A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf
.
Filters
Towards Heterogeneous Multi-core Accelerators Exploiting Fine-grained Scheduling of Layer-Fused Deep Neural Networks
[article]
2022
arXiv
pre-print
This work, therefore, enables fine-grained depth-first scheduling of layer-fused DNNs onto multi-core architectures through an open-source modeling framework called Stream. ...
Using Stream in further explorations, we demonstrate that high-level architectural decisions greatly impact hardware efficiency under the fine-grained scheduling paradigm, reducing the energy-delay product ...
Performing such fine-grained scheduling on a multi-core HW architecture comes with two major challenges: 1.) ...
arXiv:2212.10612v1
fatcat:rlriljcinneg7p7f63xiedbpxq
Trill
2014
Proceedings of the VLDB Endowment
streaming and relational queries with early results, across the latency spectrum from real-time to offline; (2) Fabric and Language Integration: Trill is architected as a high-level language library that ...
Trill fulfills a combination of three requirements for a query processor to serve the diverse big data analytics space: (1) Query Model: Trill is based on a tempo-relational model that enables it to handle ...
flow through the query plan with no fine-grained work. ...
doi:10.14778/2735496.2735503
fatcat:6cwt4vpxufgpzl2biaevkjoqay
A Survey of Multi-Tenant Deep Learning Inference on GPU
[article]
2022
arXiv
pre-print
This survey aims to summarize and categorize the emerging challenges and optimization opportunities for multi-tenant DL inference on GPU. ...
By overviewing the entire optimization stack, summarizing the multi-tenant computing innovations, and elaborating the recent technological advances, we hope that this survey could shed light on new optimization ...
Process Launch When Idle valuable multi-tenant GPU scheduling mechanisms, based on which many recent works have started to explore the fine- grained DL operator-level scheduling techniques [9, 43]. [24 ...
arXiv:2203.09040v3
fatcat:utvpoyvvajfhfghgpf45nxnbne
Text-Based Person Search with Limited Data
[article]
2021
arXiv
pre-print
Solving such a fine-grained cross-modal retrieval task is challenging, which is further hampered by the lack of large-scale datasets. ...
Text-based person search (TBPS) aims at retrieving a target person from an image gallery with a descriptive text query. ...
It is clear that appending a GRU to the pre-trained textual encoder directly allows the textual stream to learn fine-grained textual information on the basis of the previously learned coarse-grained one ...
arXiv:2110.10807v1
fatcat:l6pneuh6izcrbomek57mebh2ce
MeDSLIP: Medical Dual-Stream Language-Image Pre-training for Fine-grained Alignment
[article]
2024
arXiv
pre-print
Specifically, MeDSLIP establishes vision-language fine-grained alignments via disentangling visual and textual representations into anatomy-relevant and pathology-relevant streams. ...
MeDSLIP is evaluated under zero-shot and supervised fine-tuning settings on three public datasets: NIH CXR14, RSNA Pneumonia, and SIIM-ACR Pneumothorax. ...
The query networks in the framework are multi-layer, multi-head cross-attention modules [18, 23, 25] to align representations with query embeddings. ...
arXiv:2403.10635v1
fatcat:nfvnhuwaurbe7lmwkqgmpjvb2q
ICDE conference 2015 detailed author index
2015
2015 IEEE 31st International Conference on Data Engineering
Repair Approach
387
Diversified Top-K Clique Search
1107
AP-Tree: Efficiently Support Continuous Spatial-Keyword Queries over Stream
Ling, Tok Wang
1408
PandaSearch: A Fine-Grained Academic ...
Subscriptions
1504
Elaps: An Efficient Location-Aware Pub/Sub System
Li, Jia
1408
PandaSearch: A Fine-Grained Academic Search Engine for Research Documents
Li, Lin
1436
MARS: A Multi-Aspect ...
doi:10.1109/icde.2015.7113260
fatcat:ep7pomkm55f45j33tkpoc5asim
Cloud architecture for plant phenotyping research
2020
Concurrency and Computation
; (2) the speed layer, which processes stream data, produces views and deploys them on serving layer; (3) the serving layer receives client queries and produces serving up views merging batch and real-time ...
Digital phenotyping is an emergent science mainly based on imagery techniques. The tremendous amount of data generated needs important cloud computing for their processing. ...
The authors would also like to thank Mr Adriano Guttadauria for his technical support and for setting up all the electronic and informatic systems necessary for carrying out this cloud architecture. ...
doi:10.1002/cpe.5661
fatcat:d56lx4zndndbdmvlnajnoikdpa
Towards a unified storage and ingestion architecture for stream processing
2017
2017 IEEE International Conference on Big Data (Big Data)
Current streaming-oriented runtimes and middlewares are not flexible enough to deal with this trend, as they address ingestion (collection and pre-processing of data streams) and persistent storage (archival ...
In this position paper, we argue for a unified ingestion and storage architecture for streaming data that addresses the aforementioned challenge. ...
Moreover, we envision to build our proposal on top of robust low-level key-value stores that are capable of efficient fine-grain access (e.g., put/get, multiwrite/multi-read) to the stream records. ...
doi:10.1109/bigdata.2017.8258196
dblp:conf/bigdataconf/MarcuCAPTBN17
fatcat:w5a2fruranfflbwchhfuerqrpm
Data-Streaming and Concurrent Data-Object Co-design: Overview and Algorithmic Challenges
[chapter]
2015
Lecture Notes in Computer Science
Processing big volumes of data generated on-line, implies needs to carry out computations on-the-fly, in the streams of data. ...
suitable for a set of functions and identify new key challenges to improve data-stream processing through co-design with fine-grain efficient synchronization combined with the data exchange. ...
In this context, fine-grained synchronization mechanisms should be defined for threads M in and M proc intercommunication. Merge Physical Streams Concurrently. ...
doi:10.1007/978-3-319-24024-4_15
fatcat:mz75ythl4bfwvlyj3erhruw7fe
2021 Index IEEE Transactions on Parallel and Distributed Systems Vol. 32
2022
IEEE Transactions on Parallel and Distributed Systems
., +, TPDS Oct. 2021 2415-2433 Fine-Grained Multi-Query Stream Processing on Integrated Architectures. ...
Fan, W., +, TPDS Nov. 2021 2793-2808 Fine-Grained Multi-Query Stream Processing on Integrated Architectures. ...
Graph coloring Feluca: A Two-Stage Graph Coloring Algorithm With Color-Centric Paradigm on GPU. Zheng, Z., +, ...
doi:10.1109/tpds.2021.3107121
fatcat:e7bh2xssazdrjcpgn64mqh4hb4
Multi-perspective Attention Network for Fast Temporal Moment Localization
2021
IEEE Access
However, these fine-grained techniques require a high computational overhead, making them impractical. ...
Similar to the heavyweight fine-grained TML methods, the proposed network obtains the accurate complex relationship while being lightweight like coarse-grained TML methods. ...
Our MPAN performs the TML task in a coarse-grained manner but can learn the deeper local-global relationship between two inputs using our multi-perspective attention mechanism, as fine-grained methods ...
doi:10.1109/access.2021.3106698
fatcat:5xtxboezxrahveif6k2cwdtbee
Combined CNN Transformer Encoder for Enhanced Fine-grained Human Action Recognition
[article]
2022
arXiv
pre-print
We achieve new state-of-the-art performance on the FineGym benchmark dataset for both proposed architectures. ...
two frameworks that combine CNN vision backbone and Transformer Encoder to enhance fine-grained action recognition: 1) a vision-based encoder to learn latent temporal semantics, and 2) a multi-modal video-text ...
We also compare with a multi-task learning network (MTN) [15] that jointly predicts the coarse-to-fine grained action labels and TQN [22] with video-text query-response system. ...
arXiv:2208.01897v1
fatcat:omfjpa3qfbd6ppadbhdl54pzjq
Hyperpubsub: a decentralized, permissioned, publish/subscribe service using blockchains
2017
Proceedings of the 18th ACM/IFIP/USENIX Middleware Conference on Posters and Demos - Middleware '17
To facilitate the adoption of blockchain technologies across various business verticals, we argue that middleware plays an integral role in accelerating the development of automated business processes ...
Our system, called HyperPubSub, provides decentralized publish/subscribe messaging for a multi-federated, permissioned, environment. ...
We thank Richard Hull (IBM) for helpful discussions on Hyperledger. ...
doi:10.1145/3155016.3155018
dblp:conf/middleware/Zupan0J17
fatcat:4gckqlimafhcpguvqssw2iklua
Discriminative Feature Representation with Spatio-temporal Cues for Vehicle Re-identification
[article]
2020
arXiv
pre-print
Based on this multi-modal information, the proposed DFR-ST constructs an appearance model for a multi-grained visual representation by a two-stream architecture and a spatio-temporal metric to provide ...
Vehicle re-identification (re-ID) aims to discover and match the target vehicles from a gallery image set taken by different cameras on a wide range of road networks. ...
The appearance module is composed of a coarse-grained feature stream and a fine-grained feature stream. ...
arXiv:2011.06852v1
fatcat:iqlq53dbxbb6je66a47nsvzgha
The Database Machine: Old Story, New Slant?
2003
Conference on Innovative Data Systems Research
Therefore we suggest an alternative where database systems shake off their relatively static monolithic structure and become open sets of fine-grained components providing a collection of key information ...
Current DBMS architectures inherently do not provide this level of customisation or adaptability. ...
Essentially we believe that to have a truly fine-grained DBMS we must have fine-grained components throughout (i.e. including the OS). ...
dblp:conf/cidr/McCann03
fatcat:26j3muy5ifc6vh42qsmdh5jxbi
« Previous
Showing results 1 — 15 out of 10,639 results