Výsledky vyhledávání

Report

TURNIP: A 'Nondeterministic' GPU Runtime with CPU RAM Offload

Autor: Ding, Zhimin, Yao, Jiawen, Barrow, Brianna, Botran, Tania Lorido, Jermaine, Christopher, Tang, Yuxin, Li, Jiehui, Yao, Xinyu, Abdelghafar, Sleem Mahmoud, Bourgeois, Daniel

An obvious way to alleviate memory difficulties in GPU-based AI computing is via CPU offload, where data are moved between GPU and CPU RAM, so inexpensive CPU RAM is used to increase the amount of storage available. While CPU offload is an obvious id

Externí odkaz: http://arxiv.org/abs/2405.16283

Zobrazit plný text záznamu

Report

DeePref: Deep Reinforcement Learning For Video Prefetching In Content Delivery Networks

Autor: Alkassab, Nawras, Huang, Chin-Tser, Botran, Tania Lorido

Content Delivery Networks carry the majority of Internet traffic, and the increasing demand for video content as a major IP traffic across the Internet highlights the importance of caching and prefetching optimization algorithms. Prefetching aims to

Externí odkaz: http://arxiv.org/abs/2310.07881

Zobrazit plný text záznamu

Report

MSF-Model: Queuing-Based Analysis and Prediction of Metastable Failures in Replicated Storage Systems

Autor: Habibi, Farzad, Lorido-Botran, Tania, Showail, Ahmad, Sturman, Daniel C., Nawab, Faisal

Metastable failure is a recent abstraction of a pattern of failures that occurs frequently in real-world distributed storage systems. In this paper, we propose a formal analysis and modeling of metastable failures in replicated storage systems. We fo

Externí odkaz: http://arxiv.org/abs/2309.16181

Zobrazit plný text záznamu

Report

IPA: Inference Pipeline Adaptation to Achieve High Accuracy and Cost-Efficiency

Autor: Ghafouri, Saeid, Razavi, Kamran, Salmani, Mehran, Sanaee, Alireza, Lorido-Botran, Tania, Wang, Lin, Doyle, Joseph, Jamshidi, Pooyan

Publikováno v: Journal of Systems Research, 4(1) (2024)

Efficiently optimizing multi-model inference pipelines for fast, accurate, and cost-effective inference is a crucial challenge in machine learning production systems, given their tight end-to-end latency requirements. To simplify the exploration of t

Externí odkaz: http://arxiv.org/abs/2308.12871

Zobrazit plný text záznamu

Report

Keep It Simple: Fault Tolerance Evaluation of Federated Learning with Unreliable Clients

Autor: Huang, Victoria, Sohail, Shaleeza, Mayo, Michael, Botran, Tania Lorido, Rodrigues, Mark, Anderson, Chris, Ooi, Melanie

Federated learning (FL), as an emerging artificial intelligence (AI) approach, enables decentralized model training across multiple devices without exposing their local training data. FL has been increasingly gaining popularity in both academia and i

Externí odkaz: http://arxiv.org/abs/2305.09856

Zobrazit plný text záznamu

Akademický článek

The RICA-2 registry: design and baseline characteristics of the first 1,000 patients

Autor: Sánchez de Alcázar del Río, Á., Rubio, M.I.P., Luis, G.R., López, C.H., Ordóñez, M.Á.G., Llanes, J.O., Salamanca, P., Calle, D.G., Hueso, R.R., Candela, I.B., Bermúdez, D.F., Garcia, M.G., Luque, C.J., Gutiérrez, R.M., Iglesias, A.A., Cabello, F.G., Merlo, A.B., Cano, C.M.S., Sánchez, M., Esterellas, L., Crestelo, A., Pedreira, I.S., Solis, R.A., Vena, D.V., Armenteros, J.M., Valbuena, F.T., Niño, J.R., de Rodríguez de Olmedo, D., Franco, Á.G., Ávila, E.E.R., Martel, A.C., Vallejo, J.MªG., Sosa, S.G., Ramos, M.F.D., Luis, R.H., Garrido, J.G., Sánchez, M.G., Jiménez de la Cruz, L., Sánchez, A.B.M., Alarcón, I.Z., Ruiz De Zuazu, H.F.M., Sánchez, C.S., Bernal, L.V., Subirán, C.C., Izquierdo, M.C., Molinero, A.M., González, Á.L.M., Arxe, A.A., Camps, G.G., Calvo, X.G., Álvarez, M.M.R., Mallén, J.M.C., Biosca, A.S., Parada, D.C.Q., Lombraña, A.S., Parés, A.A., Vilaseca, V.G., Guillen, D.C., Almansa, R.M.B., Puig, E.F., Trullàs, J.C., Serrado, A., Cuartero, D., López, T.M., Muixi, F.F., Carmen Moreno García, Mª, Alquézar, V.M., Navarro, C.P., García, C.S., Luís Corcoles Satorre, J., Aldeguer, M.M., Encinar, J.C.B., Garcia, F.A.C., Albert, E.L., Pérez Silvestre, J., Ariño, A.N., María Cepeda Rodrigo, J., Birlanga, E.M., Alejandro Arazo Alcaide, J., Lorente, L., Carlos Arévalo Lorido, J., Manuel Cerqueiro González, J., del Carmén Gómez Fernández, R., Reboiro, M.L.L., Feijoo, M.L.V., Vila, M.C., Osorio, I.R., Gonzalez, B.S., Freire Castro, S.J., Paz, S.R., Rodríguez, P.V., Trinidad, R.B., Lozano, D.M., Roy, P.M., Rangel, L.S., Ronda, M.V., Yebra, M.Y., Asenjo, M., Cerrada, J.C., Pérez, D.A., Hernández, E.M., Iborra, P.L., Manzano, L., García de la Torre Rivera, M.P., Deodati, F., Bailon, M.M., Cobos, A., Morrás de la Torre, I., Fernández-Bravo, I.G., Martinez-Colubi, M.M., Ángel Satue Bartolomé, J., Barrera, S.G., Rodríguez, F.A., Martínez, J.A., Pérez, A.A., Pirla, T.C., Cano, D.A., Ferreira, J.P., Nóvoa, F., Mascarenhas, J., Pimenta, J., Moreno-García, M.C., Mittelbrunn-Alquézar, V., Conde-Martel, A., Soler-Rangel, L., Montero-Pérez-Barquero, M., Casado, J., Sánchez-Marteles, M., Arévalo-Lorido, J.C., Pérez-Silvestre, J.

Publikováno v: In Revista Clínica Española (English Edition) October 2024 224(8):522-533

Zobrazit plný text záznamu

Report

FIRE: A Failure-Adaptive Reinforcement Learning Framework for Edge Computing Migrations

Autor: Siew, Marie, Sharma, Shikhar, Li, Zekai, Guo, Kun, Xu, Chao, Lorido-Botran, Tania, Quek, Tony Q. S., Joe-Wong, Carlee

In edge computing, users' service profiles are migrated due to user mobility. Reinforcement learning (RL) frameworks have been proposed to do so, often trained on simulated data. However, existing RL frameworks overlook occasional server failures, wh

Externí odkaz: http://arxiv.org/abs/2209.14399

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání