Performance Analysis of Prefetching Thread for Linked Data Structure in CMPs

2009 ◽  
Author(s):  
Yan Huang ◽  
Zhimin Gu
Keyword(s):  
Data Structure ◽  
Performance Analysis ◽  
Linked Data ◽  
Prefetching Thread

scholarly journals Fast Key-Value Lookups with Node Tracker

2021 ◽  
Vol 18 (3) ◽  
pp. 1-26
Author(s):  
Mustafa Cavus ◽  
Mohammed Shatnawi ◽  
Resit Sendag ◽  
Augustus K. Uht
Keyword(s):  
Data Structure ◽  
Linked Data ◽  
Single Thread ◽  
Highly Effective ◽  
Execution Unit

Lookup operations for in-memory databases are heavily memory bound, because they often rely on pointer-chasing linked data structure traversals. They also have many branches that are hard-to-predict due to random key lookups. In this study, we show that although cache misses are the primary bottleneck for these applications, without a method for eliminating the branch mispredictions only a small fraction of the performance benefit is achieved through prefetching alone. We propose the Node Tracker (NT), a novel programmable prefetcher/pre-execution unit that is highly effective in exploiting inter key-lookup parallelism to improve single-thread performance. We extend NT with branch outcome streaming (BOS) to reduce branch mispredictions and show that this achieves an extra 3× speedup. Finally, we evaluate the NT as a pre-execution unit and demonstrate that we can further improve the performance in both single- and multi-threaded execution modes. Our results show that, on average, NT improves single-thread performance by 4.1× when used as a prefetcher; 11.9× as a prefetcher with BOS; 14.9× as a pre-execution unit and 18.8× as a pre-execution unit with BOS. Finally, with 24 cores of the latter version, we achieve a speedup of 203× and 11× over the single-core and 24-core baselines, respectively.

Effective Data Structure for the Multidimensional Orthogonal Bin Packing Problems

2014 ◽  
Vol 962-965 ◽  
pp. 2868-2871 ◽  
Author(s):  
Alexander V. Chekanin ◽  
Vladislav A. Chekanin
Keyword(s):  
Data Structure ◽  
Linked Data ◽  
Bin Packing ◽  
High Efficiency ◽  
Optimization Problems ◽  
Packing Problem ◽  
Strong Sense ◽  
Np Hard ◽  
Packing Problems ◽  
Orthogonal Packing Problem

The actual in industry multidimensional orthogonal packing problem is considered in the article. Solution of a large number of different practical optimization problems, including resources saving problem, optimization problems in logistics, scheduling and planning comes down to the orthogonal packing problem which is NP-hard in strong sense. One of the indicators characterizing the efficiency of packing constructing algorithm is the efficiency of the used data structure. In the article a multilevel linked data structure that increases the speed of constructing of a packing is proposed. The carried out computational experiments show the high efficiency of the new data structure. Multilevel linked data structure is applicable for multidimensional orthogonal bin packing problems any kind.

scholarly journals Performance Anatomization of Computer Algorithms Based on Iterative and Recursive Methodologies

2020 ◽  
Vol 8 (5) ◽  
pp. 1272-1276
Keyword(s):  
Data Structure ◽  
Performance Analysis ◽  
Computational Efficiency ◽  
Time Complexity ◽  
Research Paper ◽  
Space Complexity ◽  
Computer Algorithms ◽  
Solution Quality ◽  
Long Time

An algorithm is a clear specification of a sequence of instructions which when followed, provides a solution to a given problem. Writing an algorithm, depends upon various parameters, which leads to strong algorithmic performance, in terms of its computational efficiency and solution quality. This research paper presents the different methodologies of writing algorithms of data structure and also provides their performance analysis with respect to time complexity and space complexity. As we know that, for the same problem, we will have different algorithms, written using different approaches. All approaches of algorithms are important and have been an area of focus for a long time but still the question remains the same “which to use when?”, which is the main reason to perform this research. This research provides a detailed study of how algorithms being written using different approaches work and then compares them on the basis of various parameters such as time complexity and space complexity, to reach the conclusion.

Multilevel Linked Data Structure for the Multidimensional Orthogonal Packing Problem

2014 ◽  
Vol 598 ◽  
pp. 387-391 ◽  
Author(s):  
Vladislav A. Chekanin ◽  
Alexander V. Chekanin
Keyword(s):  
Data Structure ◽  
Linked Data ◽  
Bin Packing ◽  
Packing Problem ◽  
Packing Problems ◽  
Bin Packing Problem ◽  
Orthogonal Packing Problem

The actual NP-completed orthogonal bin packing problem is considered in the article. In practice a solution of a large number of different practical problems, including problems in logistics and scheduling comes down to the bin packing problem. A decision of an any packing problem is represented as a placement string which contains a sequence of objects selected to pack. The article proposes a new multilevel linked data structure that improves the effectiveness of decoding of the placement string and as a consequence, increases the speed of packing generation. The new data structure is applicable for all multidimensional orthogonal bin packing problems.

Sign in / Sign up

Export Citation Format

Share Document