Task (computing)

About: Task (computing) is a research topic. Over the lifetime, 9718 publications have been published within this topic receiving 129364 citations.

Method and system for executing batch jobs by delegating work to independent service providers

Abstract: A job management apparatus for use in a batch job execution system is provided. The job management apparatus includes a client communications part which receives a batch job from a client, an extracting part which extracts a task from the batch job, and an assigning part which receives signals from the service providers and delegates a task to a service provider who is capable of performing such task. The job management apparatus is in communication with a job database which stores the batch job after it is received from the client. At least one provider manager is in communication with the job management apparatus and in communication with the service providers which monitors the tasks as they are being performed on a service provider and provides status information about the task to the job management apparatus. The service providers are configured to output signals to the job management apparatus requesting work.

Computer system and method for sharing a job with other computers on a computer network using IP multicast

Abstract: A computer system on a network uses IP multicast to recruit other computer systems to share in the processing of a job. If a computer system on the network wants to be available to process shared jobs, it first registers for job sharing by invoking an IP multicast router at a particular IP address. All messages sent to the IP multicast router are broadcast to all computer systems that are registered with the router. When a computer system has a job to share, it recruits other computer systems to help process the job by sending a message to the IP multicast router that corresponds to a request to share the job. The candidate computer systems that receive the recruiter's broadcast determine if they can share the job according to one or more job sharing parameters. These parameters may relate to the job itself, network performance, security, or other criteria for sharing. If a computer system meets the parameters for taking on the particular job, it responds to the recruiter. If the recruiter still needs help (e.g., if not enough candidate systems have responded yet), the recruiter grants the response and delivers the job to the computer system. The computer system then performs the job (or task) and returns the results to the recruiter.

UniT: Multimodal Multitask Learning with a Unified Transformer

Abstract: We propose UniT, a Unified Transformer model to simultaneously learn the most prominent tasks across different domains, ranging from object detection to natural language understanding and multimodal reasoning. Based on the transformer encoder-decoder architecture, our UniT model encodes each input modality with an encoder and makes predictions on each task with a shared decoder over the encoded input representations, followed by task-specific output heads. The entire model is jointly trained end-to-end with losses from each task. Compared to previous efforts on multi-task learning with transformers, we share the same model parameters across all tasks instead of separately fine-tuning task-specific models and handle a much higher variety of tasks across different domains. In our experiments, we learn 7 tasks jointly over 8 datasets, achieving strong performance on each task with significantly fewer parameters. Our code is available in MMF at this https URL.

A distributed anytime algorithm for dynamic task allocation in multi-agent systems

Abstract: We introduce a novel distributed algorithm for multi-agent task allocation problems where the sets of tasks and agents constantly change over time. We build on an existing anytime algorithm (fast-max-sum), and give it significant new capabilities: namely, an online pruning procedure that simplifies the problem, and a branch-and-bound technique that reduces the search space. This allows us to scale to problems with hundreds of tasks and agents. We empirically evaluate our algorithm against established benchmarks and find that, even in such large environments, a solution is found up to 31% faster, and with up to 23% more utility, than state-of-the-art approximation algorithms. In addition, our algorithm sends up to 30% fewer messages than current approaches when the set of agents or tasks changes.

Event synchronization analysis for debugging parallel programs

Abstract: One of the major difficulties of explicit parallel programming for a shared memory machine model is detecting the potential for nondeterminacy and identifying its causes. There will often be shared variables in a parallel program, and the tasks comprising the program may need to be synchronized when accessing these variables. This paper discusses this problem and presents a method for automatically detecting non-determinacy in parallel programs that utilize event style synchronization instructions, using the Post, Wait, and Clear primitives. With event style synchronization, especially when there are many references to the same event, the difficulty lies in computing the execution order that is guaranteed given the synchronization instructions and the sequential components of the program. The main result in this paper is an algorithm that computes such an execution order and yields a Task Graph upon which a nondeterminacy detection algorithm can be applied. We have focused on events because they are a frequently used synchronization mechanism in parallel versions of Fortran, including Cray [Cray87], IBM [IBM88], Cedar [GPHL88], and PCF Fortran [PCF88].