Home › Categories › Technology› UC Berkeley AMPLab Research

UC Berkeley AMPLab Research.pdf

Mesos-A-Platform-for-Fine-Grained-Resource-Sharing-in-the-Data-Center.pdf

🔗 Source: amplab.cs.berkeley.edu

📊 Size: 1.09 MB

📄 Pages: 14 pages

⬇️ Downloads: 137

Summary

Mesos is a platform for sharing commodity clusters between multiple diverse cluster computing frameworks, such as Hadoop and MPI, improving cluster utilization and avoiding per-framework data replication. It shares resources in a fine-grained manner, allowing frameworks to achieve data locality by taking turns reading data stored on each machine. Mesos introduces a distributed two-level scheduling mechanism called resource offers, where Mesos decides how many resources to offer each framework, and frameworks decide which resources to accept and which computations to run on them. Results show that Mesos can achieve near-optimal data locality, scale to 50,000 nodes, and is resilient to failures.

Download File

Description

Mesos is a platform for sharing commodity clusters between multiple diverse cluster computing frameworks, such as Hadoop and MPI, improving cluster utilization...

Technical Information

File Format: PDF
File Size: 1.09 MB
Pages: 14
Language: EN
Total Downloads: 137
Last Updated: 2 weeks ago

Document Overview

This PDF document about UC Berkeley AMPLab research provides comprehensive information and guidance. Whether you're a beginner or advanced user, this resource offers valuable insights into UC Berkeley AMPLab research.