2022 – 2032
SMARTNESS 2030: SMART NEtworks and ServiceS for 2030.
The SMART NEtworks and ServiceS for 2030 (SMARTNESS) FAPESP Engineering Research Center (ERC) aims to advance cutting-edge research in computer networks and digital application services focused on strategic areas where scientific and technological impacts can be achieved by the year 2030, in collaboration with communities of cloud search and networks. With the deployment of 5G and the 6G vision being developed, the main challenges for SMARTNESS are how to design and operate cloud computing infrastructures and networks with adequate capabilities to leverage the next generation of Internet services and applications. The scope of end-to-end services at the Internet scale is exceptionally broad and requires contributions across multiple disciplines along with large investments in capital and human resources.
Members: Fabio Luciano Verdi (Co-PI) / Christian Esteve Rothenberg (PI) / Luiz Bittencourt / Daniel Batista / Jó Ueyama and many others.
Funding: FAPESP & Ericsson.
2022 – 2024
Design and implementation of an In-Network Load-aware Fast Rerouting mechanism.
Failures in communication networks affect the quality of the services running on the end-hosts. Although some applications may tolerate a certain delay to resolve the failure, others are highly latency-dependent and need fast solutions for traffic rerouting. Fast Rerouting (FRR) is a mechanism used to quickly reroute traffic upon failures completely in the dataplane. Although there are several works related to FRR, there is no current solution that considers the status of the network to define the backup paths. In this work, we benefit from the In-network computing and programmability to design and implement a mechanism entirely in the dataplane to define backup paths based on the load of the network. The mechanism should be aware of the status of the links in terms of, e.g., end-to-end latency to set the best backup paths. Upon a failure, our solution must fast reroute the traffic to such paths without the control plane help. We will develop the mechanism totally in the dataplane by using P4 and evaluate it using real minimalist scenarios as well as simulations.
Members: Fabio Luciano Verdi (PI) / Gustavo Venancio / Leandro Almeida / Thiago Henrique / Marco Chiesa.
Funding: FAPESP.
2017 – 2020
Leveraging the Future Internet through the coexistence of multiple architectures.
The Internet is undoubtedly one of the most important artifacts of human ingenuity. However, its main protocols remain the same as decades ago, preventing its modernization. Concerned with this stagnation, many initiatives around the world started studies of new protocols to replace the current TCP/UDP/IP. These are the so-called Future Internet Architectures (FIA). Among them are the Brazilian proposals ETArch and NovaGenesis (NG). In Europe and the United States there are dozens of projects, including RINA, XIA and CCNx. Each FIA has a particular focus and its own set of requirements, principles, etc. All of them advance in many aspects when compared to the current Internet. The integration of all aspects into a single proposal seems impossible at the current technological stage and many question whether it is really necessary. Therefore, a set of AIFs can be expected to coexist with each other and with the current Internet (including IPv6). This project aims to study scenarios of joint use of NG, RINA, ETArch architectures, specifying and implementing interconnection points, called Future Internet Exchange Points (FIXPs). The project also aims to implement a multi AIF socket in a point of presence. Finally, a simple application will be developed to test the interconnection of FIAs through RNP’s FIBRE network.
Members: Fabio Luciano Verdi (Co-Pi) / Antônio Marcos Alberti / Pedro Frosi Rosa / Augusto Suruagy Monteiro / Flavio de Oliveira Silva / José Augusto Gavazza / Michael Santos .
Funding: FAPESP.
2017 – 2020
NECOS – Novel Enablers for Cloud Slicing
Descrição: The NECOS project addresses the limitations of current cloud computing infrastructures to respond to the demand of new services, as presented in two use-cases, that will drive the whole execution of the project. The first use-case is Telco service provider focussed and is oriented towards the adoption of cloud computing in their large networks. The second use-case is targeting the use of edge clouds to support devices with low computation and storage capacity. The envisaged solution is based on a new concept Lightweight Slice Defined Cloud (LSDC) as an approach that extends the virtualization to all the resources in the involved networks and data centers and provides an uniform management with a high-level of orchestration. The NECOS approach will be manifested in a platform whose main distinguishing features are: 1. The Slice as a Service – a new deployment model. A slice is a grouping of resources managed as a whole and that can accommodate service components, independent of other slices. 2. Embedded algorithms for an optimal allocation of resources to slices in the cloud and networking infrastructure, to respond to the dynamic changes of the various service demands. 3. A management and orchestration approach making use of artificial intelligence techniques in order to tackle with the complexity of large scale virtualized infrastructure environments. 4. Making reality the lightweight principle, in terms of low footprint components deployable on large number of small network and cloud devices at the edges of the network. The NECOS platform will be based on state of the art open software platform, which will be carefully selected, rather than start from scratch. This baseline platform will be enhanced with the management and orchestration algorithms and the APIs that will constitute the research activity of the project. Finally the NECOS platform will be validated, in the context of the two proposed use cases, using the 5TONIC and FIBRE testing frameworks.
Members: Fabio Luciano Verdi (Co-PI) / Cesar Augusto Cavalheiro Marcondes / André Beltrami / Paulo Di Tarso / SERGI ALMENARA .
