Time-sensitive networking (TSN) is a promising technology for local area applications that require ultra-low latency, such as industrial automation, audio and video bridging, and 5G fronthaul transport. TSN guarantees deterministic packet transport with low latency, low packet delay jitter, and low packet loss. Correct provisioning of end-to-end latency is crucial for achieving optimal network performance and satisfying flow latency constraints. As TSN increasingly supports cloud and fog computing, managing virtualization overhead becomes critical to preserving latency performance. This work analyzes and compares TSN simulated, bare metal, and virtualized network deployments. The study focuses on an important TSN mechanism, namely, the IEEE 802.1Qav credit-based shaper (CBS). It examines its impact on end-to-end latency. Experiments show that the conducted TSN simulations provide latency results close to those from real-world setups, validating simulation as a reliable tool for TSN prototyping. Meanwhile, deploying TSN networks in virtualization-based scenarios presents specific trade-offs. All configurations achieved an average latency of less than 1 ms across all tested scenarios when the TSN idle slope parameter was tuned correctly to prioritize CBS-regulated traffic. Notably, despite the processing overhead introduced by virtualization, optimized configurations, such as those using kernel bypassing, outperformed bare metal deployments in terms of latency. However, this improvement came at the cost of reduced throughput, as fewer packets were forwarded compared to other scenarios. These findings underscore the importance of CBS parameter tuning and highlight the balance between performance and stability in virtualized TSN environments.

The rise of interactive real-time applications has exposed the latency limitations that current networks experience. Recently, the Low Latency, Low Loss, and Scalable Throughput (L4S) architecture emerged to address this problem across the Internet. This paper is the first to systematically review existing L4S experimental development efforts. It synthesizes the current state of research and identifies and outlines several remaining challenges. This analysis maps the primary research themes, dominant algorithms, and common evaluation methodologies, confirming that L4S can deliver on its promise for low-latency in controlled settings. Concurrently, the review summarizes the critical challenges and limitations reported in the literature, particularly concerning partial deployment, real-world wireless performance, and implementation robustness. By providing a structured overview of the technology’s maturity and key open questions, it serves as a guide for future work toward a robust, at-scale deployment of low-latency internet services.

Based on the need to digitalize identification documents per several institutions and companies, the segmentation task of textual information acquired importance. Commonly, Convolutional Neural Networks are applied to solve such problems. Vanilla convolution in deep learning models can provide sub-optimal performance in some learning tasks, such as semantic segmentation. Due to this, more specific proposals may be applied to better fit this context, such as self-calibrated convolutions, which consider unique characteristics of different feature maps. Considering this context, we propose a new fully convolutional network architecture based on U-Net for segmentation associated with self-calibrated convolution. We consider changing all vanilla convolution layers in this new neural network by self-calibrating convolutions. In this work, we evaluate our proposal in three different tasks of document segmentation. All experiments demonstrate an increase in the proposed neural network's segmentation performance compared with traditional U-Net in the analyzed context, but the inference time performance decreases.

Abstract The network function virtualization (NFV) paradigm replaces hardware‐dependent network functions by virtual network functions (VNFs) that can be deployed in commodity hardware, including legacy servers. Consequently, the use of NFV is expected to reduce operating and capital expenses, as well as improve service deployment operation and management flexibility. For many use cases, the VNFs must be visited and invoked following a specific order of execution in order to compose a complete network service, named service function chain (SFC). Nonetheless, despite the benefits from NFV and SFC virtualization technologies, their introduction must not harm network performance and service availability. On the one hand, redundancy is seen by network service planners as a mechanism well established to combat availability issues. At same time, there is a goal to optimize resource utilization in order to reduce operational expenditure. In this article, we share our experience in the design use of a framework, named SPIDER, focused on SFC placement that considers the network infrastructure condition and the required SFC availability to define the placement strategy. The SPIDER monitors the status of infrastructure nodes and links and defines which servers the VNFs should be placed on and the number of redundant replicas needed. We present a proof‐of‐concept of SPIDER using Kubernetes to launch the VNFs as containers. We also use Kubernetes to forward the traffic between the VNFs, composing the service chain. We perform experiments to evaluate the runtime of SPIDER and the SFC delay under different network conditions.

Modern telecommunications networks are characterised by increased dynamism and heterogeneity from the cloud to the edge. To meet these requirements and remain competitive, network operators use software defined networking (SDN) and network functions virtualisation (NFV) to make networks programmable, flexible, and agile. Service function chaining (SFC) enables NFV through the ordered sequencing of VNFs and subsequent steering of flows through them to provide end-to-end services. SFCs must meet availability constraints while avoiding resource wastage due to unnecessary redundancy. This paper proposes a novel approach to evaluate the availability of SFC placement considering the associated redundancy strategies using Stochastic Petri Net (SPN) models. We propose an algorithm to generate and resolve relatively small SPN models for SFC availability in an automatic way and then compare baseline SPN models with this algorithm across five SFC placement scenarios. In all scenarios, the proposed algorithm outperformed baseline SPN models in runtime performance.

Summary A data center infrastructure is composed of heterogeneous resources divided into three main subsystems: IT (processor, memory, disk, network, etc.), power (generators, power transformers, uninterruptible power supplies, distribution units, among others), and cooling (water chillers, pipes, and cooling tower). This heterogeneity brings challenges for collecting and gathering data from several devices in the infrastructure. In addition, extracting relevant information is another challenge for data center managers. While seeking to improve the cloud availability, monitoring the entire infrastructure using a variety of (open source and/or commercial) advanced monitoring tools, such as Zabbix, Nagios, Prometheus, CloudWatch, AzureWatch, and others is required. It is often common to use many monitoring systems to collect real‐time data for data center components from different subsystems. Such an environment brings an inherent challenge stemming from the need to aggregate and organize the whole collected infrastructure data and measurements. This first step is necessary prior to obtaining any valuable insights for decision‐making. In this paper, we present the Data Center Availability (DCA) System, a software system that is able to aggregate and analyze data center measurements aimed toward the study of DCA. We also discuss the DCA implementation and illustrate its operation, monitoring a small University research laboratory data center. The DCA System is able to monitor different types of devices using the Zabbix tool, such as servers, switches, and power devices. The DCA System is able to automatically identify the failure time seasonality and trend present in the collected data from different devices of the data center.

In this paper, we propose the use of reinforcement learning to deploy a service function chain (SFC) of cellular network service and manage the VNFs operation. We consider that the SFC is deployed by the reinforcement learning agent considering a scenario with distributed data centers, where the virtual network functions (VNFs) are deployed in virtual machines in commodity servers. The VNF management is related to create, delete, and restart the VNFs. The main purpose is to reduce the number of lost packets taking into account the energy consumption of the servers. We use the Proximal Policy Optimization (PPO2) algorithm to implement the agent and preliminary results show that the agent is able to allocate the SFC and manage the VNFs, reducing the number of lost packets.

With the continuous growth of the number of mobile devices connected to the Internet, cellular network infrastructure owners are facing several new challenges. The fifth generation of mobile technology (5G) is planned to enable a fully mobile and connected society and to empower socio-economic transformations. In 5G, different scenarios with high and strict requirements are being deployed. To deal with such heterogeneity, more advanced communication services are required. Network Function Virtualization (NFV) combined with distributed micro data centers (MDCs) can improve the management of the data flows through several 5G base stations. On the one hand, NFV provides high scalability without requiring significant architectural changes; and on the other hand, distributed MDCs decentralize architecture management, bringing the computational capabilities nearer to the base stations, hence reducing transmission delay. Nonetheless, the introduction of MDCs raises new points of failures in the architecture, impacting user experience. In this paper, we evaluate the impact of allocating MDCs to host base station functionalities; similarly, network functions run as virtual machines in the MDC, and their failures cannot be ignored. We consider a real footrace scenario as a case study. As packets are lost due to the high mobility of the runners and spectators we examine this metric. From simulations results, we discover that as the number of MDCs increases, the number of failures increases as well. However, the impact of each failure in terms of lost packets decreases considerably when considering scenarios with more MDCs. For example, by increasing the number of MDCs from one to seven, the failure impact in terms of lost packet drops by around 87.62%.

Summary Next‐generation cloud data centers are based on software‐defined data center infrastructures that promote flexibility, automation, optimization, and scalability. The Redfish standard and the Intel Rack Scale Design technology enable software‐defined infrastructure and disaggregate bare‐metal compute, storage, and networking resources into virtual pools to dynamically compose resources and create virtual performance‐optimized data centers (vPODs) tailored to workload‐specific demands. This article proposes four chassis design configurations based on Distributed Management Task Force's Redfish industry standard applied to compose vPOD systems, namely, a fully shared design, partially shared homogeneous design, partially shared heterogeneous design, and not shared design; their main difference is based on the used hardware disaggregation level. Furthermore, we propose models that combine reliability block diagram and stochastic Petri net modeling approaches to represent the complexity of the relationship between the pool of disaggregated hardware resources and their power and cooling sources in a vPOD. These four proposed design configurations were analyzed and compared in terms of availability and component's sensitivity indexes by scaling their configurations considering different data center infrastructure. From the obtained results, we can state that, in general, when one increases the hardware disaggregation, availability is improved. However, after a given point, the availability level of the fully shared, partially shared homogeneous, and partially shared heterogeneous configurations remain almost equal, while the not shared configuration is still able to improve its availability.

The next-generation data center introduces the refactoring of the traditional data center in order to create pools of disaggregated resource units, such as processors, memory, storage, network, power, and cooling sources, named composable system (CSs) with the purpose of offering flexibility, automation, optimization, and scalability. In this paper, we solve an optimization problem to allocate CSs considering next- generation data centers. The main goal is to maximize the CS availability for the application owner, having its minimum requirements (in terms of CPU, memory, network and storage), and available budget as restrictions. This problem is modeled as a bounded multidimensional knapsack problem, and we solve it using Dynamic Programming (DP), and two Soft Computing approaches: Differential Evolution (DE) and Particle Swarm optimization (PSO). We consider two different scenarios in order to analyze heterogeneity and variability aspects when allocating CSs in a data center. Moreover, we also analyze the importance of system components to give directions and priorities of actions to upgrade the system design.

Summary Emergency call services are expected to be highly available in order to minimize the loss of urgent calls and, as a consequence, minimize loss of life due to lack of timely medical response. This service availability depends heavily on the cloud data center on which it is hosted. However, availability information alone cannot provide sufficient understanding of how failures impact the service and users' perception. In this paper, we evaluate the impact of failures on an emergency call system, considering service‐level metrics such as the number of affected calls per failure and the time an emergency service takes until it recovers from a failure. We analyze a real data set from an emergency call center for a large Brazilian city. From stochastic models that represent a cloud data center, we evaluate different data center architectures to observe the impact of failures on the emergency call service. Results show that changing data center's architecture in order to improve availability from two to three nines cannot decrease the average number of affected calls per failure. On the other hand, it can decrease the probability to affect a considerable number of calls at the same time.

Summary To assess the availability of different data center configurations, understand the main root causes of data center failures and represent its low‐level details, such as subsystem's behavior and their interconnections, we have proposed, in previous works, a set of stochastic models to represent different data center architectures (considering three subsystems: power, cooling, and IT) based on the TIA‐942 standard. In this paper, we propose the Data Center Availability (DCAV), a web‐based software system to allow data center operators to evaluate the availability of their data center infrastructure through a friendly interface, without need of understanding the technical details of the stochastic models. DCAV offers an easy step‐by‐step interface to create and configure a data center model. The main goal of the DCAV system is to abstract low‐level details and modeling complexities, becoming the data center availability analysis a simple and less time‐consuming task.

The Internet of Things already comprehends billions of connected smart devices. The related issues are vast as well, ranging from the lack of standardized communication to big data management. This paper presents IoT Management and Monitoring System (IMMS), a solution focused on the management of big environments such as smart cities. The The proposed platform brings a novel database implementation to handle large volumes of data and provide quick responses to queries. Its design allows scalable and distributed deployments, while accepting diverse data types. The system was evaluated both in a real-world testbed, in a university campus, and a simulated environment. Results showed low CPU costs while consuming 1.1GB to store 21.5 million measurements. Besides, its scalability behaved as expected, maintaining its performance adequate as data volume increased.

Current Internet transport protocols are being revisited in an attempt to respond to traffic growth in size, diversity and the emergence of new applications. Designed to reduce Web TCP latency and connection establishment time, QUIC uses UDP and defines its own congestion control. To verify some of QUIC's performance claims, we carried out an extensive set of controlled experiments reflecting Internet traffic conditions by configuring various parameters, such as the round trip time (RTT), the packet loss ratio, web pages, and caching. Surprisingly, our experimental results show that QUIC presented worst page load time (PLT) results than TCP. We next present some of our interpretations of these results.

The Internet of Things has the potential of transforming health systems through the collection and analysis of patient physiological data via wearable devices and sensor networks. Such systems can offer assisted living services in realtime and offer a range of multimedia-based health services. However, lack of service availability, particularly in the cases of emergencies, can lead to adverse outcomes and in the worst case, death. In this paper, we propose an e-health monitoring architecture based on sensors and cloud and fog infrastructure scenarios. Further, we propose stochastic models to analyze how failures impact on the e-health system availability. We analyze four different scenarios and from results, we identify that the sensors and fog devices are the components that have the most significant impact on the availability of the entire e-health system in the scenarios analyzed.

The Internet of Things (IoT) has emerged as one of the prominent concepts in academic discourse in recent times reflecting a wider trend by industry to connect physical objects to the Internet and to each other. The IoT is already generating an unprecedented volume of data in greater varieties and higher velocities. Making sense of such data is an emerging and significant challenge. Infographics are visual representations that provide a visual space for end users to compare and analyze data, information, and knowledge in a more efficient form than traditional forms. The nature of loT requires a continuum modification in how end users see information to achieve such efficiency gains. Conceptualizing and implementing Infographics in an loT system can thus require significant planning and development for both data scientists, graphic designers and developers resulting in both costs in terms of time and effort. To address this problem, this paper presents SiMoNa, a domain-specific modeling language (DSML) to create, connect, interact, and build interactive infographic presentations for loT systems efficiently based on the model-driven develonment (MDD) naradiam.

Many enterprises rely on cloud infrastructure to host their critical applications (such as trading, banking transaction, airline reservation system, and credit card authorization). The unavailability of these applications may lead to severe consequences that go beyond the financial losses, reaching the cloud provider reputation too. However, to maintain high availability in a cloud data center is a difficult task due to its complexity. The power subsystem is crucial for the entire operation of the data center because it supplies power for all other subsystems, including IT components and cooling equipment. Some studies have already proposed models to evaluate the availability of the power subsystem, but none of them are based on standard redundancy models. Standards guide cloud providers regarding availability, points of failure, and watts per square foot based on components' redundancy. This paper proposes RBD and Petri Net models based on the TIA-942 standard to estimate the availability of the data center power subsystem and analyze how failures on power subsystem impact the availability of critical applications. These models are important to resource planning and decision making by the cloud providers, because they may identify which components they ought to invest in order to improve the availability level.

The Network Access Control (NAC) management is a critical task, especially in current networks that are composed of many heterogeneous things (Internet of Things) connected to share data, resources and Internet access. The Software-Defined Networking (SDN) simplifies the network design and operation, and offers new opportunities (programmability, flexibility, dy-namicity, and standardization) to manage the network. Despite this, the access control management remains a challenge, once managing security policies involves dealing with a large set of access control rules, detecting conflicting policies, defining priorities, delegating rights, and reacting against network state changes and events. This work presents the HACFlow, a novel, autonomic, and policy-based framework for access control management in OpenFlow networks. HACFlow aims to simplify and automate the network management allowing network operators to govern rights of network entities by defining dynamic, fine-grained, and high-level access control policies. We analyzed the performance of HACFlow and compared it against related approaches.

The Border Gateway Protocol (BGP) is the current standard protocol for the inter-domain routing system. Although BGP makes easy to manipulate the output traffic by an Autonomous System (AS), it lacks other desired features such as convenient management of incoming traffic to an AS. Thus, we propose a framework to evolve the control plane routing system that is compatible with BGP (smooth transition), uses SDN technologies and can be used to management the incoming traffic of multihomed ASes. Based on our framework, a load balancing application was developed and applied to a scenario to depict the proof of concept of our framework. The results were compared and discussed with BGP AS-Path Prepending technique, shown an improvement of inbound traffic bandwidth utilization.

Standard Port-based Network Access Control (NAS) with tagged Virtual Local Area Networks (VLANs) systems are useful to authenticate users within an isolated network environment. This approach on its own, however, lacks the flexibility and granularity level that new generation networks based on SDN (Software Defined Networking) can provide. The flow-based access control provides a more appropriate granularity to enforce network policies. In this paper, we propose a novel solution named SDN-based Network Access Control (S-NAC) that provides authentication and authorization of clients and servers based on high-level policies enforced at flow level. The solution has been implemented, deployed and tested over emulated and real networks.

Cloud applications are offered to users with high availability and minimal data loss. Any (hardware or software) failure must be detected and recovered quickly, in order to maintain customer trust and avoid financial losses. When we are dealing with multi-tier and stateful applications, the failure recovery process is a big challenge because the whole state of the failed application must be retrieved and restored in a new instance. This process is named as failover; it can be performed by a checkpoint service at application-level or at system-level. Depending on the location of the checkpoint data storage, it can be classified as non-collocated, collocated warm, or collocated hot. This work presents an evaluation between these two checkpoint services in both virtualized and physical environments, considering a multi-tier and stateful application.

With cloud computing consolidation, Platform-as-a-Service (PaaS) has been used as a solution for developing applications with low cost and maximum flexibility. However, an open challenge related to PaaS is the proper handling of multi-tier and stateful applications with support for high availability (HA); and scalability can be considered an essential feature for HA. However, dealing with several instances of the same application that access its state in a common area is not a simple task. This paper presents a novel PaaS framework, named NoPaaS, that supports the deployment of multi-tier and stateful applications assuring their availability according to the Service Availability Forum (SAF) redundancy model. The primary goal of this work is to present NoPaaS framework and prototype, and highlight challenges and open issues when providing multi-tier and stateful applications in high availability clouds.

A data center is divided into three basic subsystems: information technology (IT), power, and cooling. Cooling plays an important role related to data center availability, and a failure in this subsystem may cause an interruption of services. Generally, a redundant cooling subsystem is implemented based on replacing the failed component by the standby one. However, it also can be based on a rotation of computer room air conditioners (CRACs). This paper proposes scalable models that represent the cooling subsystem behavior to evaluate the impact of cooling failures on the data center availability. Models are based on the TIA-942 standard and represent Tiers I and II. We validate our model by comparing our results with the literatures. Our results show that the CRACs' rotation has similar results in availability when compared to the replace strategy.

Cloud computing has gained popularity in recent years due to its pay-as-you-go business model, high availability of services, and scalability. Service unavailability does not affect just user experience but is also translated into direct costs for cloud providers and companies. Part of this costs is due to SLA breaches, once interruption time greater than those signed in the contract generate financial penalties. Thus, cloud providers have tried to identify failure points and estimate the availability of their services. This paper proposes models to assess the availability of services running in a cloud data center infrastructure. The models follow the TIA-942 standard. We propose Tier I and IV models using the Reliability Block Diagram (RBD) to allow modeling of different types of applications, and Stochastic Petri Net (SPN) to represent the failure behavior of information technology (IT) components in a data center. We perform stationary analysis to measure the service availability, and sensitivity analysis to understand which metrics have major impacts on data center availability.

Knowing whether a room is occupied or not is crucial for improving electrical energy efficiency. For instance, if a given room is empty there is no need for the lights to be turned on. Usually in small rooms such as elevator halls, a PIR sensor is used together with the lighting, but as it lacks of precision, people often are left in the dark after a few minutes. In the present work a device capable of counting people in a room is implemented and analyzed. It is also compared with other solutions.

The popularity of Internet video streaming services has dramatically increased, leveraged by the emerging Dynamic Adaptive Streaming over HTTP (DASH) standard. DASH also favors the growth of video content access by mobile devices. However, due to fluctuating bandwidth and high latency, mobile networks, e.g. 3G, remain a challenging environment. The user's Quality of Experience (QoE) in this scenario constantly needs to improve. One way to achieve this is through the development of a better adaptation logic that reduces rebuffering events and delivers higher and more stable video qualities. In this paper, we present DBuffer, a control system developed for improving the QoE during a streaming session through mobile networks. DBuffer estimates the ideal video quality to request, during a 3G streaming session, through client-side buffer analysis. The main component of the control system is a buffer-based state machine, which manages the adaptation logic. Our experimental results show that DBuffer outperforms some of the existing advanced adaptation techniques.

Cloud outages, no matter how long, are responsible for large financial losses. Cloud providers look for solutions that provide high availability even in failure cases. Redundancy has been often used as an alternative; however it has a cost, and depending on the redundancy model, we can have different levels of availability. Beyond that, adding more resources also means adding more complexity and more point of failures. In this paper, we propose a risk-based model for estimating the availability of a redundancy setup. In order to analyze our model, we conduct Monte Carlo simulations and results show that for all models, the number of applications we deploy in a Cloud environment impacts on its availability.

Despite of Cloud infrastructures can be used as High Performance Computing (HPC) platforms, many issues from virtualization overhead had kept them unrelated. However, with advent of container-based virtualizers, this scenario acquires new perspectives because this technique promises to decrease the virtualization overhead, achieving a near-native performance. In this work, we analyzed the performance of a container-based virtualization solution - Linux Container (LXC) - against a hyper visor-based virtualization solution - KVM - under HPC activities. For our experiments, we considered CPU and communication (network and inter-process communication), and results showed the virtualizer type can impact distinctly in performance according to resource used by application.

Active measurements tools transmit probe packets between a sender and a receiver to estimate performance metrics such as round-trip time, jitter and loss. In this paper we evaluate how KVM virtualization affects measurements of performance metrics, specifically the round-trip time (RTT). To understand the impact we investigate the interplay of various environment and measurement parameters with virtualization. A number of experiments are performed in order to investigate which parameters had major impact on the RTT. The paper shows that the measurements are affected by CPU load in the host as well as network load while I/O load seemed to have limited impact.

Several works study the performance of Content Delivery Networks (CDNs) under various network infrastructure and demand conditions. Some go even further while considering demand elasticity by leveraging on Cloud elasticity where each Content Service Provider (CSP) is served only the amount of storage space and network throughput that it needs and pays accordingly. Many of today's CDN implementations are based on these proposals. However, the allocation of virtual servers in cloud computing is not yet very fine-grained and suffers from limitations and time constraints for reallocation. This work proposes an efficient cache slicing mechanism that frees the CDN provider from dedicating a different virtual surrogate for each CSP in each location. This is achieved by dealing internally with the allocation space for each CSP in a shared surrogate, while still providing services in terms of network storage allowing a CSP to select and pay the amount of content that can be stored. We show that the proposed approach deals better with peak surrogate usage, and reduces the total network traffic by around 15% and cross-traffic by as much as 35% under some scenarios.

Content Delivery Networks have gained a popular role among application service providers (ASPs) and infrastructural companies. A CDN is an overlay network that gives more control of asset delivery by strategically placing servers closer to the enduser, reducing response time and network congestion. Many strategies have been proposed to deal with aspects inherent to the CDN distribution model. Though mostly very effective, a traditional CDN approach of statically positioned elements often fails to meet quality of experience (QoE) requirements when network conditions suddenly change. In this paper, we introduce a technique to detect significant changes in a monitored metric of a CDN in order to allow provisioning adaptation of resources, using easy access information with great practical viability. Results show decrease on network resources usage without considerable changes on quality of service (QoS). Furthermore, our technique has similar performance in comparison to an omniscient strategy.

Commercial implementations of adaptive streaming systems are either too aggressive or too conservative, not being able to efficiently deliver video content to the client. Because of these issues, a great number of adaptive streaming systems have been proposed in the academia. They employ different strategies, to select the proper video level that the server should send to the client. One trend is the use of control theory, because it guarantees system stability, and enables simulation and tuning of the system prior to its implementation. However, the current proposals that employ control theory do not take full advantage of this technique, because they do not thoroughly analyze the system parameters and their influence on the overall system performance. Therefore, it is possible that they use suboptimal solutions. In this paper, we propose a control-theoretic approach for selecting the video quality in a DASH environment. We use a controller added to a state machine, to improve the QoE and to maximize bandwidth utilization. We also present a comparison of several system settings performances, by changing the controller types, their gains, and the state machine parameters. Experimental results show the effectiveness of our proposed system for buffer time control and video level smoothness, and that it outperforms a current proposal in the academia.

Network virtualization is seen as the fundamental underlying technology for enabling future Internet services. Mapping virtual networks onto a substrate network is a well-known NP-hard problem. Therefore, some authors propose heuristic based strategies by allocating virtual networks in two uncoordinated steps: mapping virtual nodes and virtual links separately. However, this strategy could result in poor performance. In this work, we propose an algorithm for the joint allocation of virtual nodes and links with the focus on balancing the load in the substrate resources. Such an algorithm is designed to cope with capacity constraints and network delay, it also offers an abstraction to consider functional requirements like geographical location or hardware characteristics. Simulation results show that the proposed algorithm has similar or better performance than other coordinated strategies from the Literature, while requiring a considerably lower processing time, finding solutions up to 269 times faster than others solutions.

