Exploiting information access patterns for context-based retrieval

Author(s):  
Travis Bauer ◽  
David B. Leake
2009 ◽  
Vol 29 (5) ◽  
pp. 1401-1404
Author(s):  
Ming SUN ◽  
Bo CHEN ◽  
Ming-tian ZHOU
Keyword(s):  

Author(s):  
Benjamin Fish ◽  
Ashkan Bashardoust ◽  
Danah Boyd ◽  
Sorelle Friedler ◽  
Carlos Scheidegger ◽  
...  

Agronomy ◽  
2021 ◽  
Vol 11 (4) ◽  
pp. 777
Author(s):  
Erythrina Erythrina ◽  
Arif Anshori ◽  
Charles Y. Bora ◽  
Dina O. Dewi ◽  
Martina S. Lestari ◽  
...  

In this study, we aimed to improve rice farmers’ productivity and profitability in rainfed lowlands through appropriate crop and nutrient management by closing the rice yield gap during the dry season in the rainfed lowlands of Indonesia. The Integrated Crop Management package, involving recommended practices (RP) from the Indonesian Agency for Agricultural Research and Development (IAARD), were compared to the farmers’ current practices at ten farmer-participatory demonstration plots across ten provinces of Indonesia in 2019. The farmers’ practices (FP) usually involved using old varieties in their remaining land and following their existing fertilizer management methods. The results indicate that improved varieties and nutrient best management practices in rice production, along with water reservoir infrastructure and information access, contribute to increasing the productivity and profitability of rice farming. The mean rice yield increased significantly with RP compared with FP by 1.9 t ha–1 (ranges between 1.476 to 2.344 t ha–1), and net returns increased, after deducting the cost of fertilizers and machinery used for irrigation supplements, by USD 656 ha–1 (ranges between USD 266.1 to 867.9 ha–1) per crop cycle. This represents an exploitable yield gap of 37%. Disaggregated by the wet climate of western Indonesia and eastern Indonesia’s dry climate, the RP increased rice productivity by 1.8 and 2.0 t ha–1, with an additional net return gain per cycle of USD 600 and 712 ha–1, respectively. These results suggest that there is considerable potential to increase the rice production output from lowland rainfed rice systems by increasing cropping intensity and productivity. Here, we lay out the potential for site-specific variety and nutrient management with appropriate crop and supplemental irrigation as an ICM package, reducing the yield gap and increasing farmers’ yield and income during the dry season in Indonesia’s rainfed-prone areas.


2021 ◽  
Vol 31 (2) ◽  
pp. 1-28
Author(s):  
Gopinath Chennupati ◽  
Nandakishore Santhi ◽  
Phill Romero ◽  
Stephan Eidenbenz

Hardware architectures become increasingly complex as the compute capabilities grow to exascale. We present the Analytical Memory Model with Pipelines (AMMP) of the Performance Prediction Toolkit (PPT). PPT-AMMP takes high-level source code and hardware architecture parameters as input and predicts runtime of that code on the target hardware platform, which is defined in the input parameters. PPT-AMMP transforms the code to an (architecture-independent) intermediate representation, then (i) analyzes the basic block structure of the code, (ii) processes architecture-independent virtual memory access patterns that it uses to build memory reuse distance distribution models for each basic block, and (iii) runs detailed basic-block level simulations to determine hardware pipeline usage. PPT-AMMP uses machine learning and regression techniques to build the prediction models based on small instances of the input code, then integrates into a higher-order discrete-event simulation model of PPT running on Simian PDES engine. We validate PPT-AMMP on four standard computational physics benchmarks and present a use case of hardware parameter sensitivity analysis to identify bottleneck hardware resources on different code inputs. We further extend PPT-AMMP to predict the performance of a scientific application code, namely, the radiation transport mini-app SNAP. To this end, we analyze multi-variate regression models that accurately predict the reuse profiles and the basic block counts. We validate predicted SNAP runtimes against actual measured times.


Sign in / Sign up

Export Citation Format

Share Document