Google Android 作業系統被大量使用在移動裝置中，其採用Dalvik虛擬機器做為執行程式的主要平台。Dalvik虛擬機器被視為一種Java虛擬機器，但需先將Java程式轉換成Google 自訂的DEX (Dalvik Executable)中間碼格式才能執行。他是一個register-based的虛擬機器，其採用trace-based的即時編譯器來加速程式執行中頻繁執行的部分。
瀏覽器的JavaScript引擎通常採用一種由method-based以及trace-based組合而成的即時編譯技術，但其較推崇trace-based 的即時編譯器，主要原因是在於trace-based即時編譯器可優化頻繁執行的指令，特別是優化迴圈的能力；此外，即時編譯器可知道JavaScript執行時期的變數型態，並在假設變數是某種型態時產生不同的原生版本使執行速度更為快速。但是，Android的Dalvik虛擬機器對於DEX指令的執行並沒有保存其變數的型態，而是利用運算元來傳達運算子的可能型態；同時採用相當簡易的trace-based即時編譯器，其中並不包含上述所提到的優化。
我們為Dalvik虛擬機器設計了一method-based即時編譯器，與以底層虛擬機(Low Level Virtual Machine)實作的提前式編譯器相比有更好的效能，且利用Android市集上公認的測試基準與Dalvik虛擬機器中的trace-based即時編譯器比較之下，我們的method-based即時編譯器擁有更好的效能表現。此外，藉由分享剖析以及編譯相關的資訊，將method-based和trace-based兩種技術適當混合，各取其優劣可以大幅度的提升效能；為了使快速的trace-based即時編譯器能和高度優化的method-based即時編譯器結合，我們一旦發現有機會達到提升效能的目的時就會利用剖析器選擇method給method-based即時編譯器編譯。

The Dalvik virtual machine is the main application platform running on Google's Android operating system for mobile devices and tablets. It is a Java Virtual Machine implementation that executes Java programs that have been translated to Google's DEX intermediate representation. This register-based virtual machine runs a basic trace-based Just-In-Time compiler to speed up the execution of programs that contain hot traces.

Web browser JavaScript engines usually run a combination of both method and trace-based JIT types. Such JavaScript engines advocate to the power of a trace-based JIT compiler because of its ability to optimize frequently executed instructions, especially inside loops. Because a JIT is able to observer the runtime type of JavaScript variables, it can have different native versions of loops that work faster under the assumption that a variable is of a certain type. Android's Dalvik virtual machine runs dex instructions, which does not preserve variable types but uses operations that reveal what the operators' types are likely to be, uses a very simple trace-based JIT but does not implement any of the above optimizations.

The method-based JIT compiler presented here delivers performance improvement comparable to that of an Ahead-Of-Time compiler based on the Low Level Virtual Machine framework. The method-based JIT was compared against Dalvik's own trace-based JIT using common benchmarks available in the Android Market. Results show that our method-based JIT is better than a basic trace-based, and that, by sharing profiling and compilation information among each other, a smart combination of both JIT techniques can achieve a great performance gain. Using a fine-tunes profiler that selects methods to be method-based JIT compiled in order to combine fast trace-based JIT compilation with highly optimized method-based JIT compilation when there is an opportunity to do so.

[1] Google android - an open handset alliance project. http://code.
google.com/android/, 2008.
[2] Arm: Floating point architecture. http://www.
arm.com/products/processors/technologies/
vector-floating-point.php, 2012.
[3] V8 javascript engine. http://code.google.com/p/v8/, 2011.
[4] S. Asghar, M. Hafeez, U. A. Malik, A. ur Rehman, and N. Riaz.
A-jump, architecture for java universal message passing. In
Proceedings of the 8th International Conference on Frontiers of
Information Technology, FIT ’10, pages 34:1–34:6, New York,
NY, USA, 2010. ACM. ISBN 978-1-4503-0342-2. doi: http:
//doi.acm.org/10.1145/1943628.1943662. URL http://doi.acm.
org/10.1145/1943628.1943662.
[5] C. S. D. at the University of Illinois at Urbana-Champaign. clang: a
c language family frontend for llvm. http://clang.llvm.org/,
2012.
[6] C. S. D. at the University of Illinois at Urbana-Champaign. The
llvm target-independent code generator. http://llvm.org/docs/
CodeGenerator.html#mc, 16 January, 2012.
[7] V. Bala, E. Duesterwald, and S. Banerjia. Dynamo: a transparent
dynamic optimization system. SIGPLAN Not., 35(5):1–12, May
2000. ISSN 0362-1340. doi: 10.1145/358438.349303. URL
http://doi.acm.org/10.1145/358438.349303.
[8] G. Benson. Zero and shark: a zero-assembly port of openjdk.
http://today.java.net/pub/a/today/2009/05/21/
zero-and-shark-openjdk-port.html, 2009.
[9] B. Cheng and B. Buzbee. A jit compiler for android’s
dalvik vm. http://dl.google.com/googleio/2010/
android-jit-compiler-androids-dalvik-vm.pdf, 2010.
[10] I. Free Software Foundation. Gcc, the gnu c compiler. http:
//gcc.gnu.org/, 05 March, 2012.
[11] A. Gal, B. Eich, M. Shaver, D. Anderson, D. Mandelin, M. R.
Haghighat, B. Kaplan, G. Hoare, B. Zbarsky, J. Orendorff, J. Ruderman,
E. W. Smith, R. Reitmaier, M. Bebenita, M. Chang, and
M. Franz. Trace-based just-in-time type specialization for dynamic
languages. SIGPLAN Not., 44:465–478, June 2009. ISSN
0362-1340. doi: http://doi.acm.org/10.1145/1543135.1542528. URL
http://doi.acm.org/10.1145/1543135.1542528.
[12] Google. Android developer tools. http://developer.android.
com/guide/developing/tools/index.html, 2007.
[13] Google. Galaxy nexus. http://www.google.com/nexus/, 2011.
[14] S.-y. Guo and J. Palsberg. The essence of compiling with traces.
SIGPLAN Not., 46(1):563–574, Jan. 2011. ISSN 0362-1340. doi:
10.1145/1925844.1926450. URL http://doi.acm.org/10.1145/
1925844.1926450.
[15] T. IEEE and T. O. Group. Posix.1-2008: The open group base specifications
issue 7. http://pubs.opengroup.org/onlinepubs/
9699919799/, 2001 - 2008.
[16] H. Inoue, H. Hayashizaki, P. Wu, and T. Nakatani. A trace-based
java jit compiler retrofitted from a method-based compiler. In Code
Generation and Optimization (CGO), 2011 9th Annual IEEE/ACM
International Symposium on, pages 246 –256, april 2011. doi:
10.1109/CGO.2011.5764692.
[17] JesusFreke. smali/baksmali: An assembler for android’s dex format.
http://code.google.com/p/smali/, 2009.
[18] D. Kurzyniec and V. Sunderam. Efficient cooperation between java
and native codes jni performance benchmark. In In The 2001
International Conference on Parallel and Distributed Processing
Techniques and Applications, 2001.
[19] C. Lattner, M. Brukman, and B. Gaeke. Jello: a retargetable just-intime
compiler for llvm bytecode, 2002.
[20] C. A. Lattner. Llvm: An infrastructure for multi-stage optimization.
Technical report, 2002.
[21] S.-W. Lee and S.-M. Moon. Selective just-in-time compilation
for client-side mobile javascript engine. In Proceedings of the
14th international conference on Compilers, architectures and
synthesis for embedded systems, CASES ’11, pages 5–14, New
York, NY, USA, 2011. ACM. ISBN 978-1-4503-0713-0. doi:
10.1145/2038698.2038703. URL http://doi.acm.org/10.1145/
2038698.2038703.
[22] J. McCarthy. Recursive functions of symbolic expressions and their
computation by machine, part i. Commun. ACM, 3(4):184–195,
Apr. 1960. ISSN 0001-0782. doi: 10.1145/367177.367199. URL
http://doi.acm.org/10.1145/367177.367199.
[23] Mozilla. Jaegermonkey. https://wiki.mozilla.org/
JaegerMonkey, 2010.
[24] S. Oaks and H. Wong. Java Threads. O’Reilly Media, Inc., 2004.
ISBN 0596007825.
[25] Oracle. Java se specifications. http://docs.oracle.com/
javase/specs/, 2012.
[26] Oracle. Openjdk’s hotspot vm. http://openjdk.java.net/
groups/hotspot/, 2012.
[27] V. K. Polychronis. Benchmarkpi - android benchmarking tool.
http://androidbenchmark.com/, 2009.
[28] T. A. O. S. Project. Dalvik optimization and verification with dexopt.
http://www.netmite.com/android/mydroid/dalvik/docs/
dexopt.html, 2008.
[29] G. Tim Bray. Renderscript. http://android-developers.
blogspot.com/2011/02/introducing-renderscript.html,
09 February, 2011.
[30] C.-S. Wang, G. Perez, Y.-C. Chung, W.-C. Hsu, W.-K. Shih, and
H.-R. Hsu. A method-based ahead-of-time compiler for android
applications. In Proceedings of the 14th international conference on
Compilers, architectures and synthesis for embedded systems, CASES
’11, pages 15–24, New York, NY, USA, 2011. ACM. ISBN 978-
1-4503-0713-0. doi: http://doi.acm.org/10.1145/2038698.2038704.
URL http://doi.acm.org/10.1145/2038698.2038704.