ترغب بنشر مسار تعليمي؟ اضغط هنا

A Survey and Annotated Bibliography of Workflow Scheduling in Computing Infrastructures: Community, Keyword, and Article Reviews -- Extended Technical Report

79   0   0.0 ( 0 )
 نشر من قبل Laurens Versluis
 تاريخ النشر 2020
  مجال البحث الهندسة المعلوماتية
والبحث باللغة English




اسأل ChatGPT حول البحث

Workflows are prevalent in todays computing infrastructures. The workflow model support various different domains, from machine learning to finance and from astronomy to chemistry. Different Quality-of-Service (QoS) requirements and other desires of both users and providers makes workflow scheduling a tough problem, especially since resource providers need to be as efficient as possible with their resources to be competitive. To a newcomer or even an experienced researcher, sifting through the vast amount of articles can be a daunting task. Questions regarding the difference techniques, policies, emerging areas, and opportunities arise. Surveys are an excellent way to cover these questions, yet surveys rarely publish their tools and data on which it is based. Moreover, the communities that are behind these articles are rarely studied. We attempt to address these shortcomings in this work. We focus on four areas within workflow scheduling: 1) the workflow formalism, 2) workflow allocation, 3) resource provisioning, and 4) applications and services. Each part features one or more taxonomies, a view of the community, important and emerging keywords, and directions for future work. We introduce and make open-source an instrument we used to combine and store article meta-data. Using this meta-data, we 1) obtain important keywords overall and per year, per community, 2) identify keywords growing in importance, 3) get insight into the structure and relations within each community, and 4) perform a systematic literature survey per part to validate and complement our taxonomies.



قيم البحث

اقرأ أيضاً

Realistic, relevant, and reproducible experiments often need input traces collected from real-world environments. We focus in this work on traces of workflows---common in datacenters, clouds, and HPC infrastructures. We show that the state-of-the-art in using workflow-traces raises important issues: (1) the use of realistic traces is infrequent, and (2) the use of realistic, {it open-access} traces even more so. Alleviating these issues, we introduce the Workflow Trace Archive (WTA), an open-access archive of workflow traces from diverse computing infrastructures and tooling to parse, validate, and analyze traces. The WTA includes ${>}48$ million workflows captured from ${>}10$ computing infrastructures, representing a broad diversity of trace domains and characteristics. To emphasize the importance of trace diversity, we characterize the WTA contents and analyze in simulation the impact of trace diversity on experiment results. Our results indicate significant differences in characteristics, properties, and workflow structures between workload sources, domains, and fields.
Scientific workflows are a cornerstone of modern scientific computing. They are used to describe complex computational applications that require efficient and robust management of large volumes of data, which are typically stored/processed at heterog eneous, distributed resources. The workflow research and development community has employed a number of methods for the quantitative evaluation of existing and novel workflow algorithms and systems. In particular, a common approach is to simulate workflow executions. In previous work, we have presented a collection of tools that have been used for aiding research and development activities in the Pegasus project, and that have been adopted by others for conducting workflow research. Despite their popularity, there are several shortcomings that prevent easy adoption, maintenance, and consistency with the evolving structures and computational requirements of production workflows. In this work, we present WorkflowHub, a community framework that provides a collection of tools for analyzing workflow execution traces, producing realistic synthetic workflow traces, and simulating workflow executions. We demonstrate the realism of the generated synthetic traces by comparing simulated executions of these traces with actual workflow executions. We also contrast these results with those obtained when using the previously available collection of tools. We find that our framework not only can be used to generate representative synthetic workflow traces (i.e., with workflow structures and task characteristics distributions that resembles those in traces obtained from real-world workflow executions), but can also generate representative workflow traces at larger scales than that of available workflow traces.
Projections onto sets are used in a wide variety of methods in optimization theory but not every method that uses projections really belongs to the class of projection methods as we mean it here. Here projection methods are iterative algorithms that use projections onto sets while relying on the general principle that when a family of (usually closed and convex) sets is present then projections (or approximate projections) onto the given individual sets are easier to perform than projections onto other sets (intersections, image sets under some transformation, etc.) that are derived from the given family of individual sets. Projection methods employ projections (or approximate projections) onto convex sets in various ways. They may use different kinds of projections and, sometimes, even use different projections within the same algorithm. They serve to solve a variety of problems which are either of the feasibility or the optimization types. They have different algorithmic structures, of which some are particularly suitable for parallel computing, and they demonstrate nice convergence properties and/or good initial behavior patterns. This class of algorithms has witnessed great progress in recent years and its member algorithms have been applied with success to many scientific, technological, and mathematical problems. This annotated bibliography includes books and review papers on, or related to, projection methods that we know about, use, and like. If you know of books or review papers that should be added to this list please contact us.
Internet supercomputing is an approach to solving partitionable, computation-intensive problems by harnessing the power of a vast number of interconnected computers. For the problem of using network supercomputing to perform a large collection of ind ependent tasks, prior work introduced a decentralized approach and provided randomized synchronous algorithms that perform all tasks correctly with high probability, while dealing with misbehaving or crash-prone processors. The main weaknesses of existing algorithms is that they assume either that the emph{average} probability of a non-crashed processor returning incorrect results is inferior to $frac{1}{2}$, or that the probability of returning incorrect results is known to emph{each} processor. Here we present a randomized synchronous distributed algorithm that tightly estimates the probability of each processor returning correct results. Starting with the set $P$ of $n$ processors, let $F$ be the set of processors that crash. Our algorithm estimates the probability $p_i$ of returning a correct result for each processor $i in P-F$, making the estimates available to all these processors. The estimation is based on the $(epsilon, delta)$-approximation, where each estimated probability $tilde{p_i}$ of $p_i$ obeys the bound ${sf Pr}[p_i(1-epsilon) leq tilde{p_i} leq p_i(1+epsilon)] > 1 - delta$, for any constants $delta >0$ and $epsilon >0$ chosen by the user. An important aspect of this algorithm is that each processor terminates without global coordination. We assess the efficiency of the algorithm in three adversarial models as follows. For the model where the number of non-crashed processors $|P-F|$ is linearly bounded the time complexity $T(n)$ of the algorithm is $Theta(log{n})$, work complexity $W(n)$ is $Theta(nlog{n})$, and message complexity $M(n)$ is $Theta(nlog^2n)$.
Improving datacenter operations is vital for the digital society. We posit that doing so requires our community to shift, from operational aspects taken in isolation to holistic analysis of datacenter resources, energy, and workloads. In turn, this s hift will require new analysis methods, and open-access, FAIR datasets with fine temporal and spatial granularity. We leverage in this work one of the (rare) public datasets providing fine-grained information on datacenter operations. Using it, we show strong evidence that fine-grained information reveals new operational aspects. We then propose a method for holistic analysis of datacenter operations, providing statistical characterization of node, energy, and workload aspects. We demonstrate the benefits of our holistic analysis method by applying it to the operations of a datacenter infrastructure with over 300 nodes. Our analysis reveals both generic and ML-specific aspects, and further details how the operational behavior of the datacenter changed during the 2020 COVID-19 pandemic. We make over 30 main observations, providing holistic insight into the long-term operation of a large-scale, public scientific infrastructure. We suggest such observations can help immediately with performance engineering tasks such as predicting future datacenter load, and also long-term with the design of datacenter infrastructure.
التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا