Parallel task-based programming models, like OpenMP, allow application developers to easily create a parallel version of their sequential codes. The standard OpenMP 4.0 introduced the possibility of describing a set of data dependences per task that the runtime uses to order the tasks execution. This order is calculated using shared graphs, which are updated by all threads in exclusive access using synchronization mechanisms (locks) to ensure the dependence management correctness. The contention in the access to these structures becomes critical in many-core systems because several threads may be wasting computation resources waiting their turn.
This paper proposes an asynchronous management of the runtime structures, like task dependence graphs, suitable for task-based programming model runtimes. In such organization, the threads request actions to the runtime instead of doing them directly. The requests are then handled by a distributed runtime manager (DDAST) which does not require dedicated resources. Instead, the manager uses the idle threads to modify the runtime structures. The paper also presents an implementation, analysis and performance evaluation of such runtime organization. The performance results show that the proposed asynchronous organization outperforms the speedup obtained by the original runtime for different benchmarks and different many-core architectures.
Maximizing parallelism level in applications can be achieved by minimizing overheads due to load imbalances and waiting time due to memory latencies. Compiler optimization is one of the most effective solutions to tackle this problem. The compiler is able to detect the data dependencies in an application and is able to analyze the specific sections of code for parallelization potential. However, all of these techniques provided with a compiler are usually applied at compile time, so they rely on static analysis, which is insufficient for achieving maximum parallelism and producing desired appl…
We evaluate and compare four contemporary and emerging runtimes for high-performance computing(HPC) applications: Cilk, Charm++, ParalleX and AM++. We compare along three bases: programming model, execution model and the implementation on an underlying machine model. The comparison study includes a survey of each runtime system's programming models, their corresponding execution models, their stated features, and performance and productivity goals. We first qualitatively compare these runtimes with programmability in mind. The differences in expressivity and programmability arising from their …
Asynchronous Many-task (AMT) runtime systems have gained increasing acceptance in the HPC community due to the performance improvements offered by fine-grained tasking runtime systems. At the same time, C++ standardization efforts are focused on creating higher-level interfaces able to replace OpenMP or OpenACC in modern C++ codes. These higher level functions have been adopted in standards conforming runtime systems such as HPX, giving users the ability to simply utilize fork-join parallelism in their own codes. Despite innovations in runtime systems and standardization efforts users face eno…
Runtime verification is the process of verifying critical behavioral properties in big complex systems, where formal verification is not possible due to state space explosion. There have been several attempts to design efficient algorithms for runtime verification. Most of these algorithms have a formally defined correctness property as a reference and check whether the system consistently meets the demands of the property or it fails to satisfy the property at some point in runtime. LTL is a commonly used language for defining these kinds of properties and is also the language of focus in thi…
We present TaskTorrent, a lightweight distributed task-based runtime in C++. TaskTorrent uses a parametrized task graph to express the task DAG, and one-sided active messages to trigger remote tasks asynchronously. As a result the task DAG is completely distributed and discovered in parallel. It is a C++14 library and only depends on MPI. We explain the API and the implementation. We perform a series of benchmarks against StarPU and ScaLAPACK. Micro benchmarks show it has a minimal overhead compared to other solutions. We then apply it to two large linear algebra problems. TaskTorrent scales v…
Distributed Stochastic Gradient Descent (SGD) when run in a synchronous manner, suffers from delays in runtime as it waits for the slowest workers (stragglers). Asynchronous methods can alleviate stragglers, but cause gradient staleness that can adversely affect the convergence error. In this work, we present a novel theoretical characterization of the speedup offered by asynchronous methods by analyzing the trade-off between the error in the trained model and the actual training runtime(wallclock time). The main novelty in our work is that our runtime analysis considers random straggling dela…
