Eight mainstream cloud management platform comparison shopping guide

According to Forrester forecasts, the private cloud market will increase from 2011 to 7.8 billion U.S. dollars to 2020 of 15 billion dollars, such a delicious cake naturally no lack of coveted people. However, the private cloud management platform as the core technology of private cloud, the individual vendors are very active, and there are some open source management solutions. This article will be the current market mainstream eight cloud management platform to introduce, and at the end of the article to compare.

1. OpenStack

OpenStack is an open source project designed to provide software for the construction and management of public and private clouds. With more than 130 companies and 1350 developers in its community, the first task of the OpenStack project is to simplify the cloud deployment process and bring it good scalability.

The OpenStack includes several community-maintained projects:

OpenStack Compute (Nova) is a set of controllers that enable you to start a virtual machine instance for a single user or using a group. It can also be used to set up a network for a specific project that contains multiple instances. OpenStack compute is comparable to Amazon EC2 in public cloud processing and is no less of a private cloud than a VMware product. In a public cloud, this set of management mechanisms will provide a prefabricated mirror or provide a storage mechanism for the mirrors created by the user, so that the user can start the mirror in the form of a virtual machine.

OpenStack object Storage (Object storage, Swift) is a set of systems for implementing object storage in large-scale scalable systems with built-in redundancy and fault-tolerant mechanisms. These objects can be recovered through a rest API or a client that can butt the object storage API like Cyberduck.

OpenStack Imaging Service (mirroring services, glance) is a set of virtual machine image lookup and retrieval system. It can be configured in three forms: using the OpenStack object storage mechanism to store mirroring, storing information directly using Amazon's simple storage solution (S3), or combining S3 storage with object storage as S3 access connectors. The OpenStack mirroring service supports a variety of virtual machine mirroring formats, including VMware (VMDK), Amazon mirrors (AKI, ARI, AMI), and various disk formats supported by VirtualBox. The container format for mirrored metadata includes Amazon's Aki, Ari, and AMI information, standard OVF format, and binary large data.

The relationship of three projects is shown in the following illustration:

Figure OpenStack The relationship between three projects

The recent inclusion of the new members of the "core" project group on the basis of the vote was a "Keystone" identity service and a set of Django-based dashboard interfaces, the latter code-named "Horizon".

2. Eucalyptus

Eucalyptus cloud computing software provides an abstraction of these resources on a single platform (split into open source and Enterprise Editions). Eucalyptus's source is open. And there are packages available to CentOS 5,debian squeeze,opensuse 11,fedora 12.

Eucalyptus chooses Xen and KVM as the hypervisor for virtualization. The current version is 2.0.3. The enterprise version of Eucalyptus has provided support for vsphere Esx/esxi.

Eucalyptus components:

The Eucalyptus contains five major components that collaborate to provide the required cloud services together. These components communicate securely with each other using a ws-security SOAP message delivery.

Cloud Controller (CLC) is within the Eucalyptus cloud, which is the main controller component responsible for managing the entire system. It is the main entry point for all users and administrators into the Eucalyptus cloud. All clients communicate with the CLC only through a SOAP-or REST-based API. The CLC is responsible for passing requests to the correct components, collecting them, and sending responses from those components back to the client. This is the external "window" of the Eucalyptus cloud.

This controller component within the Cluster Controller (CC) Eucalyptus is responsible for managing the entire virtual instance network. Requests are sent to CC via a SOAP-or REST-based interface. CC maintains all information about the Node Controller running within the system and is responsible for controlling the lifecycle of these instances. It routes requests that open virtual instances to Node Controller with available resources.

Node Controller (NC) It controls the host operating system and the corresponding hypervisor (Xen or recent KVM, which will soon support VMWare). You must run an instance of NC on each machine that hosts the actual virtual instance (instantiated from the request from CC).

Walrus (W) This controller component manages access to the storage services within the Eucalyptus. Requests are passed to Walrus via a SOAP-or REST-based interface.

This storage service within the Storage Controller (SC) Eucalyptus implements Amazon's S3 interface. SC works jointly with Walrus to store and access virtual machine images, kernel images, RAM disk images, and user data. The VM images can be public or private, and are initially stored in a compressed and encrypted format. These images are decrypted only if a node needs to start a new instance and request access to the image. The client relies on the solution, but it may appear in the form of browser scripts, a user-space application, or even a kernel module. The cloud controller is the interface from the client into the cloud and provides a logical decision for the cloud. Cloud Control performs a service that authenticates the client and translates the client's request into a transaction. A cluster controller is a collection of node controllers. It is responsible for state information and interacts with the provided virtual machines. However, the node itself interacts with both online and offline virtual machines through a hypervisor (Xen or KVM). A node controller exists as an independent instance on a separate machine. However, multiple node controllers comprise a specific cloud.

3.IBM SKC

SKC (Starter Kit for Cloud, infrastructure Cloud Rapid Deployment Solution suite) is a cloud management platform built on the POWERVM and Vmcontrol of IBM's entry-level market, which enables "Getting started cloud" based on a system resource pool (Entry Cloud) "To ensure the landing of cloud service management in a more economical and simple way.

According to IBM Cloud computing development manager Chen Hilen, SKC is from the real needs of Chinese customers, by more than 60 engineers over 1.5 of the time, developed by IBM Global Research and development agencies, created by the global release of standard products. Currently not only support the power server, but also support the x86 server, the future will also support the Z-series Big Machine.

SKC is deployed on the virtualization platform, with the Self service interface, users can access through the Web interface, because of its use of open IaaS REST API, user-friendly personalized development.

IBM SKC Features:

IBM SKC as an entry-level cloud platform management platform, and its other cloud management platform is not divided into multiple components or modules, but a whole software. The following three features are achieved:

Creating images-simplifying the storage of hundreds of images

-Easily create new golden master images and software applications with standard operating systems

-Image conversion from a physical machine or from a different x86 system

-Conserve storage space, reduce the number of images and the need for storage capacity

Deploy virtual machines-accelerate the delivery time for new applications from months to 2-3 days

-Apply an image across deployment of computing resources and storage resources

-end user self-service, improving service response

-Through resource and virtual machine isolation, Project user access control to ensure information security

-Easy to use-no need to understand the details of the entire infrastructure

-Fully support the current virtualized environment to protect your investment

-Dynamic capacity expansion to ensure continuous operation, optimize performance

Cloud operation-efficient operation and cost reduction

-Delegate delivery Management authority to cloud users to improve productivity

-Complete avoidance of negligence through automated approvals/rejections to ensure optimal operation and cloud security

-Standardize deployment and configuration, improve compliance, and reduce errors by setting policies, defaults, and templates

-Streamline management of projects, users, workloads, resources, billing, approvals, and metrology through an intuitive interface

In cloud computing billing, IBM China system and technology research and Development Center architect Cao Yu said cloud computing billing is divided into four main steps:

1. event-based monitoring system, primarily monitoring the change in the use of IT resources by each end user

2. Metering, recording data per user for each type of it resource each time period

3. Billing, according to the specified rate, the use of detailed list of resources into the cost, the formation of a detailed list of costs

4. Budget control, according to the detailed cost of the specified period of time to the user's account deduction fee, over the budget of the user account does not allow the use of IT resources, the advantage is that the application of IT resources transparent, while forcing the business system to use IT resources and return.

Figure based on the IBM SKC Cloud computing Infrastructure

4. Microsoft System Center 2012

When introducing System Center 2012, Lilin, Microsoft's senior Product marketing manager, said it was a unified management platform for cloud computing, infrastructure and applications, and emphasized that everything was application-centric. In addition, System Center 2012 is introduced to provide users with a flexible and efficient IT infrastructure, highly available IT service levels (SLAs), and to help users manage their private cloud and public cloud environments in a unified manner.

System Center 2012 enables global views to unify the enterprise's IT resources, including managing various virtualization platforms, network resources, and storage resources, as well as Cross-platform management, including Windows,unix/linux, and managing private clouds, public clouds, mixed clouds across the cloud, This helps the CIO focus more on translating IT assets into strategic resources for the enterprise.

Components and Features:

System Center2012 contains eight components that implement different functions, and users can purchase the corresponding components according to their own needs.

Microsoft Private Cloud solutions are built using Windows Server with Hyper-V and System Center. System Center, which is an important core of private cloud solutions, is described as a unified management platform for cloud computing, infrastructure, and applications, and it is noteworthy that System Center 2012 consolidates eight separate component products into one unified solution, To shorten the installation and deployment time from days to hours, the following is a detailed description of these eight components.

System Center Virtual Machine Manager

This is a core component of the Microsoft Private cloud and Datacenter management solution that will help you manage virtualized and cloud environments by empowering you with the following capabilities.

* Build a flexible and affordable infrastructure solution that will achieve infrastructure as a service (IaaS) by virtualizing your existing computing resources, network resources, and storage resources and allocating them to your business units appropriately.

* Provide service-centric management capabilities to your applications, allowing you to manage these applications independently of the current infrastructure.

* It can provide support for environments that contain multiple virtualization, including Hyper-V, VMware, and Citrix, thereby maximizing the potential value of existing data center investments.

* Maintain high availability of business-critical applications through dynamic optimization of virtualized datacenter resources.

System Center Operations Manager

The monitoring capabilities component of a private cloud solution to help users manage their own data centers and cloud environments:

* Provide flexible and affordable enterprise-class monitoring and diagnostics, while minimizing total cost ownership (TCO) by leveraging the full range of business hardware, system configurations, and heterogeneous environments.

* Make every effort to ensure the availability of critical business applications and services by leveraging the industry-leading. NET Application performance monitoring and diagnostics capabilities, and with the JEE application running state monitoring capabilities.

* Provide an integrated view of the data center, private cloud, and public cloud.

System Center App Controller

Self-service functional components in a private cloud solution that give the application owner the following capabilities:

• Easily configure, deploy, and manage services with the help of a service-centric visual interface and a standard Template library.

* Directly using Web pages as an interface to present a customized resource view based on your role in the organization, you can focus on creating, managing, and migrating services without having to bother with the server.

* View a wide range of private cloud services and public cloud services, as well as individual virtual machines, and implement fine-grained control over components located at all levels, while also being able to track various jobs and keep a detailed history of various change operations.

System Center orchestrator

The process automation component of the Microsoft Private Yunnan solution, which helps users leverage existing knowledge and resources to build flexible and economically appropriate infrastructure process automation. It will help streamline and standardize various processes in existing data centers, thereby increasing the potential return on investment.

* Optimize the use of existing data center investments by integrating, expanding, and interacting with a variety of heterogeneous tools and systems.

* help you achieve flexible and reliable data center services by allowing processes from each business area to be switched from one to the other to coordinate.

* Reduce error-prone manual operations by automating various private cloud application scenarios, reducing costs and increasing the expected ability to apply requirements.

System Center Service Manager