Funding: RNP/CTIC.
2016 – 2019
Monitoring of physical and virtual resources in cloud computing environments
After a decade of deploying cloud computing infrastructures and virtualized networks, both for experimental purposes and for production environments, we are faced with the need to monitor these environments. Monitoring consists of acting in two aspects: monitoring servers, physical and virtual, and monitoring network devices, also physical and virtual. Monitoring network devices consists of collecting data on resource usage at every interface, flow, and aggregate levels. Through these collected data, it is possible, for example, to readjust the traffic in different routes, avoid packet loss and improve the use of the network. Server monitoring consists of analyzing CPU, RAM and hard disk usage. With this data, management actions can be performed to, for example, migrate virtual machines (VMs) to more idle servers, thus performing load balancing. This project aims to develop and prototype an architecture for monitoring network resources and servers, physical and virtual, in cloud computing environments and virtualized networks, applying policies and actions that improve the use of infrastructure.
Members: Fabio Luciano Verdi (PI) / Rodolfo Vilaça / Alan Silva / Marcelo Frate / André Beltrami / Josep Lluís Lérida Monsó / Hélio Tibagi / Leobino Sampaio / Regis Martins / Alexis Saito / NILSON RUBENS DE MORAES FILHO.
Funding: FAPESP.
2013 – 2016
Study and design for Inter-Data Centers communication
The evolution and growth of large data centers developed by service providers for cloud computing has given rise to standardized and proprietary Layer 2 and Layer 3 routing solutions. We have observed that many solutions optimize existing protocols and technologies to be applied in data centers. At the same time, we realize that new approaches have emerged considering the specific characteristics found in these scenarios. Examples of existing solutions include VL2, BCube, SPB, TRILL and more recently the OpenFlow paradigm. All of these approaches, and others that exist, basically focus on solving problems internal to the data center. However, little has been studied about inter-data center communications in the most diverse aspects involved, such as efficient routing, interconnection and transfer of large masses of data.
Members: Fabio Luciano Verdi (PI) / Cesar Augusto Cavalheiro Marcondes / Rothenberg, Christian Esteve / Yeda Regina Venturini.
Funding: CNPq.
2013 – 2014
Usage of a hybrid approach for virtual machine allocation in data-center topologies.
Improving the efficiency of network traffic from cloud computing infrastructures is today a major challenge for large cloud service providers. This efficiency brings financial gains and also opens the possibility of evaluating the capabilities of the computing infrastructure against the demand for resources in the data centers. This research project seeks to optimize the allocation of virtual machines in virtualized cloud servers, considering the topology of the internal communication network of a data center. For this, we seek to optimize the distribution of these virtual machines in such a way that the traffic between the VMs is considered in the allocation process. The objective is to optimize the use of the network infrastructure in view of the demand for computing resources within the data center. The present work proposes the use of network traffic simulation, mixed integer linear programming, and heuristics to allocate virtual machines in the cloud.
Members: Fabio Luciano Verdi (PI) / Lucio Agostinho Rocha.
Funding: CAPESP.
2011 – 2014
Study and evaluation of network technologies and cloud computing platforms.
Recently there has been a change in the way we interact with services and applications. The (still evolving) paradigm of cloud computing provides services and applications over the Internet with the promise of infinite capacity and pay-as-you-go service models. However, this new computing model requires major changes in the network infrastructure and IT systems of providers, as they need a high level of flexibility to be able to effectively manage a greater computing potential for processing applications, services and data storage. This project aims to investigate the cloud computing paradigm from the point of view of network infrastructures and technologies and development and experimentation platforms. The project consists of using OpenFlow and NetFPGAs boards to study, evaluate and propose solutions for cloud computing.
Members: Fabio Luciano Verdi (PI) / Yeda Venturini / Tiago Pasqualini.
Funding: FAPESP.
2009 – 2012
Collaborative Communication: Study and design of a context-aware architecture for mobile applications.
The diversity of devices available today makes us realize a growing trend that miniaturization and ubiquity are already part of our everyday life. The Internet, when conceived almost 40 years ago, did not imagine providing the global connectivity of the most differentiated islands of applications and contents spread around the planet. These islands of applications not only access information through the Internet, but are also taking on a life of their own, making communication and interaction increasingly localized. In this sense, the term collaborative communication appears to represent the most varied forms of interaction carried out through the devices available today (PDAs, laptops, smartphones, etc.) in a local scope. Examples of this type of scenario include the interaction of devices to exchange content (songs, movies, files, etc.), peer-to-peer communication and opportunistic communication (sensor and vehicle networks). This project aims to investigate collaborative communication and understand the interoperability aspects of devices, taking into account different applications and their use in real scenarios. It is intended to study the different communication technologies including Bluetooth and Wi-Fi and how such technologies can be used to implement collaborative situations. The project also aims to carry out real experiments on the campus of the Federal University of São Carlos, based in Sorocaba, encouraging students of the Computer Science course to invest in the line of research related to collaborative communication.
Members: Fabio Luciano Verdi (PI) / Luciana Zaina / Yeda Venturini.
Funding: CNPq.
