-
Open-VERSO: a vision of 5G experimentation infrastructures, hurdles and challenges
Authors:
Angel Martin,
Pablo Losada,
Carolina Fernández,
Mikel Zorrilla,
Zaloa Fernandez,
Alvaro Gabilondo,
Juncal Uriol,
Felipe Mogollon,
Mikel Serón,
Michalis Dalgitsis,
Roberto Viola,
Luis Roca,
Carlos Giraldo,
Pablo Gonzalez,
Anxo Tato,
Joaquín Escudero,
Alvaro Vazquez,
Daniel Camps,
Andrés Cárdenas,
Carlos Herranz,
Joan Josep Aleixendri,
Rebeca Iglesias,
Gianluca Cernigliaro,
Mario Montagud,
Pau Tomàs
, et al. (1 additional authors not shown)
Abstract:
5G led to a digital revolution for networks by leveraging virtualisation techniques to manage software-based network functions through provided standard interfaces, which have matured recently for cloud infrastructure that is widely employed across domains and sectors. This undiscovered potential to adequately respond to concurrent and specialised traffic demands is promising for a wide spectrum o…
▽ More
5G led to a digital revolution for networks by leveraging virtualisation techniques to manage software-based network functions through provided standard interfaces, which have matured recently for cloud infrastructure that is widely employed across domains and sectors. This undiscovered potential to adequately respond to concurrent and specialised traffic demands is promising for a wide spectrum of industries. Moreover, it exposes the networking ecosystem to prospects beyond the traditional value chain. However, the configuration, deployment and operation of a 5G network are challenging. Thus, different scientific and research entities have built their own open, evolvable and updateable testbed infrastructure that can be used for experimentation purposes. Such testbeds enable different stakeholders to integrate new systems or features exploiting new technologies, assess the performance of innovative services, and customise operation policies to find optimal setups from a cost-effective perspective. Furthermore, federations of infrastructure allow for wider and more complex experiments to be performed in distributed domains. However, numerous technical and procedural obstacles exist during the building of 5G network testbeds. In addition, some technical barriers persist despite the testing of alternatives and ongoing efforts within open-source systems and commercial equipment portfolios. All these limitations and challenges are relevant for experimenters and stakeholders as they attempt to determine the scope of 5G set expectations.
△ Less
Submitted 28 August, 2023;
originally announced August 2023.
-
Supporting RISC-V Performance Counters through Performance analysis tools for Linux (Perf)
Authors:
Joao Mario Domingos,
Pedro Tomas,
Leonel Sousa
Abstract:
Increased attention to RISC-V in Cloud, Data Center, Automotive and Networking applications, has been fueling the move of RISC-V to the high-performance computing scenario. However, lack of powerful performance monitoring tools will result in poorly optimized applications and, consequently, a limited computing performance. While the RISC-V ISA already defines a hardware performance monitor (HPM),…
▽ More
Increased attention to RISC-V in Cloud, Data Center, Automotive and Networking applications, has been fueling the move of RISC-V to the high-performance computing scenario. However, lack of powerful performance monitoring tools will result in poorly optimized applications and, consequently, a limited computing performance. While the RISC-V ISA already defines a hardware performance monitor (HPM), current software gives limited support for monitoring performance. In this paper we introduce extensions and modifications to the Performance analysis tools for Linux(perf/perf_events), Linux kernel, and OpenSBI, aiming to achieve full support for the RISC-V performance monitoring specification. Preliminary testing and evaluation was carried out in Linux 5.7 running on a FPGA booted CVA6 CPU, formerly named Ariane, showing a monitoring overhead of 0.283%.
△ Less
Submitted 22 December, 2021;
originally announced December 2021.
-
PositNN: Training Deep Neural Networks with Mixed Low-Precision Posit
Authors:
Gonçalo Raposo,
Pedro Tomás,
Nuno Roma
Abstract:
Low-precision formats have proven to be an efficient way to reduce not only the memory footprint but also the hardware resources and power consumption of deep learning computations. Under this premise, the posit numerical format appears to be a highly viable substitute for the IEEE floating-point, but its application to neural networks training still requires further research. Some preliminary res…
▽ More
Low-precision formats have proven to be an efficient way to reduce not only the memory footprint but also the hardware resources and power consumption of deep learning computations. Under this premise, the posit numerical format appears to be a highly viable substitute for the IEEE floating-point, but its application to neural networks training still requires further research. Some preliminary results have shown that 8-bit (and even smaller) posits may be used for inference and 16-bit for training, while maintaining the model accuracy. The presented research aims to evaluate the feasibility to train deep convolutional neural networks using posits. For such purpose, a software framework was developed to use simulated posits and quires in end-to-end training and inference. This implementation allows using any bit size, configuration, and even mixed precision, suitable for different precision requirements in various stages. The obtained results suggest that 8-bit posits can substitute 32-bit floats during training with no negative impact on the resulting loss and accuracy.
△ Less
Submitted 14 May, 2021; v1 submitted 30 April, 2021;
originally announced May 2021.