GPUs are readily available in cloud computing and personal devices, but their use for data processing acceleration has been slowed down by their limited integration with common programming languages such as Python or Java. Moreover, using GPUs to their full capabilities requires expert knowledge of asynchronous programming. In this work, we present a novel GPU run time scheduler for multi-task GPU computations that transparently provides asynchronous execution, space-sharing, and transfer-computation overlap without requiring in advance any information about the program dependency structure. W…
The PGAS model is well suited for executing irregular applications on cluster-based systems, due to its efficient support for short, one-sided messages. However, there are currently two major limitations faced by PGAS applications. The first relates to scalability: despite the availability of APIs that support non-blocking operations in special cases, many PGAS operations on remote locations are synchronous by default, which can lead to long-latency stalls and poor scalability. The second relates to productivity: while it is simpler for the developer to express all communications at a fine-gra…
Swim extends the actor model to support applications composed of linked distributed actors that continuously analyze boundless streams of events from millions of sources, to respond in-sync with the real-world. Swim builds a running application from streaming events, creating a distributed dataflow graph of linked, stateful, concurrent streaming actors that is overlaid on a mesh of runtime instances. Streaming actors are vertices in the dataflow graph that concurrently analyze new events and modify their states. A link is an edge in the graph and is a URI binding to an actor's streaming API. T…
Asynchronous programming models (APM) are gaining more and more traction, allowing applications to expose the available concurrency to a runtime system tasked with coordinating the execution. While MPI has long provided support for multi-threaded communication and non-blocking operations, it falls short of adequately supporting APMs as correctly and efficiently handling MPI communication in different models is still a challenge. Meanwhile, new low-level implementations of light-weight, cooperatively scheduled execution contexts (fibers, aka user-level threads (ULT)) are meant to serve as a bas…
Open-source matters, not just to the current cohort of HPC users but also to potential new HPC communities, such as machine learning, themselves often rooted in open-source. Many of these potential new workloads are, by their very nature, far more asynchronous and unpredictable than traditional HPC codes and open-source solutions must be found to enable new communities of developers to easily take advantage of large scale parallel machines. Task-based models have the potential to help here, but many of these either entirely abstract the user from the distributed nature of their code, placing e…
Benchmarking and comparing performance of a scientific simulation across hardware platforms is a complex task. When the simulation in question is constructed with an asynchronous, many-task (AMT) runtime offloading work to GPUs, the task becomes even more complex. In this paper, we discuss the use of a uniquely suited performance measurement library, APEX, to capture the performance behavior of a simulation built on HPX, a highly scalable, distributed AMT runtime. We examine the performance of the astrophysics simulation carried-out by Octo-Tiger on two different supercomputing architectures. …
IBM T. J. Watson Research Center（IBM托马斯·J·沃森研究中心）；Weill Cornell Medical College（威尔康奈尔医学院）
This paper presents Rudra, a parameter server based distributed computing framework tuned for training large-scale deep neural networks. Using variants of the asynchronous stochastic gradient descent algorithm we study the impact of synchronization protocol, stale gradient updates, minibatch size, learning rates, and number of learners on runtime performance and model accuracy. We introduce a new learning rate modulation strategy to counter the effect of stale gradients and propose a new synchronization protocol that can effectively bound the staleness in gradients, improve runtime performance…
The next generation HPC and data centers are likely to be reconfigurable and data-centric due to the trend of hardware specialization and the emergence of data-driven applications. In this paper, we propose ARENA -- an asynchronous reconfigurable accelerator ring architecture as a potential scenario on how the future HPC and data centers will be like. Despite using the coarse-grained reconfigurable arrays (CGRAs) as the substrate platform, our key contribution is not only the CGRA-cluster design itself, but also the ensemble of a new architecture and programming model that enables asynchronous…
University of California, Berkeley（加州大学伯克利分校）；Duke University（杜克大学）
Task-based distributed frameworks (e.g., Ray, Dask, Hydro) have become increasingly popular for distributed applications that contain asynchronous and dynamic workloads, including asynchronous gradient descent, reinforcement learning, and model serving. As more data-intensive applications move to run on top of task-based systems, collective communication efficiency has become an important problem. Unfortunately, traditional collective communication libraries (e.g., MPI, Horovod, NCCL) are an ill fit, because they require the communication schedule to be known before runtime and they do not pro…
