Proceedings of the ACM SIGPLAN 1999 conference on Programming language design and implementation
Java bytecode compression for low-end embedded systems
ACM Transactions on Programming Languages and Systems (TOPLAS)
Bytecode compression via profiled grammar rewriting
Proceedings of the ACM SIGPLAN 2001 conference on Programming language design and implementation
Grammar-based compression of interpreted code
Communications of the ACM - Program compaction
Heap compression for memory-constrained Java environments
OOPSLA '03 Proceedings of the 18th annual ACM SIGPLAN conference on Object-oriented programing, systems, languages, and applications
More efficient network class loading through bundling
JVM'01 Proceedings of the 2001 Symposium on JavaTM Virtual Machine Research and Technology Symposium - Volume 1
Optimization technique for commercial mobile MMORPG
Edutainment'10 Proceedings of the Entertainment for education, and 5th international conference on E-learning and games
Hi-index | 0.00 |
Cellular phones used to provide voice-only services, offline computing, games, schedulers, notes, and calculators, etc. Current generation of multi-functioned handsets including cellular phones and PDAs support high speed multimedia data transfers, MP3 music and video, and give the user direct access to multi-information services such as share process, news, sports results and M-commerce by Wireless Java. Many of these applications require a large volume of data and thus higher non-volatile memory capacity. One solution is to integrate Multi-chip package (MCP) & First Cycle Random Access Memory (FCRAM), or Flexible Flash Memories into the small handsets. The increasing memory capacity, however, causes higher cost and difficulty due to a limit of memory size in the small handsets. Alternative solution is to reduce memory space required for executing applications. Runtime memory compression technique is one candidate for achieving the goal. We propose a Java computation mechanism with a de / compression module that reduces Java heap memory demand. We also introduce the use of a shared Java virtual machine (VM) for wireless Java. The runtime heap compression mechanism could achieve 48.54% of compression ratio. The heap memory, then, could be saved 51.46% and the garbage collection (GC) time could be eliminated. This memory saving leads low power consumption with the multi-memory bank organization. The space efficiency of the proposed computation mechanism is 1.23 to 2.13. The speed up in Java heap memory management part varies among benchmarks and is up to 5.20. The modified LRU page replacement policy and the shared VM make the use of de / compression technique further less affect the entire performance. The proposed computation mechanism with Java heap de / compression module makes the wireless Java web applications faster and will offer good user experience and low power consumption.