Concurrent sets with range query operations are highly desirable in applications such as in-memory databases. However, few set implementations offer range queries. Known techniques for augmenting data structures with range queries (or operations that can be used to build range queries) have numerous problems that limit their usefulness. For example, they impose high overhead or rely heavily on garbage collection. In this work, we show how to augment data structures with highly efficient range queries, without relying on garbage collection. We identify a property of epoch-based memory reclamation algorithms that makes them ideal for implementing range queries, and produce three algorithms, which use locks, transactional memory and lock-free techniques, respectively. Our algorithms are applicable to more data structures than previous work, and are shown to be highly efficient on a large scale Intel system.

Replicas 1 of a vertex play an important role in existing distributed graph processing systems which make a single vertex to be parallel processed by multiple machines and access remote neighbors locally without any remote access. However, replicas of vertices introduce data coherency problem. Existing distributed graph systems treat replicas of a vertex v as an atomic and indivisible vertex, and use an eager data coherency approach to guarantee replicas atomicity. In eager data coherency approach, any changes to vertex data must be immediately communicated to all replicas of v, thus leading to frequent global synchronizations and communications. In this paper, we propose a lazy data coherency approach, called LazyAsync, which treats replicas of a vertex as independent vertices and maintains the data coherency by computations, rather than communications in existing eager approach. Our approach automatically selects some data coherency points from the graph algorithm, and maintains all replicas to share the same global view only at such points, which means the replicas are enabled to maintain different local views between any two adjacent data coherency points. Based on PowerGraph, we develop a distributed graph processing system LazyGraph to implement the LazyAsync approach and exploit graph-aware optimizations. On a 48-node EC2-like cluster, LazyGraph outperforms PowerGraph on four widely used graph algorithms across a variety of real-world graphs, with a speedup ranging from 1.25x to 10.69x.

It is well known that modern functional programming languages are naturally amenable to parallel programming. Achieving efficient parallelism using functional languages, however, remains difficult. Perhaps the most important reason for this is their lack of support for efficient in-place updates, i.e., mutation, which is important for the implementation of both parallel algorithms and the run-time system services (e.g., schedulers and synchronization primitives) used to execute them. In this paper, we propose techniques for efficient mutation in parallel functional languages. To this end, we couple the memory manager with the thread scheduler to make reading and updating data allocated by nested threads efficient. We describe the key algorithms behind our technique, implement them in the MLton Standard ML compiler, and present an empirical evaluation. Our experiments show that the approach performs well, significantly improving efficiency over existing functional language implementations.

Many numerical algorithms are naturally expressed as operations on tensors (i.e. multi-dimensional arrays). Hence, tensor expressions occur in a wide range of application domains, e.g. quantum chemistry and physics; big data analysis and machine learning; and computational fluid dynamics. Each domain, typically, has developed its own strategies for efficiently generating optimized code, supported by tools such as domain-specific languages, compilers, and libraries. However, strategies and tools are rarely portable between domains, and generic solutions typically act as ''black boxes'' that offer little control over code generation and optimization. As a consequence, there are application domains without adequate support for easily generating optimized code, e.g. computational fluid dynamics. In this paper we propose a generic and easily extensible intermediate language for expressing tensor computations and code transformations in a modular and generative fashion. Beyond being an intermediate language, our solution also offers meta-programming capabilities for experts in code optimization. While applications from the domain of computational fluid dynamics serve to illustrate our proposed solution, we believe that our general approach can help unify research in tensor optimizations and make solutions more portable between domains.

This paper examines the causes and extent of code size overhead caused by the ARM calling convention in Thumb-2 binaries. We show that binaries generated from C++ source files generally have higher amounts of calling convention overhead, and present a binary file optimizer to eliminate some of that overhead. Calling convention overhead can negatively impact power consumption, flash memory costs, and chip size in embedded or otherwise resource-constrained domains. This is particularly true on platforms using "compressed" instruction sets, such as the 16-bit ARM Thumb and Thumb-2 instruction sets, used in virtually all smartphones and in many other smaller-scale embedded devices. In this paper, we examine the extent of calling convention overhead in practical software, and compare the results of C and C++ programs, and find that C++ programs generally have a higher percentage of calling-convention overhead. Finally, we demonstrate a tool capable of eliminating some of this overhead, particularly in the case of C++ programs, by modifying the calling conventions on a per-procedure basis.

