SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
We develop a non-cooperative game-theoretic model for the problem of graph minor-embedding to show that optimal compiling of adiabatic quantum programs in the sense of Nash equilibrium is possible.
The current phase of quantum computing is in the Noisy Intermediate-Scale Quantum (NISQ) era. On NISQ devices, two-qubit gates such as CNOTs are much noisier than single-qubit gates, so it is essential to minimize their count. Quantum circuit synthesis is a process of decomposing an arbitrary unitary into a sequence of quantum gates, and can be used as an optimization tool to produce shorter circuits to improve overall circuit fidelity. However, the time-to-solution of synthesis grows exponentially with the number of qubits. As a result, synthesis is intractable for circuits on a large qubit scale. In this paper, we propose a hierarchical, block-by-block opti-mization framework, QGo, for quantum circuit optimization. Our approach allows an exponential cost optimization to scale to large circuits. QGo uses a combination of partitioning and synthesis: 1) partition the circuit into a sequence of independent circuit blocks; 2) re-generate and optimize each block using quantum synthesis; and 3) re-compose the final circuit by stitching all the blocks together. We perform our analysis and show the fidelity improvements in three different regimes: small-size circuits on real devices, medium-size circuits on noisy simulations, and large-size circuits on analytical models. Our technique can be applied after existing optimizations to achieve higher circuit fidelity. Further, using a set of NISQ benchmarks, we show that QGo can reduce the number of CNOT gates by 29.9% on average and up to 50% when compared with industrial compiler optimizations such as t|ket). When executed on the IBM Athens system, shorter depth leads to higher circuit fidelity. We also demonstrate the scalability of our QGo technique to optimize circuits of 60+ qubits, Our technique is the first demonstration of successfully employing and scaling synthesis in the compilation tool chain for large circuits. Overall, our approach is robust for direct incorporation in production compiler toolchains to further improve the circuit fidelity.
One of the many benefits of using a high-performance computing (HPC) system at a Department of Energy (DOE) Office of Science (SC) HPC facility is the large number of software products, built and optimized for the system. The HPC center staff and HPC vendors provide optimized software such as libraries and even full scientific applications, ready to be used by users as building blocks to accelerate scientific discovery. Behind each provided packaged software module are a large number of decisions - which compiler, optimizations, variants/options - to build the software on the target system. And, even before the software gets deployed, the software must be developed, tested, and maintained, including deprecating old versions and ensuring compatibility across versions. The software lifecycle is complex and is further convoluted by a web of interdependencies on other software.
We present our experience in porting optimized CUDA implementations to oneAPI. We focus on the use case of numerical integration, particularly the CUDA implementations of PAGANI and $m$-Cubes. We faced several challenges that caused performance degradation in the oneAPI ports. These include differences in utilized registers per thread, compiler optimizations, and mappings of CUDA library calls to oneAPI equivalents. After addressing those challenges, we tested both the PAGANI and m-Cubes integrators on numerous integrands of various characteristics. To evaluate the quality of the ports, we collected performance metrics of the CUDA and oneAPI implementations on the Nvidia V100 GPU. We found that the oneAPI ports often achieve comparable performance to the CUDA versions, and that they are at most 10% slower.
In quantum computing, transpilation plays a crucial role in converting high-level, machine-independent quantum circuits into circuits specially for a quantum device, considering factors such as basis gate set, topology, error profile, etc. Yet, the efficiency of transpilation remains a significant bottleneck, particularly when dealing with very large QASM level input files. In this paper, we present QASMTrans, a C++ based high-performance quantum transpiler framework that can demonstrate on average 50-100× speedups compared to the internal transpiler of Qiskit. Particularly, for large dense circuits such as ’uccsd n24’ and ’qft n320’ incorporating millions of gates, QASMTrans can successfully transpile in 69s and 31s, respectively, while Qiskit failed to finish in one hour. Using QASMTrans as the baseline, it becomes more feasible to explore much larger design space and impose more comprehensive compiler optimizations.
Abstract The program state is formed by the values that the program manipulates. These values are stored in the stack, in the heap, or in static memory. The ability to inspect the program state is useful as a debugging or as a verification aid. Yet, there exists no general technique to insert inspection points in type‐unsafe languages such as C or C++. The difficulty comes from the need to traverse the memory graph in a so‐called uncooperative environment. In this article, we propose an automatic technique to deal with this problem. We introduce a static code transformation approach that inserts in a program the instrumentation necessary to report its internal state. Our technique has been implemented in LLVM. It is possible to adjust the granularity of inspection points trading precision for performance. In this article, we demonstrate how to use inspection points to debug compiler optimizations; to augment benchmarks with verification code; and to visualize data structures.
RDataFrame is ROOT’s high-level interface for Python and C++ data analysis. Since it first became available, RDataFrame adoption has grown steadily and it is now poised to be a major component of analysis software pipelines for LHC Run 3 and beyond. Thanks to its design inspired by declarative programming principles, RDataFrame enables the development of highperformance, highly parallel analyses without requiring expert knowledge of multi-threading and I/O: user logic is expressed in terms of self-contained, small computation kernels tied together by a high-level API. This design completely decouples analysis logic from its actual execution, and opens several interesting avenues for workflow optimization. In particular, in this work we explore the benefits of moving internal data processing from an event-by-event to a bulkby-bulk loop. This refactoring dramatically reduces the framework’s runtime overheads; in collaboration with the I/O layer it improves data access patterns; it exposes information that optimizing compilers might use to auto-vectorize the invocation of user-defined computations; finally, while existing user-facing interfaces remain unaffected, it becomes possible to additionally offer interfaces that explicitly expose bulks of events, useful e.g. for the injection of GPU kernels into the analysis workflow. In order to inform similar future R&D, design challenges will be presented, as well as an investigation of the relevant timememory trade-off backed by novel performance benchmarks.
Large parallel machines generally offer the best parallel performance with "native execution" that is achieved using codes developed with the optimized compilers, communication libraries, and runtimes offered on the machines. In this paper, we report and analyze performance results from native execution of deep learning on a leadership-class high-performance computing (HPC) system. Using our new code called DeepEx, we present a study of the parallel speed up and convergence rates of learning achieved with native parallel execution. In the trade-off between computational parallelism and synchronized convergence, we first focus on maximizing parallelism while still obtaining convergence. Scaling results are reported from execution on up to 15,000 GPUs using two scientific data sets from atom microscopy and protein folding applications, and also using the popular ImageNet data set. In terms of the traditional measure of parallel speed up, excellent scaling is observed up to 12,000 GPUs. Additionally, accounting for convergence rates of deep learning accuracy or error, a deep learning-specific metric called "learning speed up" is also tracked. The performance results indicate the need to evaluate parallel deep learning execution in terms of learning speed up, and point to additional directions for improved exploitation of high-end HPC systems.
In applications where efficiency is critical, developers may examine their compiled binaries, seeking to understand how the compiler transformed their source code and what performance implications that transformation may have. This analysis is challenging due to the vast number of disassembled binary instructions and the many-to-many mappings between them and the source code. These problems are exacerbated as source code size increases, giving the compiler more freedom to map and disperse binary instructions across the disassembly space. Interfaces for disassembly typically display instructions as an unstructured listing or sacrifice the order of execution. Here, we design a new visual interface for disassembly code that combines execution order with control flow structure, enabling analysts to both trace through code and identify familiar aspects of the computation. Central to our approach is a novel layout of instructions grouped into basic blocks that displays a looping structure in an intuitive way. We add to this disassembly representation a unique block-based mini-map that leverages our layout and shows context across thousands of disassembly instructions. Finally, we embed our disassembly visualization in a web-based tool, DisViz, which adds dynamic linking with source code across the entire application. DizViz was developed in collaboration with program analysis experts following design study methodology and was validated through evaluation sessions with ten participants from four institutions. Participants successfully completed the evaluation tasks, hypothesized about compiler optimizations, and noted the utility of our new disassembly view. Our evaluation suggests that our new integrated view helps application developers in understanding and navigating disassembly code.
Novel "converged" applications combine phases of scientific simulation with data analysis and machine learning. Each computational phase can benefit from specialized accelerators. However, algorithms evolve so quickly that mapping them on existing accelerators is suboptimal or even impossible. This paper presents the SODA (Software Defined Accelerators) framework, a modular, multi-level, open-source, no-human-in-the-loop, hardware synthesizer that enables end-to-end generation of specialized accelerators. SODA is composed of SODA-Opt, a high-level frontend developed in MLIR that interfaces with domain-specific programming frameworks and allows performing system level design, and Bambu, a state-of-the-art high-level synthesis engine that can target different device technologies. The framework implements design space exploration as compiler optimization passes. We show how the modular, yet tight, integration of the high-level optimizer and lower-level HLS tools enables the generation of accelerators optimized for the computational patterns of converged applications. We then discuss some of the research opportunities that such a framework allows, including system-level design, profile driven optimization, and supporting new optimization metrics.
Explore the source record for details and available documents.
This project helped narrow the gap between near-term hardware and practical applications of quantum computing through the development and implementation of optimized compilation and error mitigation techniques. These advances were enabled by low-level control available on the Advanced Quantum Testbed, and led to methods that characterized and improved the platform’s performance on quantum algorithms. Results of this work were published in the scientific literature and used to inform internal software development of partner Super.tech.
This project builds toward a comprehensive error-mitigating toolkit that makes quantum programming more robust and adaptive to the noisy, resource-limited nature of today’s quantum hardware. To that end, it integrates established error-mitigation methods — such as zero-noise extrapolation and dynamical decoupling — directly into compiler infrastructures. These techniques will be packaged as modules that can automatically adjust and combine based on performance analysis, enabling compilers to explore large design spaces and produce optimized, low-noise quantum programs with minimal manual intervention. In parallel, this project also explores new approaches to analog quantum programming or quantum simulation, and has developed the programming language SimuQ which treats quantum Hamiltonian evolution as the central object.
This report summarizes Unitary Fund’s contributions to the Department of Energy’s TEAM project (DE-SC0020266) under Thrust 2: Quantum Programming and Compilation. The central outcomes of this work have been the development of Mitiq, an open-source Python toolkit for applying quantum error mitigation (QEM) techniques to noisy quantum programs, and the invention, benchmarking and theoretical investigation of novel QEM techniques. Additional outcomes include the development of other open source software packages for the usage, simulation and control of quantum computers.
This report summarizes our contributions to the Department of Energy’s Tough Errors are no Match (TEAM) project (DE-SC0020377) under Thrust 2: Quantum Programming and Compilation. The central outcomes of this work included a novel efficient quantum compiling algorithm which works without requiring the quantum computer to exactly invert its operations, answering a longstanding open problem in quantum compiling. Additional results include the implementation of zero-noise extrapolation error mitigation in collaboration with the Unitary Fund, as well as novel quantum algorithms for entanglement detection and pseudorandomness.
Following the mass adoption of external accelerators for high performance computing, the overall performance of many applications has become increasingly dependent on relatively small accelerated kernels. As static analysis is fundamentally limited by dynamic values and external definitions, standard ahead-of-time compilation is not always sufficient to achieve the best performance. Furthermore, many users looking to port an existing application to run on an external accelerator will not want to fundamentally restructure their programs. These and other problems can be addressed through both link-time optimization (LTO) and just-in-time (JIT) compilation, but until now had sparse and inconsistent support from the compiler. In this work, we present a new compilation method that enables device-side LTO as well as a transparent JIT compilation tool-chain for OpenMP target offloading. Our contributions include an entirely new device linking and embedding scheme to enable LTO as well as a novel JIT engine to efficiently optimize OpenMP offloading regions at run-time. We also introduce a persistent caching system to improve end-to-end runtime using the JIT engine and minimize kernel launching overheads. We measure the performance of our LTO and JIT implementation via several real-world scientific applications. With our optimizations we observe significant improvements through LTO on large applications as well as significant end-to-end execution time improvement using JIT.