Search results
-
Pay Migration Tax to Homeland: Anchor-based Scalable Reference Counting for Multicores
Seokyong Jung, Jongbin Kim, Minsoo Ryu, Sooyong Kang, and Hyungsoo Jung, Hanyang University The operating system community has been combating scalability bottlenecks for the past 10 years with victories or all the then-new multicore hardware. File systems ...admin - December 2, 2021 - 11:30 pm
-
SLM-DB: Single-Level Key-Value Store with Persistent Memory
Young-ri Choi, UNIST This paper investigates how to leverage emerging byte-addressable persistent memory ...admin - December 2, 2021 - 11:30 pm
-
Cluster storage systems gotta have HeART: improving storage efficiency by exploiting disk-reliability heterogeneity
configured in a one-scheme-for-all fashion. In this paper, we make a case for exploiting reliability ...admin - December 2, 2021 - 11:30 pm
-
Speculative Encryption on GPU Applied to Cryptographic File Systems
Vandeir Eduardo, Federal University of Paraná and University of Blumenau; Luis C. Erpen de Bona and Wagner M. Nunan Zola, Federal University of Paraná Due to the processing of cryptographic functions, Cryptographic File Systems (CFSs) may require signific ...admin - December 2, 2021 - 11:30 pm
-
tensorflow-tracing: A Performance Tuning Framework for Production
Sayed Hadi Hashemi, University of Illinois at Urbana-Champaign and National Center for Supercomputing Applications; Paul Rausch; Benjamin Rabe, University of Illinois at Urbana-Champaign and National Center for Supercomputing Applications; Kuan-Yen Chou, ...admin - December 2, 2021 - 8:30 am
-
MLOp Lifecycle Scheme for Vision-based Inspection Process in Manufacturing
an operational lifecycle. In this paper, a practical machine learning operation scheme to build the ...admin - December 2, 2021 - 8:30 am
-
Transfer Learning for Performance Modeling of Deep Neural Network Systems
another. In this paper, we perform an empirical study to understand the effectiveness of different transfer ...admin - December 2, 2021 - 8:30 am
-
Low-latency Job Scheduling with Preemption for the Development of Deep Learning
Hidehito Yabuuchi, The University of Tokyo; Daisuke Taniwaki and Shingo Omura, Preferred Networks, Inc. Efficient job scheduling of trial-and-error (TE) jobs is a challenging problem in deep learning projects. Unfortunately, existing job schedulers to dat ...admin - December 2, 2021 - 8:30 am
-
Towards Taming the Resource and Data Heterogeneity in Federated Learning
owning a partial set of all the data sources could hope to learn in isolation. In this paper, we study ...admin - December 2, 2021 - 8:30 am
-
Disdat: Bundle Data Management for Machine Learning Pipelines
Ken Yocum, Sean Rowan, and Jonathan Lunt, Intuit, Inc.; Theodore M. Wong, 23andMe, Inc. Modern machine learning pipelines can produce hundreds of data artifacts (such as features, models, and predictions) throughout their lifecycle. During that time, data ...admin - December 2, 2021 - 8:30 am
-
Opportunities and Challenges Of Machine Learning Accelerators In Production
This paper discusses some high level observations from experience internally at Google. Rajagopal ...admin - December 2, 2021 - 8:30 am
-
KnowledgeNet: Disaggregated and Distributed Training and Serving of Deep Neural Networks
environment. In this paper, we propose KnowledgeNet (KN), which is a new architectural technique for a simple ...admin - December 2, 2021 - 9:30 am
-
Katib: A Distributed General AutoML Platform on Kubernetes
Jinan Zhou, Andrey Velichkevich, Kirill Prosvirov, and Anubhav Garg, Cisco Systems; Yuji Oshima, NTT Software Innovation Center; Debo Dutta, Cisco Systems Automatic Machine Learning (AutoML) is a powerful mechanism to design and tune models. We present Ka ...admin - December 2, 2021 - 9:30 am
-
Stratum: A Serverless Framework for the Lifecycle Management of Machine Learning-based Data Analytics Tasks
cloud-fog-edge spectrum. This paper describes the Stratum architecture highlighting the problems it resolves. ...admin - December 2, 2021 - 9:30 am
-
TonY: An Orchestrator for Distributed Machine Learning Jobs
configurations, monitoring, and fault tolerance. In this paper, we describe TonY, an open-source orchestrator for ...admin - December 2, 2021 - 6:30 am
-
MPP: Model Performance Predictor
Sindhu Ghanta, Sriram Subramanian, Lior Khermosh, Harshil Shah, Yakov Goldberg, Swaminathan Sundararaman, and Drew Roselli, ParallelM; Nisha Talagala, Pyxeda AI Operations is a key challenge in the domain of machine learning pipeline deployments involving ...admin - December 2, 2021 - 6:30 am
-
Continuous Training for Production ML in the TensorFlow Extended (TFX) Platform
degrade the quality of downstream services supported by these models. In this paper we describe the ...admin - December 2, 2021 - 6:30 am
-
Shooting the moving target: machine learning in cybersecurity
Ankit Arun and Ignacio Arnaldo, PatternEx We introduce a platform used to productionize machine learning models for detecting cyberthreats. To keep up with a diverse and ever-evolving threat landscape, it is of paramount importance to seamlessly iterate o ...admin - December 2, 2021 - 6:30 am
-
Deep Learning Inference Service at Microsoft
Oltean, Maya Mosyak, Chris Barnes, Thomas Liu, and Junhua Wang, Microsoft This paper introduces the Deep ...admin - December 2, 2021 - 6:30 am
-
Accelerating Large Scale Deep Learning Inference through DeepCPU at Microsoft
significant improvement to many Microsoft services and products. In this paper, we introduce our experience ...admin - December 2, 2021 - 6:30 am
-
Caching in the Multiverse
predict one future, we can use it to select a future that is best for caching. This paper provides ...admin - December 1, 2021 - 6:30 pm
-
Respecting the block interface – computational storage using virtual objects
Ian F. Adams, John Keys, and Michael P. Mesnier, Intel Labs Computational storage has remained an elusive goal. Though minimizing data movement by placing computation close to storage has quantifiable benefits, many of the previous attempts failed to take ...admin - December 1, 2021 - 6:30 pm
-
Jungle: Towards Dynamically Adjustable Key-Value Store by Combining LSM-Tree and Copy-On-Write B+-Tree
critical to server applications. This paper suggests a novel approach that transplants copy-on-write ...admin - December 1, 2021 - 6:30 pm
-
Analyzing the Impact of GDPR on Storage Systems
Aashaka Shah, University of Texas at Austin; Vinay Banakar, Hewlett Packard Enterprise; Supreeth Shastri, Melissa Wasserman, and Vijay Chidambaram, University of Texas at Austin The recently introduced General Data Protection Regulation (GDPR) is forcing ...admin - December 1, 2021 - 6:30 pm
-
A Tale of Two Abstractions: The Case for Object Space
Daniel Bittman, Peter Alvaro, Darrell D. E. Long, and Ethan L. Miller, UC Santa Cruz The increasing availability of byte-addressable non-volatile memory on the system bus provides an opportunity to dramatically simplify application interaction with persis ...admin - December 1, 2021 - 6:30 pm