Dynamic Translation Research Articles

Online optimizations driven by hardware performance monitoring

Hardware performance monitors provide detailed direct feedback about application behavior and are an additional source of infor-mation that a compiler may use for optimization. A JIT compiler is in a good position to make use of such information because it is running on the same platform as the user applications. As hardware platforms become more and more complex, it becomes more and more difficult to model their behavior. Profile information that captures general program properties (like execution frequency of methods or basic blocks) may be useful, but does not capture sufficient information about the execution platform. Machine-level performance data obtained from a hardware performance monitor can not only direct the compiler to those parts of the program that deserve its attention but also determine if an optimization step actually improved the performance of the application. This paper presents an infrastructure based on a dynamic compiler+runtime environment for Java that incorporates machine-level information as an additional kind of feedback for the compiler and runtime environment. The low-overhead monitoring system provides fine-grained performance data that can be tracked back to individual Java bytecode instructions. As an example, the paper presents results for object co-allocation in a generational garbage collector that optimizes spatial locality of objects on-line using measurements about cache misses. In the best case, the execution time is reduced by 14% and L1 cache misses by 28%.

HDTrans

Dynamic translation is a general purpose tool used for instrumenting programs at run time. Many current translators perform substantial rewriting during translation in an attempt to reduce execution time. When dynamic translation is used as a ubiquitous policy enforcement mechanism, the majority of program executions have no dominating inner loop that can be used to amortize the cost of translation. Even under more favorable usage assumptions, our measurements show that such optimizations offer no significant benefit in most cases. A simpler, more maintainable, adaptable, and smaller translator may be preferable to more complicated designs. In this paper, we present HDTrans, a light-weight IA-32 to IA-32 binary translation system that uses some simple and effective translation techniques in combination with established trace linearization and code caching optimizations. We also present an evaluation of translation overhead under non-ideal conditions, showing that conventional benchmarks do not provide a good prediction of translation overhead when used pervasively. A further contribution of this paper is an analysis of the effectiveness of post-compile static pre-translation techniques for overhead reduction. Our results indicate that static pre-translation is effective only when expensive instrumentation or optimization is performed, and that efficient reload of pre-translated code incurs a substantial execution-time penalty.

Changes in knee motion pattern after anterior cruciate ligament injury – A case report

After an anterior cruciate ligament injury, the contra-lateral non-injured leg has been found to adapt towards the injured leg. Accordingly, in order to study changes in knee motion pattern after an anterior cruciate ligament injury, the ideal is to compare the same leg prior to and after the injury. However, this is very seldom possible. The purpose of the present study was to describe changes in static and dynamic sagittal tibial translation, electromyographic activity and muscle torque relevant to an anterior cruciate ligament tear in one patient evaluated both before and after the injury. A male soccer player was examined 11 weeks before and eight weeks after an anterior cruciate ligament injury. Sagittal tibial translation was measured with the CA-4000 electrogoniometer; statically during Lachman's test, and dynamically during isokinetic muscle testing, one-legged squat and level walking. The electromyographic activity of mm. quadriceps and hamstrings, was registered simultaneously during the one-legged squat test. Static tibial translation was increased by approximately 2 mm, while dynamic tibial translation was decreased by 0.4 mm at isokinetic testing, 0.9 mm at one-legged squat and 2.4 mm during level walking compared to before the injury. Muscle torque decreased 30% and 35% for the quadriceps and the hamstrings muscle, respectively. The electromyographic activity revealed similar activation levels in quadriceps and a doubled level of activation in hamstring compared to before the injury. In spite of an increase in static tibial translation eight weeks after an anterior cruciate ligament injury, the tibial translation decreased during activity, thus indicating that the patient could stiffen the knee in order to protect it against increased shear forces.

Sagittal Plane Knee Motion in the ACL-Deficient Knee During Body Weight Shift Exercises on Different Support Surfaces

Experimental design with group comparisons. To compare anterior tibial translation and muscle activity among different exercises for early weight-bearing and neuromuscular training in individuals with a unilateral anterior cruciate ligament (ACL) injury and in uninjured controls. The effects of exercise and activity on tibial translation must be taken into consideration during rehabilitation after ACL injury. Twelve patients with an ACL-deficient knee and 12 age- and gender-matched controls participated in the study. Sagittal tibial translation and muscle activity were registered during the Lachman test (static translation) and 4 body weight shift exercises (dynamic translation). A Student t test with Bonferroni correction and analysis of variance were used for the statistical analysis. Forward-backward body weight shift exercise resulted in smaller anterior tibial translation compared to body weight shift from side to side. Analysis of EMG activity could not explain this difference in anterior tibial translation. The amount of anterior tibial translation or EMG activity did not change when the exercises were performed on a trampoline compared to a firm surface. Forward-backward weight shifting may be preferable in initial rehabilitation after ACL injury compared to body weight shift from side to side.

UP-02.06: Urology in Islamic medicine

Islamic medicine represents an important period of development of the medical knowledge in world history. It could be one of the most comprehensive summations of the medical knowledge and experiences from prior civilizations. The time frame, which extended over eight centuries, and the interaction with all other civilizations encountered during expansion of Islam in the Arab countries, Persia, North Africa, India, China, and Europe, provide a variety of cultures interacting in a dynamic translation and transformation process.

Replay compilation

The performance of Java has been tremendously improved by the advance of Just-in-Time (JIT) compilation technologies. However, debugging such a dynamic compiler is much harder than a static compiler. Recompiling the problematic method to produce a diagnostic output does not necessarily work as expected, because the compilation of a method depends on runtime information at the time of compilation.In this paper, we propose a new approach, called replay JIT compilation , which can reproduce the same compilation remotely by using two compilers, the state-saving compiler and the replaying compiler . The state-saving compiler is used in a normal run, and, while compiling a method, records into a log all of the input for the compiler. The replaying compiler is then used in a debugging run with the system dump, to recompile a method with the options for diagnostic output. We reduced the overhead to save the input by using the system dump and by categorizing the input based on how its value changes. In our experiment, the increase of the compilation time for saving the input was only 1%, and the size of the additional memory needed for saving the input was only 10% of the compiler-generated code.

Method-specific dynamic compilation using logistic regression

Determining the best set of optimizations to apply to a program has been a long standing problem for compiler writers. To reduce the complexity of this task, existing approaches typically apply the same set of optimizations to all procedures within a program, without regard to their particular structure. This paper develops a new method-specific approach that automatically selects the best optimizations on a per method basis within a dynamic compiler. Our approach uses the machine learning technique of logistic regression to automatically derive a predictive model that determines which optimizations to apply based on the features of a method. This technique is implemented in the Jikes RVM Java JIT compiler. Using this approach we reduce the average total execution time of the SPECjvm98 benchmarks by 29%. When the same heuristic is applied to the DaCapo+ benchmark suite, we obtain an average 33% reduction over the default level O2 setting.

Tibial translation in exercises used early in rehabilitation after anterior cruciate ligament reconstruction: Exercises to achieve weight-bearing

Accelerated rehabilitation after ACL reconstruction involves early weight-bearing. Exercises designed to achieve weight-bearing should be effective for the neuromuscular system and protect the knee from excessive anterior tibial translation. The aim of this study was to assess the anterior tibial translation under two different body-weight shift exercises at 2 weeks post-ACL reconstruction and compare this with healthy controls. Methods Seven patients at 2 weeks after ACL reconstruction and seven controls participated in the study. Dynamic sagittal tibial translation was registered during two body weight shift exercises (from side to side and forward–backward). Results There was no significant difference in maximal translation between the two exercises in the ACL-reconstructed group. In the control group, the body weight shift from side-to-side exercise resulted in larger anterior translation compared to the forward–backward body weight shift. Conclusions Two weeks after ACL reconstruction, both body weight shift exercises can be used to train body weight acceptance.