Just-in-time (JIT) compilation during program execution and ahead-of-time (AOT) compilation during software installation are alternate techniques used by managed language virtual machines (VM) to generate optimized native code while simultaneously achieving binary code portability and high execution performance. Profile data collected by JIT compilers at run-time can enable profile-guided optimizations (PGO) to customize the generated native code to different program inputs. AOT compilation removes the speed and energy overhead of online profile collection and dynamic compilation, but may not be able to achieve the quality and performance of customized native code. The goal of this work is to investigate and quantify the implications of the AOT compilation model on the quality of the generated native code for current VMs. First, we quantify the quality of native code generated by the two compilation models for a state-of-the-art (HotSpot) Java VM. Second, we determine how the amount of profile data collected affects the quality of generated code. Third, we develop a mechanism to determine the accuracy or similarity for different profile data for a given program run, and investigate how the accuracy of profile data affects its ability to effectively guide PGOs. Finally, we categorize the profile data types in our VM and explore the contribution of each such category to performance.

As future computing hardware progresses towards extreme-scale technology, new challenges arise for addressing heterogeneous compute and memory resources, for providing application resilience in the presence of more frequent failures, and for working within strict energy constraints. While C++ has gained popularity in recent years within the HPC community, some concepts of object-oriented program design may be at odds with the techniques we use to address the challenges of extreme-scale computing. In this work, we focus on the challenges related to using aggregate data structures that include pointer values within a programming model where the runtime may frequently relocate data, and traditional serialization techniques are not practical. We propose and evaluate a marshalled encoding for relocatable data blocks, and present a C++ library and other tools to simplify the work of the application programmer developing new applications or porting existing applications to such emerging programming models.

Traditional scientific and emerging data analytics applications require fast, power-efficient, large, and persistent memories. Combining all these characteristics within a single memory technology is expensive and hence future supercomputers will feature different memory technologies side-by-side. However, it is a complex task to program hybrid-memory systems and to identify the best object-to-memory mapping. We envision that programmers will probably resort to use default configurations that only require minimal interventions on the application code or system settings. In this work, we argue that intelligent, fine-grained data placement can achieve higher performance than default setups. We present an algorithm for data placement on hybrid-memory systems. Our algorithm is based on a set of single-object allocation rules and global data placement decisions. We also present RTHMS, a tool that implements our algorithm and provides recommendations about the object-to-memory mapping. Our experiments on a hybrid memory system, an Intel Knights Landing processor with DRAM and HBM, show that RTHMS is able to achieve higher performance than the default configuration. We believe that RTHMS will be a valuable tool for programmers working on complex hybrid-memory systems.

A memory consistency model specifies which writes to shared memory a given read may see. Ambiguities or errors in these specifications can lead to bugs in both compilers and applications. Yet architectures usually define their memory models with prose and litmus tests—small concurrent programs that demonstrate allowed and forbidden outcomes. Recent work has formalized the memory models of common architectures through substantial manual effort, but as new architectures emerge, there is a growing need for tools to aid these efforts. This paper presents MemSynth, a synthesis-aided system for reasoning about axiomatic specifications of memory models. MemSynth takes as input a set of litmus tests and a framework sketch that defines a class of memory models. The sketch comprises a set of axioms with missing expressions (or holes). Given these inputs, MemSynth synthesizes a completion of the axioms—i.e., a memory model—that gives the desired outcome on all tests. The MemSynth engine employs a novel embedding of bounded relational logic in a solver-aided programming language, which enables it to tackle complex synthesis queries intractable to existing relational solvers. This design also enables it to solve new kinds of queries, such as checking if a set of litmus tests unambiguously defines a memory model within a framework sketch. We show that MemSynth can synthesize specifications for x86 in under two seconds, and for PowerPC in 12 seconds from 768 litmus tests. Our ambiguity check identifies missing tests from both the Intel x86 documentation and the validation suite of a previous PowerPC formalization. We also used MemSynth to reproduce, debug, and automatically repair a paper on comparing memory models in just two days.