Dissertation > Excellent graduate degree dissertation topics show

Research on Algorithm of Parallel Copying Garbage Collection Based on Lisp 2 for Multicore System

Author: WuChangMao
Tutor: ZhangCongPin
School: Henan Normal
Course: Computer Software and Theory
Keywords: multicore garbage collection Lisp 2 copying algorithm parallelization
CLC: TP332
Type: Master's thesis
Year: 2011
Downloads: 15
Quote: 0
Read: Download Dissertation

Abstract


As object-oriented languages and programming approach are widely used at present, garbage collection is drawing people’s more attention. Garbage collection which is also called Dynamic Memory Management is a mechanism of deallocating unreachable objects by Runtime System. The prevailing object-oriented languages such as C# and Java all support garbage collection. With garbage collection, programmers only pay attention to allocate memory for objects without taking destroying objects into consideration. The unreachable objects are deallocated automatically by garbage collector without the intervention of programmers, which considerably decreases the burden of programmers. Garbage collection avoids memory leak and errors caused by incorrect memory operations, e.g. dangling reference, improves the robustness of software as well. The performance and efficiency of garbage collector, however, directly affect the program’s executive efficiency and user’s experience. Therefore, improving the efficiency of garbage collector is of great importance in application fields.At the present time, multicore CPU and multicore GPU increasingly prevail, consequently their parallel computing ability sharply rise, all these offer the parallelization of garbage collection a solid basisWith the study on garbage collection, we have mainly accomplished the following works in this paper:1. We have deeply researched on the algorithms of garbage collectionBy means of looking up a large number of papers and treatises about garbage collection, we do many researches on the garbage collection algorithms e.g. Reference-Count algorithm, Mark-Sweep Algorithm, Copying Algorithm as well Generational Algorithm, especially parallel garbage collection algorithm, and compare these methods in the aspects of disadvantages, advantages and applicable surroundings.2. We have presented a novel copying garbage collection algorithm (Lisp 2-Copying-GC-Algorithm) based on Lisp 2 through deallocating unreachable objects on FromSpace and copying live objects to ToSpace.3. By means of dividing heap into equal blocks and defining 6 statuses for blocks, we have presented the parallel algorithm of Lisp 2-Copying-GC-Algorithm, called Lisp 2-Parallel-Copying-GC.4. The implementation of Lisp 2-Parallel-Copying-GC in multicore CPU systemIn multicore CPU system, we have implemented the algorithm of Lisp 2-Parallel-Copying-GC using the OpenMP 3.0 specification. OpenMP is a parallel language for shared memory system and distributed shared memory system, and also is a programming interface for multithread programming. The experimental results show that in multicore CPU system the proposed algorithm is able to improve the garbage collection efficiency in a large scale.5. The implementation of Lisp 2-Parallel-Copying-GC in multicore GPU systemCUDA is NVIDIA’s parallel computing architecture. It enables dramatic increases in computing performance by harnessing the power of the GPU. For the first time we conduct an effective exploration in garbage collection field using GPU system. In multicore GPU system, we have implemented Lisp 2-Parallel-Copying-GC. From the results we can see that the garbage collection efficiency is much improved in multicore GPU system than that in multicore CPU system.From the works above, we realize that with the multicore framework (multicore CPU system and multicore GPU system) prevailing, if garbage collection could employ the strong computing ability in multicore system, the efficiency of garbage collector will be largely improved so that the user’s experience will be better.

Related Dissertations

  1. Research of the Important Technology of Flash File System Based on Large Capacity NAND Flash Memory,TP333
  2. A Cache Scheme for Improving Random Write Performance in SSD,TP333
  3. Research of Main Memory Database Parallel Query Processing on SMP,TP311.13
  4. Resarch of Task-level Data Processing Based on Multicore CPU and Test of Its Performance on Cluster Platform,TP274
  5. Tile data recording disk drive technology research organization and its implementation,TP333.35
  6. Dynamic analysis based on multi-core acceleration method,TP332
  7. High dynamic conditions SINS Moving Base Transfer Alignment Parallel Algorithms,V249.322
  8. TilePro64 multicore processors based H.264 HD video decoding software design,TN919.81
  9. Based on multi-core SMP cluster environment parallel ray tracing simulation of satellite imagery and Implementation,TP391.41
  10. Multi-core processor based on RSA Research and Implementation of Parallel Algorithms,TP309.7
  11. Efficient multi- GPU based acoustic wave simulator and its application,TP391.41
  12. Combination of multi-core based on the generalized Gaussian image classification method,TP391.41
  13. Multi-core parallel cluster environment GeoComputation execution time prediction technology research,TP338.6
  14. Front of the flow based on shared key technology of multi-core architecture,TP332
  15. Core temperature for the public perception of the Linux scheduler Research and Implementation,TP316.81
  16. Short-term Urban Traffic Forecasting Based on Multi-kernel SVM Model,U491.14
  17. EFI-based software model of heterogeneous multi-core systems and its security applications,TP316
  18. FCM combines multi-resolution image cone multicore parallel image segmentation algorithm,TP391.41
  19. Fault-Tolerant Routing Algorithms for 2D-MESH Based Network-on-Chip,TP302.8
  20. A FTL Algorithm Combined with Hot Data Identifier,TP333
  21. LDPC decoder, the realization of research,TN911.22

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Electronic digital computer (not a continuous role in computer ) > Arithmetic unit and the controller (CPU)
© 2012 www.DissertationTopic.Net  Mobile