Resource management - Cloud-Scale Resource Management...

Info icon This preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Cloud-Scale Resource Management: Challenges and Techniques Ajay Gulati VMware, Inc. [email protected] Ganesha Shanmuganathan VMware, Inc. [email protected] Anne Holler VMware, Inc. [email protected] Irfan Ahmad VMware, Inc. [email protected] Abstract Managing resources at large scale while providing per- formance isolation and efficient use of underlying hard- ware is a key challenge for any cloud management soft- ware. Most virtual machine (VM) resource management systems like VMware DRS clusters, Microsoft PRO and Eucalyptus, do not currently scale to the number of hosts and VMs supported by cloud service providers. In addi- tion to scale , other challenges include heterogeneity of systems, compatibility constraints between virtual ma- chines and underlying hardware, islands of resources cre- ated due to storage and network connectivity and limited scale of storage resources. In this paper, we shed light on some of the key issues in building cloud-scale resource management systems, based on five years of research and shipping cluster re- source management products. Furthermore, we discuss various techniques to provide large scale resource man- agement, along with the pros and cons of each technique. We hope to motivate future research in this area to de- velop practical solutions to these issues. 1 Introduction Managing compute and IO resources at large scale in both public and private clouds is quite challenging. The success of any cloud management software critically de- pends on the flexibility, scale and efficiency with which it can utilize the underlying hardware resources while pro- viding necessary performance isolation [ 5 , 8 , 11 ]. Cus- tomers expect cloud service providers to deliver quality of service (QoS) controls for tenant VMs. Thus, resource management at cloud scale requires the management plat- form to provide a rich set of resource controls that balance the QoS of tenants with overall resource efficiencies of datacenters. For public clouds, some systems (e.g., Amazon EC2) provide largely a one-to-one mapping between virtual and physical CPU and memory resources. This leads to poor consolidation ratios and customers are unable to exploit the benefits from statistical multiplexing that they enjoy in private clouds. Some recent studies [ 4 ] have done a cost comparison of public vs. private clouds and have come up with a mantra of don’t move to the cloud, but virtualize . For private clouds, resource management solutions like VMware DRS [ 2 ] and Microsoft PRO [ 1 ] have led to better performance isolation, higher utilization of un- derlying hardware resources via over-commitment and overall lower cost of ownership. Unfortunately, the rich set of controls provided by these solutions does not eas- ily scale to cloud environments. As part of a team, we have worked on VMware’s shipping Distributed Re- source Scheduler (DRS) product and are in the process of prototyping and evaluating a scaled version of DRS.
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern