Filters








6,384 Hits in 6.9 sec

Dynamic Resource Allocation for Database Servers Running on Virtual Storage

Gokul Soundararajan, Daniel Lupei, Saeed Ghanbari, Adrian Daniel Popescu, Jin Chen, Cristiana Amza
2009 USENIX Conference on File and Storage Technologies  
We show that our multi-resource allocation approach improves application performance by up to factors of 2.9 and 2.4 compared to state-of-the-art singleresource controllers, and their ad-hoc combination  ...  We use a combination of on-line modeling and sampling to arrive at near-optimal configurations within minutes.  ...  Finally, we acknowledge the generous support of our sponsors: Natural Sciences and Engineering Research Council (NSERC), Ontario Centers of Excellence (OCE), Ontario Ministry of Research and Innovation  ... 
dblp:conf/fast/SoundararajanLGPCA09 fatcat:e4jtjgca25bjdbpbryojgjdilm

Self-tuning histograms

Ashraf Aboulnaga, Surajit Chaudhuri
1999 Proceedings of the 1999 ACM SIGMOD international conference on Management of data - SIGMOD '99  
Although similar in structure to traditional histograms, these histograms infer data distributions not by examining the data or a sample thereof, but by using feedback from the query execution engine about  ...  In this paper, we describe the techniques for initializing and refining self-tuning histograms.  ...  Acknowledgement We thank Vivek Narasayya for useful discussions and for his help with the experiments.  ... 
doi:10.1145/304182.304198 dblp:conf/sigmod/AboulnagaC99 fatcat:prdio3atnfgkna6fex6jp4wsyu

Self-tuning histograms

Ashraf Aboulnaga, Surajit Chaudhuri
1999 SIGMOD record  
Although similar in structure to traditional histograms, these histograms infer data distributions not by examining the data or a sample thereof, but by using feedback from the query execution engine about  ...  In this paper, we describe the techniques for initializing and refining self-tuning histograms.  ...  Acknowledgement We thank Vivek Narasayya for useful discussions and for his help with the experiments.  ... 
doi:10.1145/304181.304198 fatcat:ee7t35na2beqbkdlck572642vm

Investigating Autonomic Runtime Management Strategies for SAMR Applications

Sumir Chandra, Manish Parashar, Jingmei Yang, Yeliang Zhang, Salim Hariri
2005 International journal of parallel programming  
Dynamic structured adaptive mesh refinement (SAMR) techniques along with the emergence of the computational Grid offer the potential for realistic scientific and engineering simulations of complex physical  ...  KEY WORDS: GridARM autonomic runtime management framework; structured adaptive mesh refinement; application/system sensitive reactive and proactive partitioning.  ...  The current application state is characterized in terms of application-level metrics such as Conceptual model of the GridARM framework 2 Runtime improvement using system-sensitive partitioning 3 Workload  ... 
doi:10.1007/s10766-005-3589-z fatcat:hbmjzz5gwnb33dlpygxn3m5z4a

Self-Tuning Database Systems: A Decade of Progress

Surajit Chaudhuri, Vivek R. Narasayya
2007 Very Large Data Bases Conference  
This paper primarily focuses on the problem of automated physical database design. We also highlight other areas where research on self-tuning database technology has made significant progress.  ...  We thank the VLDB 10-year Best Paper Award Committee for selecting our paper for the award. Working with members of the AutoAdmin research team has been a great experience for both of us.  ...  The shipping of the Index Tuning Wizard and subsequently the Database Engine Tuning Advisor in Microsoft SQL Server was made possible due to the commitment and support of many people in the SQL Server  ... 
dblp:conf/vldb/ChaudhuriN07 fatcat:sclwmmhdijdhvgagprmgbycqcm

Pragma: an infrastructure for runtime management of grid applications

M. Parashar, S. Hariri
2002 Proceedings 16th International Parallel and Distributed Processing Symposium  
, predictive models for system behavior and application performance, and an agent based control network.  ...  The overarching motivation for this research is to enable the next generation of very largescale dynamically adaptive scientific and engineering simulations on widely distributed and highly heterogeneous  ...  system and application state, predictive models for system behavior and application performance, and an agent based control network.  ... 
doi:10.1109/ipdps.2002.1016579 dblp:conf/ipps/ParasharH02 fatcat:4ws7iextyzc4rfjyzvfiuq7hgi

Heterogeneous CPU-GPU Epsilon Grid Joins: Static and Dynamic Work Partitioning Strategies

Benoit Gallet, Michael Gowanlock
2020 Data Science and Engineering  
We also propose a performance model for each static partitioning strategy to perform the distribution of the work between the processors.  ...  We evaluate the performance of all three partitioning methods by considering the execution time and the load imbalance between the CPU and GPU as performance metrics.  ...  of candidate points to refine), and w est the workload of the query points determined by the model (and where w total est is the total estimated workload of HEGJoin).  ... 
doi:10.1007/s41019-020-00145-x fatcat:jikurrovynfgfdflyoskceghlu

Hybrid Runtime Management of Space-Time Heterogeneity for Parallel Structured Adaptive Applications

Xiaolin Li, Manish Parashar
2007 IEEE Transactions on Parallel and Distributed Systems  
Further, to address dynamic resource situations for long running applications, AHMP provides a hybrid partitioning strategy (HPS), which involves application-level pipelining, trading space for time when  ...  for characterizing the partitioning requirements of these regions, and a runtime system for selecting, configuring and applying the most appropriate partitioner to each region.  ...  INTRODUCTION Simulations of complex physical phenomena, modeled by systems of partial differential equations (PDE), are playing an increasingly important role in science and engineering.  ... 
doi:10.1109/tpds.2007.1038 fatcat:7jn4yt5ijzbzdccoso5blowyja

Efficient Design Exploration Framework of SW/HW Systems Based on Tightly-coupled Thread Model

Arif Ullah Khan, Tsuyoshi Isshiki, Dongju Li, Hiroaki Kunieda
2015 IPSJ Transactions on System LSI Design Methodology  
The proposed framework drastically reduces the exponential size design space into near-linear size by utilizing the accurate HW timing models as the indicator for system bottleneck and guiding the enumeration  ...  In this paper we propose a design space exploration framework for heterogeneous MPSoC based on tightly-coupled thread (TCT) parallel programing model which can handles system partition exploration and  ...  Acknowledgments This work is supported by VLSI Design and Education Center (VDEC), the University of Tokyo, Synopsys, Inc., and Y Explorations, Inc.  ... 
doi:10.2197/ipsjtsldm.8.38 fatcat:gkv5i4gszrbc7fqbbbxpsr7dyi

A yoke of oxen and a thousand chickens for heavy lifting graph processing

Abdullah Gharaibeh, Lauro Beltrão Costa, Elizeu Santos-Neto, Matei Ripeanu
2012 Proceedings of the 21st international conference on Parallel architectures and compilation techniques - PACT '12  
Synchronous Parallel (BSP) model that offers a convenient environment to simplify the implementation of graph algorithms on heterogeneous platforms; and, (iii) demonstrates TOTEM'S efficiency by implementing  ...  TOTEM achieves speedups close to the model's prediction, and applies a number of optimizations that enable linear speedups with respect to the share of the graph offloaded for processing to accelerators  ...  We evaluate the performance of the system to demonstrate the feasibility of graph partitioning and offloading to GPUs and validate and refine the performance model.  ... 
doi:10.1145/2370816.2370866 dblp:conf/IEEEpact/GharaibehCSR12 fatcat:oiym75qbwzgvfcembm4u76t4j4

Program mapping onto network processors by recursive bipartitioning and refining

Jia Yu, Jingnan Yao, Laxmi Bhuyan, Jun Yang
2007 Proceedings - Design Automation Conference  
In this paper, we introduce a multilevel balancing and refining algorithm for NP program mapping.  ...  Then the algorithm iteratively refines the solution by migrating tasks from the bottleneck stage to other stages.  ...  , which will be used in the local refinement for pipeline workload balancing.  ... 
doi:10.1145/1278480.1278681 dblp:conf/dac/YuYBY07 fatcat:f2boqepclbauho5twbui4ymjlq

Program Mapping onto Network Processors by Recursive Bipartitioning and Refining

Jia Yu, Jingnan Yao, Laxmi Bhuyan, Jun Yang
2007 Proceedings - Design Automation Conference  
In this paper, we introduce a multilevel balancing and refining algorithm for NP program mapping.  ...  Then the algorithm iteratively refines the solution by migrating tasks from the bottleneck stage to other stages.  ...  , which will be used in the local refinement for pipeline workload balancing.  ... 
doi:10.1109/dac.2007.375275 fatcat:dcod5jzhkfffnn3eg7zcpdtf4u

System-level power estimation and optimization

Luca Benini, Robin Hodgson, Polly Siegel
1998 Proceedings of the 1998 international symposium on Low power electronics and design - ISLPED '98  
The model consists of a set of resources, an environmental workload specification, and a power management policy, which serves as the heart of the system model.  ...  The optimization algorithm can be used in a tight loop with the estimation engine to derive new power-management policy algorithms for a given system-level description.  ...  For this system, the workload can be modeled by collecting actual traces of packet arrival times on a real LAN, and by using these traces to drive the model.  ... 
doi:10.1145/280756.280881 dblp:conf/islped/BeniniHS98 fatcat:o3vzpztyxfhezhgjmbdw6hjkwi

Multidimensional Adaptive & Progressive Indexes

Matheus Agio Nerone, Pedro Holanda, Eduardo C. de Almeida, Stefan Manegold
2021 2021 IEEE 37th International Conference on Data Engineering (ICDE)  
Our extensive experimental evaluation using both synthetic and real-life data sets and workloads shows that (a) the Adaptive KD-Tree reduces total workload time by up to a factor 2 compared to the state-of-the-art  ...  three orders of magnitude lower than the state-of-the-art.  ...  Cost Model. The cost model considers the query and the state of the index in a way that is not affected by different data distributions, workload patterns, or query selectivities.  ... 
doi:10.1109/icde51399.2021.00060 fatcat:6luyf5l7dbhxjiofbfnbpqt4tu

Stripe: Tensor Compilation via the Nested Polyhedral Model [article]

Tim Zerrell, Jeremy Bruestle
2019 arXiv   pre-print
Unfortunately, this approach requires excessive engineering effort to scale or maintain with any degree of state-of-the-art performance.  ...  To mitigate, engineers develop hand-tuned kernels for each ML library update and hardware upgrade.  ...  Acknowledgments We would like to gratefully acknowledge the contributions and feedback of a number of people without whom this paper would not have been possible.  ... 
arXiv:1903.06498v1 fatcat:feqgth6yxvdhxb652jv5vwwlea
« Previous Showing results 1 — 15 out of 6,384 results