دانلود رایگان مقاله لاتین محاسبات داده فشرده از سایت الزویر
عنوان فارسی مقاله:
تخصیص منابع چند لایه برای محاسبات داده های فشرده
عنوان انگلیسی مقاله:
Multi-Tier Resource Allocation for Data-Intensive Computing
سال انتشار : 2015
برای دانلود رایگان مقاله محاسبات داده فشرده اینجا کلیک نمایید.
مقدمه انگلیسی مقاله:
1. Introduction
As the scale and size of applications continues to increase with the explosive growth in data volume (dubbed ‘Big Data’), distributed processing/computing has become rather essential. The availability of virtually unlimited resource capacity with cloud computing has greatly enabled such distributed computing. As a result, a wide range of distributed systems has been developed. Some popular examples are MapReduce data processing framework [1], Pregel graph processing system [2] and Montage astronomical image mosaic engine [3,4]. Applications in these distributed systems exhibit much heterogeneity in terms particularly of resource usage characteristics, e.g., CPU-intensive applications and data-intensive applications. Within distributed systems, resource allocation and management is an ongoing research concern; it is well known that improving resource allocation mechanisms can result in considerable real-world improvements in performance and efficiency. However, different applications have different resource requirements, mean-ing that effective resource management in heterogeneous environments is much more difficult. While there has been a large amount of research into resource allocation, most resource allocation strategies do not account for application heterogeneity. In particular, the majority of classic optimisation techniques for resource management assume homogeneity in either application or resource, or both. And thus, they often do not perform optimally in heterogeneous environments. In this paper, we address the problem of fine-grained resource allocation to distributed systems where jobs are composed of many small tasks taking into account application heterogeneity. The traditional approach to resource management in these systems has been to divide resources into logical partitions (called ‘slots’ or ‘containers’), and allocate tasks to partitions [1,5]. However, since different jobs have different resource usage characteristics, this approach can lead to both resource under-utilisation and resource contention, resulting in decreased performance. A general purpose, scheduler-independent solution to this problem is highly desirable, especially as we see increasing heterogeneity of both applications and execution environments.
برای دانلود رایگان مقاله محاسبات داده فشرده اینجا کلیک نمایید.
کلمات کلیدی:
Scholarly articles for Allocation for Data-Intensive Computing Market-oriented cloud computing: Vision, hype, and … - Buyya - Cited by 2444 … and data scheduling in distributed data-intensive … - Ranganathan - Cited by 552 Topology-aware resource allocation for data-intensive … - Lee - Cited by 100 Search Results Multi-Tier Resource Allocation for Data-Intensive Computing www.sciencedirect.com/science/article/pii/S2214579615000167 by T Ryan - 2015 - Cited by 6 - Related articles Mar 23, 2015 - In this paper, we present Multi-Tier Resource Allocation (MTRA) as a novel fine-grained resource management technique for distributed systems. ... We demonstrate the efficacy of this technique in a data-intensive computing environment, MapReduce data processing framework in Hadoop ... Fair Resource Allocation for Data-Intensive Computing ... - IEEE Xplore ieeexplore.ieee.org/document/7412778/ by S Tang - 2016 - Cited by 4 - Related articles To address the computing challenge of 'big data', a number of data-intensive computing frameworks (e.g., MapReduce, Dryad, Storm and Spark) h. Fair Resource Allocation for Data-Intensive Computing in the Cloud www.comp.nus.edu.sg/~hebs/pub/fairresource_tsc16.pdf by S Tang - Cited by 4 - Related articles allocation mechanism called Long-Term Resource Fairness (LTRF) for such computing. .... 2.1 Data-Intensive Computation in the Cloud. To enable the query ... Fair Resource Allocation for Data-Intensive Computing in the Cloud https://www.researchgate.net/.../295084089_Fair_Resource_Allocation_for_Data-Intensi... Fair Resource Allocation for Data-Intensive Computing in the Cloud on ResearchGate, the professional network for scientists.
