The explosive growth of Mobility, Cloud, Social, Big Data, and Internet of Things (IOT) presents tremendous opportunities as well as treats for businesses today. Enterprises of all industries face the urgency to innovate in new business models and services. Innovation speed, business agility, and operational efficiency are becoming the primary success factors for businesses to stay relevant and to compete in today's fast growing digital economy.
“Data center”, the cornerstone of a business, needs to transform in order to effectively enable business innovation with fast time-to-market, agility, and efficiency. Virtualization and Cloud Computing play a key role in today's data center transformation. They help enable data center resource consolidation, increase resource sharing, simplify IT management and operation, and improve data center automation, which leads to less human errors and downtime.
Today, most of large enterprises and service providers have acquired experiences in implementing their first generation cloud projects based on proprietary, commercial, off-the-shelf Virtualization and Cloud Computing technologies. Moving forward, in order to meet future’s fast growth in number of users, subscriptions, application services, devices, storage, traffic, bandwidth needs, enterprises and service providers will need to look for an alternative that can help them meet the needs of scalability, faster innovation in features and functionalities, cost effectiveness, globalization, and etc.
OpenStack is an ubiquitous Open Source Cloud Computing platform that is designed to meet the needs of public and private clouds regardless of size, by being simple to implement and massively scalable.
The technology consists of a group of interrelated projects that control large pools of computing, storage, and networking resources throughout a data center managed through a dashboard or via the OpenStack API. OpenStack works with popular enterprises and open source technologies making it ideal for heterogeneous infrastructure. OpenStack APIs are also compatible with Amazon EC2 and Amazon S3, and thus client applications written for Amazon Web Services can be used with OpenStack with minimal porting effort.
OpenStack has a modular architecture with various code names for its components: Compute (Nova), Image Service (Glance), Object Storage (Swift), Dashboard (Horizon), Identify Service (Keystone), Networking (Neutron), Block Storage (Cinder), Orchestration (Heat), Telemetry (Ceilometer), Database (Trove), Elastic Map Reduce (Sahara), Bare Metal Provisioning (Ironic), Multiple Tenant Cloud Messaging (Zaqar), Shared File System Service (Manila), DNSaaS (Designate), Security API (Barbican), and etc.
OpenStack is developed by an open global community, and managed by the OpenStack Foundation, a non-profit corporate entity established in September 2012 to promote OpenStack software and its community. More than 800 companies have joined the project, including major leading telecommunication companies, IT technology vendors, service providers, enterprises, and etc. from North America, EMEA, APAC, and Latin America. The OpenStack community collaborates around a six-month, time-based release cycle with frequent development milestones.
The OpenStack Foundation along with its supporting companies including AT&T, Ubuntu, HP, IBM, Intel, Rackspace, Redhat, SUSE, Huawei, Apitra, CCAT, Cisco, DreamHost, EMC, Ericsson, Hitachi, Juniper, Mirantis, NEC, NetApp, Symantec, Yahoo, Odin, and the recently joined Google, etc., serving more than 18,000 individual members from over 140 countries around the world.
Enterprises and service providers today are looking at OpenStack as the next generation cloud platform that can help their data center transform into a large, highly scalable, automated fabric of physical resources and virtualized services in a heterogeneous data center, controlled by software automation and APIs to enable new development model, DevOps, and to create innovative services for Mobility, Social, Big Data, IOT, Network Function Virtualization (NFV), and etc.
• Faster Speed of Innstrong>tion - Within only 3 years since its inception, OpenStack has achieved tremendous progress in technology innovation and maturity. This is due to the large size of global developer and user communities as well as the leaderships and resource support from leading sponsoring companies. The speed of innovation definitely beats any single company trying to create a proprietary solution with their own R&D team and limited resour br /> />
• More Quality Assurance - Since OpenStack is an open source project developed in an "open" community. There are more users testing all type of scenarios, and more developers working on bug fixes. In addition, a wide range of vendors supporting the community also help with interoperability testing and certification. For example, Huawei has many Centers of Excellence in Santa Clara (US), Xian (China), Beijing (China), Israel, and Europe dedicated for quality assurance and interoperability testing with other vendors' OpenStack supporting products and technologies.
• Interoperability - OpenStack allows for mixed hypervisor IT environments which will become more fragmented over time. Having the benefit of "Interoperability" simplifies management and support, and hence decreases data center cost and risks.
