NIT3222 Virtualization in Computing Tutorial 1

What is the differences between virtualisation and private cloud?

The management of compute resources.

Virtualisation 不需要配置。是一个关键组件。虚拟化是云计算的基础

Private cloud 是需要配置的。虚拟化

虚拟化是一种具体技术,指把硬件资源虚拟化,实现隔离性、可扩展性、安全性、资源可充分利用等特点的产品。

CPU虚拟化的技术:Intel的VT-x,AMD的AMD-V。

虚拟化的软件:VMWare,Hyper-V,Xen,KVM,OpenVZ等

云计算

云计算是个概念,而不是具体技术。

目前云计算,大多是依赖虚拟化,通过把多台服务器实体虚拟化后,构成一个资源池,实现共同计算,共享资源。也就是现在所谓“云计算”,其实这个词提出来之前,过去的“服务器集群”就已经实现这些功能了,只不过没有现在那么先进而已。

云计算 (cloud computing)是基于互联网的相关服务的增加、使用和交付模式,通常涉及通过互联网来提供动态易扩展且经常是虚拟化的资源。云是网络、互联网的一种比喻说法。过去在图中往往用云来表示电信网,后来也用来表示互联网和底层基础设施的抽象。因此,云计算甚至可以让你体验每秒10万亿次的运算能力,拥有这么强大的计算能力可以模拟核爆炸、预测气候变化和市场发展趋势。用户通过电脑、笔记本、手机等方式接入数据中心,按自己的需求进行运算。

私有云

私有云是为了解决安全问题的一种云计算方案。虽然公有云很方便,但是毕竟是把数据和资源都交给其它人了,这对于企业管理来说,是一个很大的隐患。毕竟,很多数据是无价的,很多企业把用户数据作为他们的灵魂支柱。因此,当今的企业常常不希望把数据和资源部署到公有云上,他们往往会自己出资搭建机房,这时的解决方案,就是私有云方案。私有云比起公有云,更加安全,但是,成本也更高。

What is the difference between a parent and child partition?

存储空间分区:根分区和子分区。

Parent partition (or root partition)

Runs the virtualisation stack and has direct access to hardware

The parent partition is created when Hyper-V is installed

The parent partition executes the Virtualisation Service Provider (VSP)

Communicates directly with device drivers and services hardware requests made by child partitions

Child partition

Primarily runs the applications that are needed for the business

Ultimately, child partitions run applications

A hypervisor aware operating system will run the Virtualised Service Client (VSC)

This is the client for the Virtualisation Service Provider (VSP) made available by the parent partition

A hypervisor aware operating system is like a paravirtualised operating system

What is the NUMA? How to scale up server with NUMA?

NUMA (Non-uniform memory access)

Helps hosts scale up the number of cores and memory access

Partitions cores and memory into “nodes”

Allocation and latency depends on the memory location relative to a processor

High performance applications detect NUMA and minimize cross-node memory access

Scaling up: Physical NUMA

为了提高内存的读写速度可以将某些内存直接分配给CPU,形成一些逻辑结点,实现物理physical NUMA。

This is optimal:

Memory allocation and thread allocations within the same NUMA node

Memory populated in each NUMA node

Guest NUMA

Presenting NUMA topology within VM

Guest operating systems & apps can make intelligent NUMA decisions about thread and memory allocation

Guest NUMA nodes are aligned with host resources

Policy driven per host –
best effort, or force alignment

How to? 通过兼容性设置:

Allows live migration across different CPU versions within the same processor family

Configure compatibility on a per-VM basis

Presents a normalized virtual processor within the VM

What is dynamic memory in virtualization?

Startup RAM

Amount of memory configured at startup

Maximum RAM

Maximum memory VM is allowed to use

Memory Buffer

Percentage of extra memory to reserve for a VM

Percentage is based on memory committed to the guest OS

Can be used for file caching and I/O intensive applications

Memory Weight (aka Memory Priority)

Prioritizes memory allocation to VMs when physical memory is under pressure

Dynamic Memory is a memory allocation feature for Microsoft Hyper-V which pools the host’s memory and sends resources to virtual machines (VMs) that need it. When Hyper-V Dynamic Memory is enabled, the VM declares how much memory it needs and the host provides that quantity plus an extra amount to be used as a buffer.

文章已创建 16

发表评论

电子邮件地址不会被公开。 必填项已用*标注

相关文章

开始在上面输入您的搜索词,然后按回车进行搜索。按ESC取消。

返回顶部