Management of associations in the IRBAC 2000 model

Secure interaction and interoperability between two or more administrative domains is a major concern. The IRBAC 2000 model accomplishes secure interaction and interoperability by flexibly dynamic inter-domain role translations. Associations are the key element of the IRBAC 2000 model, which have a great impact on security and efficiency of dynamic role translations. Therefore, it is a crucial problem how to manage the associations in the IRBAC 2000 model. There are two cases under which some matters will emerge. One is where conflicting associations may result in a security hazard. Another is where redundant associations may reduce the efficiency of dynamic role translations and increase the difficulty of management of associations. The formal definitions on conflicting associations and redundant associations are given, and the methods are discusses to judge whether there are conflicting associations or redundant associations in IRBAC 2000 model. The protective mechanism is presented, which utilizes prerequisite conditions to prevent conflicting or redundant associations from appearing in IRBAC 2000 model.

Exploration of the KT1000 as a Tool to Assess Dynamic Anterior Tibial Translation

Exploration of the KT1000 as a Tool to Assess Dynamic Anterior Tibial Translation

Compiler and runtime support for efficient software transactional memory

Programmers have traditionally used locks to synchronize concurrent access to shared data. Lock-based synchronization, however, has well-known pitfalls: using locks for fine-grain synchronization and composing code that already uses locks are both difficult and prone to deadlock. Transactional memory provides an alternate concurrency control mechanism that avoids these pitfalls and significantly eases concurrent programming. Transactional memory language constructs have recently been proposed as extensions to existing languages or included in new concurrent language specifications, opening the door for new compiler optimizations that target the overheads of transactional memory.This paper presents compiler and runtime optimizations for transactional memory language constructs. We present a high-performance software transactional memory system (STM) integrated into a managed runtime environment. Our system efficiently implements nested transactions that support both composition of transactions and partial roll back. Our JIT compiler is the first to optimize the overheads of STM, and we show novel techniques for enabling JIT optimizations on STM operations. We measure the performance of our optimizations on a 16-way SMP running multi-threaded transactional workloads. Our results show that these techniques enable transactional memory's performance to compete with that of well-tuned synchronization.

Dynamic code reordering method for improvement of Java performance

AbstractWhen the Java language is used in a large‐scale real‐time communications system, processing power presents a problem. The authors have been investigating an essential solution for this problem. This paper proposes a technique in which native code which has already been on‐line compiled by a JIT compiler is dynamically reordered, based on profiling data obtained in the Java execution environment. The realization procedure is clearly described. The technique is to place segments of compiled code with high execution frequency close to each other as a group in memory. In addition, codes with low execution frequency are allocated to another area, so that the codes with high frequencies are more compactly placed. By this measure, the instruction cache hit rate of the processor is greatly increased and the processing performance in a large‐scale program can be improved. The extent of performance improvement by applying the dynamic code reordering technique is evaluated, using applications with characteristics close to those of the real large‐scale real‐time communications program, and the effectiveness of the proposed method is demonstrated. © 2006 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 89(11): 78–88, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecja.20185

Catechizing Culture: Missionaries, Aymaras, and the "New Evangelization" (review)