Content Distribution Networks (CDN) are a well-known and accepted way to deliver content. Considering the ever growing number of users and also the growing number of small localized CDNs, there is a strong trend for collaboration among them. CDN collaboration may cause cost reduction for CDN providers and better quality of experience for the end user. However, CDN collaboration raises performance issues. This paper presents an analytical view of collaboration among CDNs. Our analytical approach provides studies of collaboration performance through metrics such as cache hit, latency, and number of caches. Then, we present a collaboration overhead representation, considering two collaboration request routing strategies. From our performance analyses, we show that CDN collaboration can reduce response time and could also help a primary CDN deal with a flash crowd event.

Resource management in distributed environments, such as Cloud virtualized networks, is one of the main concerns of infrastructure providers. Despite the many advantages of centralized solutions, their unique point of failure and the lack of scalability make us consider the use of distributed and autonomic solutions. Beyond this, centralized solutions would be inefficient to deal with dynamic and peak situations. By using simple rules, each autonomic node can independently act and react according to the scenario's changes without the knowledge of the whole system, making autonomic management robust and adaptable. The main goal of this paper is to present the Role-Based Self-Appointment for Live Streaming (RBSA4LS) prototype, developed considering Software Defined Network technologies. The prototyped was developed and used to evaluate RBSA4LS performance. Our results show a significant decrease in network traffic and an increase of client Quality of Experience (QoE).

Dynamic system adaptation to continuously changing business requirements and environment conditions has been a common demand for IT over the last years. In spite of constant innovation in service computing, current technology is not able to keep up with business ever growing need faster time to market. The Service Refinement Cycle (SRC) has been proposed for providing flexibility and power for creating new services by composing existing services with integrated policies. We developed and deployed a Service Execution Environment (SEE) for the SRC using existing technology, such as Web Services, XACML and WS-BPEL. As a proof of concept prototype, we also developed an Online Radio Service that makes extensive use of the key SRC features and is executed by our SEE.

Considerable effort has been made by researchers in the area of network traffic classification, since the Internet is constantly changing. This characteristic makes the task of traffic identification not a straightforward process. Besides that, encrypted data is being widely used by applications and protocols. There are several methods for classifying network traffic such as known ports and Deep Packet Inspection (DPI), but they are not effective since many applications constantly randomize their ports and the payload could be encrypted. This paper proposes a hybrid model that makes use of a classifier based on computational intelligence, the Extreme Learning Machine (ELM), along with Feature Selection (FS) and Multi-objective Genetic Algorithms (MOGA) to classify computer network traffic without making use of the payload or port information. The proposed model presented good results when evaluated against the UNIBS data set, using four performance metrics: Recall, Precision, Flow Accuracy and Byte Accuracy, with most rates exceeding 90%. Besides that, presented the best features and feature selection algorithm for the given problem along with the best ELM parameters.

Over the past years, multimedia applications have dramatically increased their share in the Internet traffic composition. Therefore, there is a strong interest from both industry and academia to develop techniques for improving user's quality of experience for multimedia services, especially video content. Adaptive streaming is a technology that allows adaptation of video quality to the current network conditions and to the client device. In this paper we characterize the network traffic behavior of two popular adaptive streaming environments, namely Netflix and YouTube. In our experiments we observed their behavior under different constraints and varying network conditions. We found out that Netflix has overall better performance, even though both players behave poorly under bandwidth and packet loss constraints. In the delay constrained scenario, they nearly reached their regular behavior.

The Internet has rapidly evolved, becoming a virtual place where it is simple to handle tasks that were known to be cumbersome or complex beforehand. However the development of new services remains a time consuming activity. This paper presents E2ECloud, a computing environment that allows the development of solutions for creation and provision of composite services in the cloud. Users create a new service through a SaaS tool, which generates a Service Code and deploys it onto an IaaS to be executed. A prototype of E2ECloud was developed and qualitative and quantitative evaluations were conducted, highlighting the potential of our approach to the creation of a new market for cloud service development.

Network Management depends on precise characterization of the traffic profile of networked applications. When the identification and classification of network flows is done using machine learning, the characterization of traffic still requires an approach that is capable of providing a balance between accuracy and processing speed in real-time scenarios. This paper proposes an architecture to classify network traffic based on Stream Data Mining techniques using Graphic Processing Units (GPU), in order to meet the requirements of both classification accuracy and speed. Our proposal combines the characteristics of data mining techniques with a continuous stream of input data, and with high processing performance GPU architecture. Results show that our approach provides accuracy comparable to or better than existing related work (e.g., above 95%) while ramping up performance (e.g., up to 62x speed up), comparing the different implementations of our approach. These facts allow the deployment of the proposed technique to the real-time management of high speed backbone links.

Traffic classification and identification plays an important role for several activities in network traffic management, where DPI (Deep Packet Inspection) is one of the most accurate and used techniques. However, inspection of packet payload is highly computing intensive. Several research studies have evaluated different components of DPI systems for application detection, in order to increase the classification speed. Nonetheless, the arrangement of the signatures in the signature set is an open issue and can degrade performance. Depending on the order of signatures, the overall performance of the DPI system can be degraded, leading to loss of packets and incorrect traffic identification. To the best of our knowledge, no previous research has analyzed the impact of the order of the application signatures and how it could be modified to improve the identification speed in a given DPI. In this work, we evaluate the impact of the ordering of signatures in a list and propose a method to dynamically adapt the signature list according to the traffic dynamics. We show the effectiveness of our approach with the most reactive proposed setup, saving more than 50% of processing time. We demonstrate the importance of the order of signatures and propose an effective method that can be used to save processing time. Finally, our method can be combined with other state-of-the-art techniques to achieve an optimal utilization of DPI features.

Compressed Deterministic Finite Automata (DFA) promises same representation power as traditional DFAs while using less memory for representing Regular Expressions (RE). Experimental evaluations of DFA-based Deep Packet Inspection (DPI) systems focus mainly on memory consumption without observing other important related aspects, such as the matching speed. Proper design of DPI systems requires the assessment of several performance metrics at hardware level, in order to make sure that its implementation will not compromise the overall performance. This paper proposes a novel and systematic evaluation of DPIs and reveals the impact of DFA's data-structures and the correspondent memory layout implementation to hardware-level metrics. Experimental results show that some DFA model and memory layout combinations are almost 100 times faster than others. Results also show that choosing the incorrect model-layout pair can lead to significant performance issues. Our methodology and results will certainly help researchers and developers to design efficient DPI engines, through the selection of the best DFA model and memory layout combination to achieve the targeted overall performance.

Considerable effort has been made by researchers in the area of network traffic classification, since the Internet grows exponentially in both traffic volume and number of protocols and applications. The task of traffic identification is a complex task due to the constantly changing Internet and an increase in encrypted data. There are several methods for classifying network traffic such as port-based and Deep Packet Inspection (DPI), but they are not effective since many applications use random ports and the payload could be encrypted. This paper proposes an Optimized Hybrid Model (OHM) that makes use of a rule-based model (Apriori) along with a self-organizing map (SOM) model to tackle the problem of traffic classification without making use of the payload or ports. The proposed method also allows the generation of association rules for new unknown applications and further labeling by experts. Besides that, a optimizer called Firefly Algorithm was also used to enhance the results by optimizing both Apriori and SOM parameters and a comparative study was performed on both optimized and non-optimized models. The OHM showed to be superior to a non-optimized model for both eMule and Skype applications, reaching levels superior to 94% for correctness rate. The OHM was also validated against another model based on computational intelligence, named Realtime, and the OHM proposed in this work presented better results when tested in real time.

The last two decades have witnessed major advances in the wireless communication area, which have been supported by the increasing capabilities of portable devices, such as notebooks, tablets, cellular phones, smart phones and personal digital assistants (PDAs). Today, they provide a rich computing environment and incorporate different radio technologies. Despite of network diversity, a great deal of research in the field of mobile ad hoc networks focuses on a single and isolated wireless technology at a time, with homogeneous elements identified by IP addresses. This work proposes a Heterogeneous Technology Routing Framework (HTR), targeted towards scenarios where the heterogeneity of devices and networking technologies is present. Our contribution is twofold. It consists of a framework, which encompasses a process for bootstrapping networks, and a routing protocol capable of dealing with multiple network interfaces. We then evaluate the performance of this protocol by taking measurements in a real test-bed.

Traffic Identification is a key function performed by Internet Service Providers' (ISP) administrators to evaluate and improve network services. However, traffic identification needs to be done in real-time and at wire speed to be useful for network tuning. Deep Packet Inspection (DPI) is widely used for identifying normal applications and attacks in the network by looking for well-known patterns within the packets. Such patterns are mostly expressed by Regular Expressions (RE), which are then evaluated by abstract machines known as Deterministic Finite Automata (DFA). Some previous studies grouped DFAs together to evaluate multiple patterns on a single DFA match's run. Efficient grouping algorithms would combine several DFAs without exceeding the available machine's memory. This work proposes and evaluates a new method to combine several DFAs into a single one. Additionally we compared this algorithm to state-of-the-art approaches using a compressed DFA model. Experimental results show that our algorithm generates less groups and transitions than existent algorithms.

Current network traffic analysis systems heavily rely on Deep Packet Inspection (DPI) techniques, such as Finite Automata (FA), to detect patterns carried by regular expression (regex). However, traditional Finite Automata cannot keep up with the ever-growing speed of the Internet links. Although there are a number of efficient FA compressing mechanisms for DPIs, there is no standardized or common way to evaluate and compare them. In this scenario, this paper proposes a methodology to evaluate and compare automaton models and the data-structures that materialize them. We also adapt state-of-the-art memory layouts to better fit in today's computer architectures. Finally, we apply our methodology to most important automaton models, memory layouts, and well-known signature sets. The results show us that some memory layouts are not efficient for regexes that represent small automata and other ones which fit only with uncompressed automata. Further, we also found out that theoretical studies about memory usage from memory encodings are not as accurate as they should be.

In this paper we introduce a novel framework for traffic identification that employs machine learning techniques focusing on the estimation of multiple traffic influencing factors. The effect of these factors is handled with the training of several machine learning models. We utilize the outcome of the multiple models via a recombination algorithm to achieve high overall true positive and true negative and low overall false positive and false negative classification ratio. The proposed method can improve the performance of every kind of machine learning based traffic identification engine making them capable of efficient operation in changing network environment i.e., when the probing node is trained and tested in different sites.

Deep Packet Inspection (DPI) systems have been becoming an important element in traffic measurement ever since port-based classification was deemed no longer appropriate, due to protocol tunneling and misuses of well-defined ports. Current DPI systems express application signatures using regular expressions and it is usual to perform pattern matching through the use of Finite Automaton (FA). Although DPI systems are essentially more accurate, they are also resource-intensive and do not scale well with link speeds. Looking to this area of interest, this paper proposes a novel Deterministic Finite Automaton, called Ranged Compressed Deterministic Finite Automaton (RCDFA), that compresses transitions without additional memory lookups. Experimental results show that RCDFA yields space savings of 97% over the original DFA and up to 93% better compression when compared to the DFA's state-of-the-art compression techniques.

In this paper we present implementation guidelines of the Distributed Cloud Resource Allocation System (D-CRAS). D-CRAS ensures an automatic monitoring and control of resources to guarantee the optimal functioning of the Cloud while meeting developers' requirements. Additionally, this work establishes the necessary technologies that meet the construction of the proposed system.

Autonomic Clouds emerge as a result of applying four self-management properties of Autonomic Computing (self configuration, self-healing, self-optimization, and self-protection) in Cloud environment. In this way, Autonomic Cloud Computing is seen as a Cloud with autonomy to take important decisions about resource management, such as the resource allocation for incoming requests, and the optimization decisions of resource utilization. The main goal of this paper is to make a brief discuss about issues of creating self-adaptive systems for Autonomic Clouds, focusing on infrastructure management level.

Distributed Clouds, or just D-Clouds, can be seen as a paradigm that is able to exploit the potential of sharing resources across geographic boundaries and provide latency-bound allocation of resources to third-party developers. The representation of D-Cloud resources is a challenge that involves the careful choice of characteristics that drive the mapping of requests on the substrate resources. Regarding these problems, this paper introduces the Cloud Modeling Language (CloudML), a vendor-neutral XML-based language intended to integrate the description of different cloud related aspects such as computational and network resources, services profiles, and developers' requests in an integrated way. Furthermore, the CloudML provides a way to describe geographical location aware services, seen particularly indispensable in D-Cloud scenarios.

