Data size optimizations for java programs

Ananian, C. Scott; Rinard, Martin

doi:10.1145/780732.780741

Cited by 32 publications

(16 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Type analysis and coloring are also implemented in FLEX [AR03], which includes various optimizations for Java programs, and FijiVM, an ahead-of-time compiler for Java which generates C code and uses 0-CFA rather than RTA [PZV09]. However, coloring is only used for interface methods.…”

Section: Embedded Runtime Systemsmentioning

confidence: 99%

Efficient compilation of .NET programs for embedded systems

Sallenave¹,

Ducournau

2010

Proceedings of the Workshop on the Implementation, Compilation, Optimization of Object-Oriented Languages, Programs and Systems

View full text Add to dashboard Cite

Compiling under the closed-world assumption (CWA) has been shown to be an appropriate way for implementing object-oriented languages such as Java on low-end embedded systems. In this paper, we explore the implications of using whole program optimizations such as Rapid Type Analysis (RTA) and coloring on programs targeting the .Net infrastructure. We extended RTA so that it takes into account .Net specific features such as (i) array covariance, a language feature also supported in Java, (ii) generics, whose specifications in .Net impacts type analysis and (iii) delegates, which encapsulate methods within objects. We also use an intraprocedural control flow analysis in addition to RTA . We evaluated the optimizations that we implemented on programs written in C#. Preliminary results show a noticeable reduction of the code size, class hierarchy and polymorphism of the programs we optimize. Array covariance is safe in almost all cases, and some delegate calls can be implemented as direct calls.

show abstract

Section: Embedded Runtime Systemsmentioning

confidence: 99%

Efficient compilation of .NET programs for embedded systems

Sallenave¹,

Ducournau

2010

Proceedings of the Workshop on the Implementation, Compilation, Optimization of Object-Oriented Languages, Programs and Systems

View full text Add to dashboard Cite

show abstract

“…The source of nonlinearity is the repeated posting of field members from a super-class to its instantiated subtypes (4), which happens at most once per field per subtype, which in the worst case is quadratic. For simple hierarchies, the algorithm runs in expected linear time.…”

Section: Algorithm Complexitymentioning

confidence: 99%

“…If it is the same value across all object instances, the compiler can inline it as a constant where reads occur. If it is constant by subclass [4], the compiler can move it to the meta-object, and otherwise, the compiler could introduce a constant table stored in ROM that is indexed by the object handle obtained with reference compression.…”

Section: Rom-izationmentioning

confidence: 99%

“…Ananian and Rinard [4] present a suite of both static and dynamic techniques to reduce data size for Java programs. They propose field reduction, a whole-program static analysis that bounds the ranges of values that primitive fields may contain over any execution in order to reduce their size, unread and constant field elimination, that removes unused and constant fields, static specialization, which eliminates fields that are constant by subclass, externalization, which removes frequently constant fields and puts them in a hash table, and class pointer compression, which is essentially a single compression table for object headers.…”

Section: Object Compressionmentioning

confidence: 99%

“…For example, a points-to analysis [30] with live objects would likely give more precise results than reachable members analysis and might allow objects of the same type to be laid out differently depending on which parts of the program manipulate them. There are a number of other reference compression and object layout techniques [4][10] [14] that could be applied with various tradeoffs between size and performance; further study is required to determine what works the best in practice for this domain. Such tradeoffs are likely to be fruitful research topics in the future if the initialization time model becomes more widely employed.…”

Section: Future Workmentioning

confidence: 99%

See 2 more Smart Citations

Virgil

Titzer¹

2006

SIGPLAN Not.

View full text Add to dashboard Cite

Embedded microcontrollers are becoming increasingly prolific, serving as the primary or auxiliary processor in products and research systems from microwaves to sensor networks. Microcontrollers represent perhaps the most severely resource-constrained embedded processors, often with as little as a few bytes of memory and a few kilobytes of code space. Language and compiler technology has so far been unable to bring the benefits of modern object-oriented languages to such processors. In this paper, I will present the design and implementation of Virgil, a lightweight objectoriented language designed with careful consideration for resource-limited domains. Virgil explicitly separates initialization time from runtime, allowing an application to build complex data structures during compilation and then run directly on the bare hardware without a virtual machine or any language runtime. This separation allows the entire program heap to be available at compile time and enables three new data-sensitive optimizations: reachable members analysis, reference compression, and ROM-ization. Experimental results demonstrate that Virgil is well suited for writing microcontroller programs, with five demonstrative applications fitting in less than 256 bytes of RAM with fewer than 50 bytes of metadata. Further results show that the optimizations presented in this paper reduced code size between 20% and 80% and RAM size by as much as 75%.

show abstract

Coloring, a versatile technique for implementing object‐oriented languages

Ducournau

2010

Softw Pract Exp

View full text Add to dashboard Cite

Late binding and subtyping create run-time overhead for object-oriented languages. Dynamic typing and multiple inheritance create even more overhead. Static typing and single inheritance lead to two major invariants-of reference and position-that make the implementation as efficient as possible. Coloring is a technique that preserves these invariants for dynamic typing or multiple inheritance at minimal spatial cost.Coloring has been introduced and applied, more or less independently, to method invocation-under the name of selector coloring-to subtype tests under the name of pack encoding and to attribute access. This paper reviews a number of uses of coloring for optimizing object-oriented programming and generalizes them. It specifies several variations of coloring, such as bidirectional coloring and n-dimensional coloring. Coloring is NP-hard, so compilers that use it depend on heuristics. The paper describes some experimental results which indicate that bidirectional coloring gives the best results.

show abstract

Data size optimizations for java programs

Cited by 32 publications

References 29 publications

Efficient compilation of .NET programs for embedded systems

Efficient compilation of .NET programs for embedded systems

Virgil

Coloring, a versatile technique for implementing object‐oriented languages

Contact Info

Product

Resources

About