Andrew Orta, Catechizing Culture: Missionaries, Aymaras, and the Evangelization. New York: Columbia University Press, October 2004. Andrew Orta's book deals (ostensibly, at least) with the implementation of Catholic missionary practices inspired by a of among Aymara-speaking peasants in the Bolivian altiplano.1 These practices constitute the latest phase in a embarked upon by the Church in Latin America after World War II (particularly after the Second Vatican Council of 1962-65) and imagined as a rectification of the flawed and violent conversion that followed the conquest. During the 1960s and 70s the New Evangelization in the altiplano, as elsewhere in Latin America, was oriented primarily by liberation theology, with its emphasis on class differences and its preferential option for the poor requiring the empowerment of lay pastoral agents and grass-roots organizations. By the time Orta began his fieldwork in the late 1980s, re-evangelization efforts had been recast in inculturationist terms, a move which involved a valorization of the Aymara peasantry's Aymaraness as opposed to a generic class condition. Inculturationism posits Christianity as a force that must become incarnated in local cultural forms if conversion is to be effective, and views culture in general-any given culture-as being already Christian in essence or potentiality. Inculturationism as a missionary strategy can take a variety of forms-particularly liturgical adaptations involving native performances and symbols, and dynamic translation practices incorporating native religious categories (things which have, of course, been going on since long before the development of a theology of inculturation). As stated in the book's opening paragraph, the application of inculturation theology results in a surprising twist to previous pastoral policies: To become more Christian, the missionaries declare, the Aymara must become more 'Indian' and return to 'the ways of the ancestors'-ways that the missionaries have come to see as local cultural expressions of Christian values (vii). Orta approaches his topic through the theoretical lense of globalization and the production of locality in globalizing contexts. Firstly, the transition from liberation theology to inculturationism is presented as symptomatic of the global shift from a politics of class to a politics of ethnicity. Orta also argues that even though inculturationism may seem to be an inherently localizing move, it is still part and parcel of global forces and tends to impose standardized and thus fictional models of local culture, in this case a generic Aymaraness that does always sit well with local practices. It is the Aymara catechists deputized by the mostly North American and European inculturationist missionaries who are engaged in the real task of the production of locality, and who are the real protagonists of the book. A second dominant theme is that of cultural and historical essentialisms, binarisms, and presuppositions of purity or continuity. Orta notes the importance of such essentialisms for missionary models and practices-in particular, inculturationst missionaries see Aymara culture, and the Aymara person, as a composite consisting of an unchanged Aymara core with an overlay of colonial or creole Christianity. At the same time, Orta is very much engaged in the critique of such essentialisms as they pervade anthropological and ethnohistorical research on the Andes. The critique is closely tied to the question of locality, which, he emphasizes, must be seen not as an embattled site of primordial authenticity but as an evanescent project continually produced by a range of situated actors (204). …

Optimizing code‐copying JIT compilers for virtual stack machines

AbstractJust‐in‐time (JIT) compilers are widely used to implement stack‐based virtual machines, such as the Java and .NET virtual machines. One disadvantage of most JIT compilers is that they are unportable; much of the back‐end is specific to the target machine. An alternative to machine‐specific code generation methods is to define a routine in a high‐level language for each virtual machine instruction. These can be compiled to native code using a normal C compiler. The native code for these routines can then be strung together, allowing very simple, unoptimized code to be produced just in time. In this paper we present such a system based on an existing implementation of the Forth language. We present a novel system of optimizations for the system based on exploiting common sequences of virtual machine instructions. We use a small domain specific language and tool to generate stack‐optimized code for sequences of virtual machine instructions, and for choosing the most useful sequences for a code‐copying compiler. By measuring the length of the resulting executable code, we allow machine‐specific sequences to be chosen without any machine‐dependent code in our system. Experimental results show that best (average) speedups of 47.2% (15.75%) are possible on a Pentium 4 machine, and even higher an a PowerPC based machine. Furthermore, our optimizations allow the size of the generated code to be reduced by an average of 17.9% on the Pentium 4, and 20.5% on the PowerPC over a wide range of programs. Copyright © 2006 John Wiley & Sons, Ltd.

Mumbo: A Rule-Based Implementation of a Run-time Program Generation Language

We describe our efforts to use rule-based programming to produce a model of Jumbo, a run-time program generation (RTPG) system for Java. Jumbo incorporates RTPG following the simple principle that the regular compiler — or, rather, its back-end — can be used both for ordinary, static compilation and for run-time compilation. This tends to produce a run-time compiler that is inefficient but potentially subject to improvement by partial evaluation. However, the complexity of the language and compiler have made it difficult for us to achieve actual optimization. The model, written in Maude, preserves all the essential ingredients of Jumbo, but operates on a simplified language, called Mumbo. The simplification in the language together with Maude's support for code rewriting has allowed us to make rapid progress. We discuss the model in detail, the kinds of optimizations we have obtained, and the impact on the Jumbo project.

Network security mechanisms utilising network address translation

A new protocol technology is just starting to emerge from the laboratory environment. Its stated purpose is to provide a means whereby networks, and the services that reside on them, can be protected from adversarial compromise. This protocol called Dynamic Network Address Translation (Dynat) is designed to protect computer networks against cyber attacks. Briefly, Dynat changes network parameters, such as the IP address and port numbers, between communication sessions and even during sessions. As a result, an adversary cannot associate activity on a given port of a given IP address with an application on a particular computer, thereby presenting a significant barrier to network attacks. This paper identifies the major components or attributes that are associated with the Dynat protocol and describes some of the potential implementations and associated network architectures that can deploy Dynat. It examines inter-operability issues associated with Dynat's interaction with other network protocols along with its impact on standard security implementations, such as IPSec and Intrusion Detection Systems.

Separation of Duty in Dynamic Role Translations Between Administrative Domains

Separation of Duty in Dynamic Role Translations Between Administrative Domains

A region-based compilation technique for dynamic compilers

Method inlining and data flow analysis are two major optimization components for effective program transformations, but they often suffer from the existence of rarely or never executed code contained in the target method. One major problem lies in the assumption that the compilation unit is partitioned at method boundaries. This article describes the design and implementation of a region-based compilation technique in our dynamic optimization framework, in which the compiled regions are selected as code portions without rarely executed code. The key parts of this technique are the region selection, partial inlining, and region exit handling. For region selection, we employ both static heuristics and dynamic profiles to identify and eliminate rare sections of code. The region selection process and method inlining decisions are interwoven, so that method inlining exposes other targets for region selection, while the region selection in the inline target conserves the inlining budget, allowing more method inlining to be performed. The inlining process can be performed for parts of a method, not just for the entire body of the method. When the program attempts to exit from a region boundary, we trigger recompilation and then use on-stack replacement to continue the execution from the corresponding entry point in the recompiled code. We have implemented these techniques in our Java JIT compiler, and conducted a comprehensive evaluation. The experimental results show that our region-based compilation approach achieves approximately 4% performance improvement on average, while reducing the compilation overhead by 10% to 30%, in comparison to the traditional method-based compilation techniques.

Runtime specialization with optimistic heap analysis

We describe a highly practical program specializer for Java programs. The specializer is powerful, because it specializes optimistically, using (potentially transient) constants in the heap; it is precise, because it specializes using data structures that are only partially invariant; it is deployable, because it is hidden in a JIT compiler and does not require any user annotations or offline preprocessing; it is simple, because it uses existing JIT compiler ingredients; and it is fast, because it specializes programs in under 1s.These properties are the result of (1) a new algorithm for selecting specializable code fragments, based on a notion of influence; (2) a precise store profile for identifying constant heap locations; and (3) an efficient invalidation mechanism for monitoring optimistic assumptions about heap constants. Our implementation of the specializer in the Jikes RVM has low overhead, selects specialization points that would be chosen manually, and produces speedups ranging from a factor of 1.2 to 6.4, comparable with annotation-guided specializers.

Evaluation of Partial File Modification for Java-Based Realtime Communication Systems

This paper proposes a Java online plug-in mechanism that can be used to modify any part in a system file coded in Java, even while the part is running, without service interruption. The Java-related plug-in capabilities are devised by using the plug-in functional elements offered by the existing C++ online plug-in that we proposed. In particular, measures on how to deal with the use of Just In Time compilation and inline expansion are considered. New linkage and file-back up techniques are proposed for this purpose. Case studies reveal its wide applicability and the degree of memory area saving effects. Evaluation proves this mechanism does not affect the performance of ordinary service processing. It is expected to be used in practice for Java-based service processing such as VoIP and Instant Messaging.