Traffic identification and classification are essential tasks performed by Internet Service Provider (ISPs) administrators. Deep Packet Inspection (DPI) is currently playing a key role in traffic identification and classification due to its increased expressive power. To allow fair comparison among different DPI techniques and systems, workload generators should have the following characteristics: (i) synthetic packets with meaningful payloads; (ii) TCP and UDP traffic generation; (iii) a configurable network traffic profile, and (iv) a high-speed sending rate. This paper proposes a workload generator framework which inherits all of the above characteristics. A performance evaluation shows that our flexible workload generator system achieves very high sending rates over a 10Gbps network, using a commodity Linux machine. Additionally, we have configured and tested our workload generator following a real application traffic profile. We then analyzed its results within a DPI system, proving its accuracy and efficiency.

Service composition is an important feature enabled by Service Oriented Computing but it remains mainly a manual process. Proposals for complete or partial automation of composition exist, the majority relying on artificial intelligence-based planning, automated proof techniques, or graph-based solutions. Although sound, these approaches present practical issues that prevent their use in production scenarios. This paper presents a composition tool that allows for partial service specification along with a composition algorithm based on the semantic matching of services inputs and outputs. Early performance numbers are also presented.

There are several strategies in the literature to geolocate Internet hosts, however the accuracy of their estimates is a common issue. Furthermore, accurate Internet host location remains a challenge due to incomplete domain registers. In this paper, we propose a classification for geolocation strategies and present the Whois Based Geolocation (WBG) approach, a strategy developed to geolocate Internet hosts given their IP address information. This new mechanism is based on different strategies combination and uses a heuristic to improve the results. An analysis that considers both accuracy and completeness is made showing the effectiveness of our tool.

This paper discusses the modeling of the effect of noises which are, in the time-domain, non-Gaussian. We propose models for their effect on each frequency based on simple assumptions about their time-domain characteristics. The models show analytically that composite noises made, in the time-domain, of Gaussian and non-Gaussian sources may have, on the QAM-symbols, an effect whose distribution is still Gaussian.

A 3D underlay routing protocol, for scenarios with device and networking technology heterogeneity is presented. It provides mechanisms for the seamless interoperation of dynamic, heterogeneous ad hoc networks. This underlay protocol aggregates and exchanges device context information, such as hardware resources, channel state information and network interfaces to build its routing decisions. 3D Routing considers the following fundamental emergency networking aspects: the bootstrapping process of the network(s) and nodes, the building and spreading of context information between the nodes, and the assignment of roles to particular nodes. To this end, we discuss how these ideas are designed and implemented in order to provide seamless interoperability for this type of emerging networking scenarios.

Service composition is today mostly a manual process and automated service creation techniques are not ready for prime time in IT and network system development. This paper presents the Service Refinement Cycle as an integrated framework for dealing with the service lifecycle composed of stages like service creation, deployment, execution and management. The main realization of the cycle is the Service Code, which contains executable specifications (code) that may run at any compatible engine.

In this paper we share our experience in building an ADSL video delivery test-bed. As a result, we are able to measure the impact of important parameters such as the loop length, the presence of background traffic and line protection against transmission errors on video quality. We show that although these access technologies promise a broadband pipe and low delay communication, a careful configuration of line parameters is paramount to achieve optimal access and is needed for the success of multimedia triple play (3P) services. We particularly analyze the delay, jitter, packet loss and bitrate consumption obtained from real ADSL measurements. These results are an important step towards understanding the adequate deployment of such services. As a special case, we study the effect of repetitive noise on SDTV.

Recently, we have been faced the rise of several technologies that aim to communicate easier with the users. Solutions to support specific requirements, such as mobility and security, were designed. However, few works can be found regarding to automate the discovery of environment features, as well as to automate the network configuration to allow a better utilization of the network resources. This also includes the adaptation to changes in the network environment and adapting to new requirements from the users. In this paper, we propose XDNM, a method to discover network elements' capabilities in a network and negotiate configurations to be used by them and also the optimization of these configuration according to changes mentioned above. Some questions related with the implementation of this kind of solution are also discussed in the paper in order to give a more practical view of this method in current technologies.

Ambient Networks propose the concept of network composition for providing dynamic and instant resource access for roaming users. policy-based management (PBM) is a good candidate for a service management platform in such scenarios. However, PBM infrastructures for large-scale service management in 3G/4G environments are still to be developed. This paper presents a prototype implementation of a PBM approach for network composition that uses a tree-based information model mapped on a peer-to-peer storage mechanism. A preliminary, measurement-based performance analysis of the prototype is presented along with a discussion on the strategies used, their benefits and pitfalls.

The management of large-scale 3G and 4G wireless systems, such as ambient networks (AN), in the Internet is a challenge, due to their inherent heterogeneous and distributed nature. In this paper we present our experience with the design and implementation of PBMAN, a Policy-based Management framework for AN, which gives special attention to network composition. We discuss the main choices as well as the lessons we learned during this process and the next steps.

Recently, peer-to-peer (P2P) streaming applications have drawn attention to both users and research community. The P2P approach for IPTV has the appealing feature of providing TV-like quality content distribution service (for both on-demand and live TV), which has been attracting a large number of users from across the globe. However, there is very little understanding about their network architecture and resource consumption. This work aims at having an in-depth understanding of Joost, a video-on-demand P2P system for distributing TV content, created by Skype and Kazaa founders. Results show that Joost relies on UDP for content delivery with an asymmetric bandwidth consumption of until 750 Kbps (download) and 480 Kbps (upload). We also evaluated the Joost control plane, which is mainly based on TCP. Along with other P2P IPTV technologies, such systems could have a great impact on network bandwidth utilization, even more than P2P file transfer ones.

The problem of access selection (AS) for multi-access networks has for long been addressed by both the standardization and research communities. As a result, a number of papers have proposed efficient AS algorithms that can take into account radio resource efficiency, overall capacity and quality of service (QoS) requirements in a multi-service environment. However, only a few works have developed on-line AS algorithms that do not require a priori knowledge of the traffic mix when delay sensitive (e.g. voice) and best effort (data) applications are supported. In this work, we present an online AS algorithm that performs well in a multiaccess network supporting two service classes and specifically takes into account the elastic nature of data applications. We model AS as a binpacking problem and realize that the problem is NP-complete. Therefore, we develop a heuristic algorithm called LessDamage that calculates a damage parameter and uses it as a metric for the allocation strategy. Simulation results show that LessDamage performs better in terms of blocking probability and elastic data throughput than available online binpacking heuristics, independently of the number of the available access technologies.

This article presents a routing protocol that uses physical location information in diverse ways. This information can generate local working sub-areas, limiting the target of messages that pass through the network. Also it allows the exchange of data between different sub-areas, through the use of nodes that control the borders of the work areas. The knowledge of neighboring areas allows routing based on physical location and the exchange of information only to nodes that are physically adjacent

It is expected that future wireless systems will consist of several distinct radio access technologies (including WCDMA/HSDPA, GSM/EDGE/GPRS, WLAN and others) forming a multi-access system that offers advanced voice and multimedia services. Previous works have shown that the combined capacity region of such systems depend on the service allocation policy that assigns user sessions to the available subsystems. The currently available service allocation policies typically operate off-line implying that the actual service mix is assumed to be known prior to the service allocation taking place. In this paper we consider the on-line problem according to which sessions arrive one after the other and no assumptions on the service mix can be made. We adopt four on-line bin-packing algorithms to the multi-access environment and study their performance by means of simulation in terms of the class-wise blocking probability and throughput. We find that the algorithm termed Less Voice provides the best performance in terms of the blocking probabilities and imposes the least slow down for elastic sessions

Fieldwork environments possess diverse restrictions, such as lack of infrastructure and low capacity devices. This article describes an architecture comprised of middleware and services designed to support fieldwork requirements. Services have been projected to support the diversity of devices types, with diverse mobility patterns, in a way to not overload devices with lesser battery and processing capabilities. The data routing inside the work area considers the aspects of types of devices, indicating to the content adaptation service the types of nodes in the environment, for the content generation not to overload the weaker devices. The division of the work area into context areas also minimizes the coverage area of each work group. The use of diverse location information update strategies also reduces the data load that passes through the environment.

This paper presents how to model and carry out the performance analysis of message-oriented middleware (MOM) using generalised stochastic Petri Net (GSPN) models. The results obtained from the Petri net analysis are compared against ones measured in a commercial MOM. Additionally, some results are presented in order to demonstrate the flexibility and the benefits of the proposed model. This research also focuses on how to improve the MOM performance by suggesting appropriated techniques and adjustments to the MOM basic architecture. Finally, we point out some decisions usually taken by systems administrators that may have a major impact on the performance of MOM systems.

The emergence of Bluetooth as a default radio interface has allowed handheld electronic devices to be instantly interconnected as ad hoc networks. These short-range ad hoc wireless networks, called piconets, operate in the unlicensed 2.4-GHz ISM (Industrial-Scientific-Medical) band where devices may be used to configure single or overlapping piconets, known as scatternet. As all piconets operate in the same frequency band, the presence of multiple piconets in the vicinity may create interference on signal reception. This paper employs a signal capture model to study the piconet MAC performance, taking inter-piconet interference into consideration. This model leads to several important mathematical relationships for Bluetooth networks, including successful packet transmission probability. Furthermore, our model and anticipated throughput are validated using extensive simulation. These results indicate that Bluetooth throughput is affected by multiple piconet interference. Definitely, our model can be considered to provide a solid foundation for future interference aware Bluetooth protocols.

The emergence of Bluetooth as a default radio interface has allowed handheld electronic devices to be instantly interconnected as ad-hoc networks. These short range ad-hoc wireless networks, called piconets, operate in the unlicensed 2.45 GHz ISM (industrial-scientific-medical) band where up to eight devices may be used to configure single or overlapping piconets. As all piconets operating in the same frequency band, the presence of multiple piconets creates interference on signal reception. This paper employs a signal capture model to study the piconet MAC performance, taking inter-piconet interference into consideration. Furthermore, extensive simulation is performed to validate the throughput obtained from our model. These results indicate that Bluetooth throughput is sensibly affected by multiple piconet interference.

The emergence of Bluetooth as a default radio interface allows handheld electronic devices to be instantly interconnected into ad hoc networks. These short range ad-hoc wireless networks, called piconets, operate in the unlicensed 2.45 GHz ISM (Industrial-Scientific-Medical) band where up to eight devices may be used to configure single or overlapping piconets. This creates interference on the device from other devices operating in the same frequency band including microwaves and devices enabling various wireless LAN standards. This paper uses a signal capture model to study piconet MAC performance. Furthermore, simulations are used to validate the throughput obtained from this model. These results reveal important performance implications of the effect of inter-piconet interference on throughput.

This paper presents a simulation environment for mobile cellular networks called CELSA, which implements a mobility model based on users, regions and time period characteristics. This environment was used to obtain quality of service (QoS) parameters of a mobile cellular network and to evaluate the performance of a non-uniform channel allocation scheme called CA-STV compared to fixed and dynamic channel allocation.

The advent of second and third generation cellular systems forces cellular operators to face a considerable increase in the signaling traffic over their mobile networks. Service providers are therefore required to have appropriate planning tools in order to maintain adequate service quality levels. This paper presents a proposal of a simulation architecture for the analysis of signaling in GSM and IS41/136 mobile cellular networks. Procedures like power up, power down, call setup, handover and location updating are implemented. The simulation tool CELSA (cellular signaling analyzer) has implemented this architecture. Simulation results for exchanged signaling load in an example network architecture are presented.

The conventional approach to reliable multicast in computer networks relies on the retransmission on demand of lost packets. Existing multicast protocols adopt a static packet retransmission scheme (unicast or multicast) to retransmit these packets often leading to performance loss due to wasting bandwidth resources. This is worse if we consider the limited bandwidth resources wireless interfaces currently offer. Therefore, careful continuous monitoring and control is fundamental in these wireless mobile environments. This paper looks at solutions to the retransmission problem while taking into consideration network topology and the number of fixed and mobile users. Precise conditions and rules for packet retransmission are set taking into account the dynamics of the network state. Finally, this algorithm has been incorporated into a reliable multicast protocol with support to mobile users, known as the reliable mobile multicast protocol (RM2), for performance analysis.

Current distributed computing is based on rigid paradigms such as client server computing, that lack run-time adaptive behaviour. The designer of such systems has to consider all possibilities during the design stage and the services that are offered remain constant until a new re-design is done. Although there have been attempts to provide different programming languages that consider programming both distributed and non distributed applications the same, the two remain however different. New paradigms are needed that consider factors such as dynamic adaptability, interaction, ability to monitor and react to the environment, mobility, network latency, partial failure and concurrency. We show that the agent based programming paradigm, agent-ware, is closer to fulfilling some of these requirements and describe the proposed Agentware Virtual Machine (AVM), a generic distributed agent-based computing environment. With the emergence of intelligent agent-based technology, a number of distributed applications are being re-engineered taking advantage of this new design methodology.