HJ-OpenCL

The cloud is an increasingly popular platform to deploy applications as it lets cloud users to provide resources to their applications as needed. Furthermore, cloud providers are now starting to offer a pay-as-you-use model in which users are only charged for the resources that are really used instead of paying for a statically sized instance. This new model allows cloud users to save money, and cloud providers to better utilize their hardware. However, applications running on top of runtime environments such as the Java Virtual Machine (JVM) cannot benefit from this new model because they cannot dynamically adapt the amount of used resources at runtime. In particular, if an application needs more memory than what was initially predicted at launch time, the JVM will not allow the application to grow its memory beyond the maximum value defined at launch time. In addition, the JVM will hold memory that is no longer being used by the application. This lack of dynamic vertical scalability completely prevents the benefits of the pay-as-you-use model, and forces users to over-provision resources, and to lose money on unused resources. We propose a new JVM heap sizing strategy that allows the JVM to dynamically scale its memory utilization according to the application's needs. First, we provide a configurable limit on how much the application can grow its memory. This limit is dynamic and can be changed at runtime, as opposed to the current static limit that can only be set at launch time. Second, we adapt current Garbage Collection policies that control how much the heap can grow and shrink to better fit what is currently being used by the application. The proposed solution is implemented in the OpenJDK 9 HotSpot JVM, the new release of OpenJDK. Changes were also introduced inside the Parallel Scavenge collector and the Garbage First collector (the new by-default collector in HotSpot). Evaluation experiments using real workloads and data show that, with negligible throughput and memory overhead, dynamic vertical memory scalability can be achieved. This allows users to save significant amounts of money by not paying for unused resources, and cloud providers to better utilize their physical machines.

Many widely used, latency sensitive, data-parallel distributed systems, such as HDFS, Hive, and Spark choose to use the Java Virtual Machine (JVM), despite debate on the overhead of doing so. This paper analyzes the extent and causes of the JVM performance overhead in the above mentioned systems. Surprisingly, we find that the warm-up overhead, i.e., class loading and interpretation of bytecode, is frequently the bottleneck. For example, even an I/O intensive, 1GB read on HDFS spends 33% of its execution time in JVM warm-up, and Spark queries spend an average of 21 seconds in warm-up.The findings on JVM warm-up overhead reveal a contradiction between the principle of parallelization, i.e., speeding up long running jobs by parallelizing them into short tasks, and amortizing JVM warm-up overhead through long tasks. We solve this problem by designing HotTub, a new JVM that amortizes the warm-up overhead over the lifetime of a cluster node instead of over a single job by reusing a pool of already warm JVMs across multiple applications. The speed-up is significant. For example, using HotTub results in up to 1.8X speedups for Spark queries, despite not adhering to the JVM specification in edge cases.

While the Java Virtual Machine (JVM) plays a vital role in ensuring correct executions of Java applications, testing JVMs via generating and running class files on them can be rather challenging. The existing techniques, e.g., ClassFuzz and Classming, attempt to leverage the power of fuzzing and differential testing to cope with JVM intricacies by exposing discrepant execution results among different JVMs, i.e., inter-JVM discrepancies, for testing analytics. However, their adopted fuzzers are insufficiently guided since they include no well-designed seed and mutator scheduling mechanisms, leading to inefficient differential testing. To address such issues, in this paper, we propose SJFuzz, the first JVM fuzzing framework with seed and mutator scheduling mechanisms for automated JVM differential testing. Overall, SJFuzz aims to mutate class files via control flow mutators to facilitate the exposure of inter-JVM discrepancies. To this end, SJFuzz schedules seeds (class files) for mutations based on the discrepancy and diversity guidance. SJFuzz also schedules mutators for diversifying class file generation. To evaluate SJFuzz, we conduct an extensive study on multiple representative real-world JVMs, and the experimental results show that SJFuzz significantly outperforms the state-of-the-art mutation-based and generation-based JVM fuzzers in terms of the inter-JVM discrepancy exposure and the class file diversity. Moreover, SJFuzz successfully reported 46 potential JVM issues, and 20 of them have been confirmed as bugs and 16 have been fixed by the JVM developers.

In Memory Computing is moving data into dynamic memory which is traditionally stored in secondary storage devices. In this paper we try to improve the capability of Java Virtual Machine (JVM) to process large data collections/application by pertinence of In-Memory Computing concepts. It helps JVM/processor to process huge data sets efficiently due to high-speed access and reduction of latency of slow secondary storage devices. It is implemented by providing more dynamic memory to JVM or by the implementation of In Memory Database concepts. In this paper we correlate Actual machine with JVM and try to improve ability of JVM to process large data collections by eliminating disk access and replacing Database with JAVA Collections in the Model Layer hence using `in memory' concepts, Henceforth a high performance system is designed.

Many embedded systems feature processors coupled with a small and fast scratchpad memory. To the difference with caches, allocation of data to scratchpad memory must be handled by software. The major gain is to enhance the predictability of memory accesses latencies. A compile-time dynamic allocation approach enables eviction and placement of data to the scratchpad memory at runtime. Previous dynamic scratchpad memory allocation approaches aimed to reduce average-case program execution time or the energy consumption due to memory accesses. For real-time systems, worst-case execution time is the main metric to optimize. In this paper, we propose a WCET-directed algorithm to dynamically allocate static data and stack data of a program to scratchpad memory. The granularity of placement of memory transfers (e.g. on function, basic block boundaries) is discussed from the perspective of its computation complexity and the quality of allocation.

Memory allocation is the process of allocating physical or virtual memory space to computer applications and processes. There are two types of memory allocation: static memory allocation and dynamic memory allocation. Static memory allocation in C allocates memory during compile time using stack. It is mainly used when the amount of memory that is required by the program is known. Once the memory is allocated using static memory allocation then the size of the memory allocated cannot change whereas dynamic memory allocation allocates memory to a process/application during its run time. It is used when the amount of memory that a program requires is unknown. It uses a heap for managing memory and the size of the memory allocated can be altered. In this paper, a dynamic memory manager is proposed in Linux OS. The proposed dynamic memory manager is based upon the internal design and implementation of malloc() and free() library functions in C. It is capable of allocating and deallocating memory to a user program/process. It also shows the memory statistics of every object malloc()-ed by a program/process that is currently in use The proposed memory manager overcomes the problem of internal fragmentation.

Dynamic memory allocators are critical components of modern systems, and developers strive to find a balance between their performance and their security. Unfortunately, vulnerable allocators are routinely abused as building blocks in complex exploitation chains. Most of the research regarding memory allocators focuses on popular and standardized heap libraries, generally used by high-end devices such as desktop systems and servers. However, dynamic memory allocators are also extensively used in embedded systems but they have not received much scrutiny from the security community.In embedded systems, a raw firmware image is often the only available piece of information, and finding heap vulnerabilities is a manual and tedious process. First of all, recognizing a memory allocator library among thousands of stripped firmware functions can quickly become a daunting task. Moreover, emulating firmware functions to test for heap vulnerabilities comes with its own set of challenges, related, but not limited, to the re-hosting problem.To fill this gap, in this paper we present HEAPSTER, a system that automatically identifies the heap library used by a monolithic firmware image, and tests its security with symbolic execution and bounded model checking. We evaluate HEAPSTER on a dataset of 20 synthetic monolithic firmware images — used as ground truth for our analyses — and also on a dataset of 799 monolithic firmware images collected in the wild and used in real-world devices. Across these datasets, our tool identified 11 different heap management library (HML) families containing a total of 48 different variations. The security testing performed by HEAPSTER found that all the identified variants are vulnerable to at least one critical heap vulnerability. The results presented in this paper show a clear pattern of poor security standards, and raise some concerns over the security of dynamic memory allocators employed by IoT devices.

Frequent dynamic memory allocations (DyMAs) can significantly hinder the scalability of parallel multi-threaded programs. As the number of threads grows, DyMAs can even become the main performance bottleneck. We introduce modern tools and methods for evaluating the impact of DyMAs and present techniques for its reduction, which include scalable heap implementations, small buffer optimization, and memory pooling. Additionally, we provide a survey of state-of-the-art implementations of these techniques and study them experimentally by using a benchmark program, server simulator software, and a real-world high-performance computing application. As a result, we show that relatively small modifications in parallel program's source code or a way of its execution may substantially reduce the runtime overhead associated with the use of dynamic data structures.

High-level languages such as Java increase both productivity and portability with productive language features such as managed runtime, type safety, and precise exception semantics. Additionally, Java 8 provides parallel stream APIs with lambda expressions to facilitate parallel programming for mainstream users of multi-core CPUs and many-core GPUs. These high-level APIs avoid the complexity of writing natively running parallel programs with OpenMP and CUDA/OpenCL through Java Native Interface (JNI). The adoption of such high-level programming models offers opportunities for enabling compilers to perform parallel-aware optimizations and code generation.While many prior approaches have the ability to generate parallel code for both multi-core CPUs and many-core GPUs from Java and other high-level languages, selection of the preferred computing resource between CPUs and GPUs for individual kernels remains one of the most important challenges since a variety of factors affecting performance such as datasets and feature of programs need to be taken into account.This paper explores the possibility of using machine learning to address this challenge. The key idea is to enable a Java runtime to select a preferable hardware device with performance heuristics constructed by supervised machine-learning techniques. For this purpose, if our JIT compiler detects a parallel stream API, 1) our compiler records features of its computation such as the parallel loop range and the number of instructions and 2) our Java runtime generates these features for constructing training data. For the results reported in this paper, we constructed a prediction model with support vector machines (SVMs) after obtaining 291 samples by running 11 applications with different data sets and optimization levels. Our Java runtime then uses the SVMs to make predictions for unseen programs.Our experimental results on an IBM POWER8 platform with NVIDIA Tesla GPUs show that our prediction model predicts a faster configuration with up to 99.0% accuracy with 5-fold cross validation. Based on these results, we conclude that supervised machine-learning is a promising approach for building performance heuristics for mapping Java applications onto accelerators.

Subject of Research. The paper presents the study of vulnerability exploitation techniques in the implementation of dynamic memory allocation algorithms (glibc library allocator): Poisoned Null-byte, Overlapped Chunks, Fastbin Attack,Unsafe Unlink, House of Einherjar, House of Force, House of Spirit, House of Lore, Unsorted Bin Attack. Examples of vulnerability exploitation code and classification of the presented techniques are given in accordance with the Common Weakness Enumeration list. The modern methods and means of vulnerabilities detection are studied; their advantages and disadvantages are shown using the Heap Hopper framework as an example. Modern methods of appropriate software verification are considered. Method. The proposed software verification method combines the approaches of static analysis and symbolic execution using an accurate model of algorithms for dynamic memory allocation. In the compilation process of program being tested, the Kripke structure is created. Dynamic memory vulnerabilities are described by temporal logic formulas. The resulting structure and formulas are passed at the input of the model checking algorithm. Concrete-symbolic execution of the assembled binary file is performed. Vulnerability conditions expressed in the form of propositional logic formulas are checked for symbolic execution paths. Main Results. The practical use of the proposed approach to detection of dynamic memory vulnerabilities in software applications is shown. Symbolic execution is implemented in the form of a low-level debugger, which reduces the operating time of algorithms due to the execution of the application being tested on a real processor. Practical Relevance. The paper presents an integrated approach for solving the problem of automatic vulnerabilities detecting at different stages of the software development life cycle. This approach is applicable for verification of the similar implementations of dynamic memory allocators, such as ptmalloc, dlmalloc, tcmalloc, jemalloc and musl.

Recent compilers allow a general-purpose program (written in a conventional programming language) that handles private data to be translated into a secure distributed implementation of the corresponding functionality. The resulting program is then guaranteed to provably protect private data using secure multi-party computation techniques. The goals of such compilers are generality, usability, and efficiency, but the complete set of features of a modern programming language has not been supported to date by the existing compilers. In particular, recent compilers PICCO and the two-party ANSI C compiler strive to translate any C program into its secure multi-party implementation, but they currently lack support for pointers and dynamic memory allocation, which are important components of many C programs. In this work, we mitigate the limitation and add support for pointers to private data and consequently dynamic memory allocation to the PICCO compiler, enabling it to handle a more diverse set of programs over private data. Because doing so opens up a new design space, we investigate the use of pointers to private data (with known as well as private locations stored in them) in programs and report our findings. Aside from dynamic memory allocation, we examine other important topics associated with common pointer use such as reference by pointer/address, casting, and building various data structures in the context of secure multi-party computation. This results in enabling the compiler to automatically translate a user program that uses pointers to private data into its distributed implementation that provably protects private data throughout the computation. We empirically evaluate the constructions and report on the performance of representative programs.

FPGA behavioral synthesis has gained significant momentum recently with the growing interests in accelerating high-performance computing applications. While the latest generation of high-level synthesis (HLS) tools has made significant progress, they still lack the support for certain high-level language features such as dynamic memory allocation, despite the fact that efficiently utilization of the on-chip memory resources in FPGAs is critical to achieve the performance and power consumption target for many designs.

In this chapter, we present our enhancements of one of the most efficient exact algorithms for the maximum clique problem—MCS algorithm by Tomita, Sutani, Higashi, Takahashi and Wakatsuki (in Proceedings of WALCOM’10, 2010, pp. 191–203). Our enhancements include: applying ILS heuristic by Andrade, Resende and Werneck (in Heuristics 18:525–547, 2012) to find a high-quality initial solution, fast detection of clique vertices in a set of candidates, better initial coloring, and avoiding dynamic memory allocation. A good initial solution considerably reduces the search tree size due to early pruning of branches related to small cliques. Fast detecting of clique vertices is based on coloring. Whenever a set of candidates contains a vertex adjacent to all candidates, we detect it immediately by its color and add it to the current clique avoiding unnecessary branching. Though dynamic memory allocation allows to minimize memory consumption of the program, it increases the total running time. Our computational experiments show that for dense graphs with a moderate number of vertices (like the majority of DIMACS graphs) it is more efficient to store vertices of a set of candidates and their colors on stack rather than in dynamic memory on all levels of recursion. Our algorithm solves p_hat1000-3 benchmark instance which cannot be solved by the original MCS algorithm. We got speedups of 7, 3000, and 13000 times for gen400_p0.9_55, gen400_p0.9_65, and gen400_p0.9_75 instances, correspondingly.

Array is an important feature in Java and Java Virtual Machine. In spite of its importance, it has not been modeled by any existing Java Virtual Machine Models. In this paper, we define an extending model which uses an existing model as a basis and give the hierarchy of these two models, to model the array. In the extending model, we model the array in three steps. The first step is adding array related instructions in a formal way. The second step is refining the type compatibility to include array types. The last step is implementing array loading process also in a formal way. In the last part of thesis, we give the future work of extending other important features in Java and Java Virtual Machine.

We present a scheme which explores the parallelism on a Java virtual machine (JVM). An algorithm, called dynamic-duplication scheduling is developed for solving the static scheduling and code generation for data flow graphs on the parallel JVM. Experimental results show that the schedule produced by the algorithm on the parallel JVM is significantly improved compared with the traditional JVM.