasticity can be seen as the ability of a system to increase or decrease computing resources allocated in dynamic and on demand way. In order explore this feature, several works addressed development frameworks platforms focusing construction elastic parallel distributed applications for IaaS clouds. However, none these exploration elasticity multithreaded applications. paper, we propose mechanism provide support OpenMP applications, making possible provisioning cloud taking into account...

Network Neutrality is essential for ensuring a level playing field the development of new applications and services on Internet. Laws rules alone might not be enough to protect innovation, fair competition consumer's freedom choice online. The research community has responsibility propose solutions that reveal discriminatory traffic management mechanisms We present potential risks non-neutral Internet, identify several open challenges designing detect differentiation, model addresses such by...

Nowadays, Big Data applications exchange huge amounts of data, highly demanding network guarantees for bandwidth and low latency. However, equipments did not provide a standard interface to control dynamically the resources. Software-Defined Network (SDN) has emerged support programmability, but it provides programming model devoted operators. This paper presents Overlay Framework (NoF), which enables networks be defined based on application requirements. NoF language allows specialists...

The Barnes-Hut algorithm is a widely used approximation method for the N-Body simulation problem. irregular nature of this tree walking code presents interesting challenges its computation on parallel systems. Additional problems arise in effectively exploiting processing capacity GPU architectures. We propose and investigate applicability software Simulated Wide-Warps (SWW) context. To extent, we explicitly deal with dynamic patterns data accesses remapping transformation, by controlling...

This paper presents HyperBone, an overlay network based on a virtual hypercube that offers services such as monitoring and routing, allowing the execution of distributed applications across Internet hypercubes are scalable by definition, presenting several properties symmetry logarithmic diameter, advantageous for parallel applications. HyperBone nodes run algorithm (DiVHA) in order to maintain topology. DiVHA keeps even when number is not power two, or under dynamic fault situation, which...

This work presents a dependable fully distributed network management tool based on the Internet standard protocol, SNMP (Simple Network Management Protocol). Multiple agents running Hi-ADSD with Timestamps, Hierarchical Distributed System-Level Diagnosis algorithm monitor themselves and configurable set of services devices, issuing controlling commands depending results. The system is in sense that it continues working even if only one agent fault-free. A MIB (Management Information Base)...