• No Vendor Lock-In - OpenStack supports multiple hardware products and software technologies as long as they support and adhere to OpenStack standards and APIs. This benefit significantly decreases or minimizes integration work and spending.
• Investment Protection - In addition to the support of multiple hypervisors, and large number of hardware and software technologies and products in its ecosystem, OpenStack will have more longevity than any proprietary cloud platforms due to its large member of user support, and investments and commitments made by significant number of vendors. The success of OpenStack will exponentially grow as the enterprise adoption continues to grow. OpenStack is definitely here to stay, according to not just developers and users, but also industry leading analysts.
Huawei is fully committed to OpenStack's success and its entire ecosystem starting from the OpenStack Grizzly release timeframe, June 2013. Huawei's Cloud OS, FusionSphere 5.0 and beyond, is built based on OpenStack. Within only 2 years, Huawei's code contribution for the later releases of OpenStack (Juno, Liberty) has been ranked in one of the top 10 companies in the OpenStack community.
Huawei, currently a Gold member of the OpenStack Foundation, has been a loyal and significant contributor of OpenStack in providing vision, software development, testing, bug fixes, resources, and the protection and promotion of the OpenStack brand. Moving forward, Huawei will continue our leadership, commitment, contribution to the OpenStack Foundation, and developer and user communities.
Huawei built it's commercially used Cloud OS, FusionSphere, based on native OpenStack, thus it is compatible with the OpenStack ecosystem. This enables Huawei FusionSphere to support all hypervisors and software and hardware products that are certified on OpenStack. In addition, Huawei added OpenStack expansions to improve the performance, reliability, availability, security, ease-of-use, compatibility, automation, and management of OpenStack for its customers' commercial use. All the expansions and enhancements are implemented based on the native OpenStack plug-ins/drivers. There are no changes made in the OpenStack trunk code. Drivers of other vendors' hardware can be easily integrated into Huawei's FusionSphere. This ensures the integrity of OpenStack core platform within Huawei FusionSphere while providing Huawei's customers a production ready, enterprise grade, cloud platform.
Mobility, Cloud, Social, and IOT are rapidly blurring the boundaries between telecommunication (CT) and IT. We are experiencing the height of ICT convergence in data centers. Huawei FusionSphere differentiates itself from other OpenStack based cloud operating systems in the area of ICT support:
• Reduced Latency - FusionSphere hypervisor is built with the technology to improve communication performance of virtual machines (VMs) by 1.5 to 3 times, significantly reducing latency.
• Higher Performance - FusionSphere's cloud resource scheduling allocates virtualized resources according the affinity between network elements to improve VM performance. In addition, a "performance priority" policy is available to ensure that all the VMs for a logical network element (multiple
VMs working together to provide a complete service function) stay close physically in the network topology to guarantee optimized latency and throughput between VMs.
• Higher Reliability - High Availability (HA) and Fault Tolerance (FT) features are built in FusionSphere to include resource redundancy, fault detection, fault recovery to minimize or eliminate system and application downtime.
FusionSphere adopts SDS (software defined storage), a state-of-the-art distributed storage system that provides low latency, high performance, scale-out storage capability. FusionSphere virtualizes various storage systems and enables storage resources to form resource pools, building application-oriented storage systems with an unified management.
In addition, FusionSphere uses an intelligent storage resource scheduling algorithm to deliver optimal performance and resource utilization. Compared to traditional disk arrays, FusionSphere SDS increases the IOPS of applications by 2 to 3 times, improves the throughput by 6 times, reduces the latency between server and storage by 20% to 50%, and expands the peak I/O bandwidth by 10 times.
FusionSphere uses an overlayed virtualization network component that is based on software-defined networking (SDN) and virtual extensible LAN (VxLAN) to construct a full mesh Layer 2 network in one, and across multiple data centers, simplifying application development and making resource scheduling flexible. Being aware of topologies and QoS, the SDN controller dynamically selects the correct routing path and adjusts the WAN routing policy and bandwidth among multiple data centers, reducing cross-data-center link bandwidth cost by 60% to 70% on average.
Huawei's SDN is directly programmable, agile, centrally managed, programmatically configured, and based on open standards and is vendor neutral. Its differentiated offerings are DC-SDN, DCI-SDN, WAN-SDN, and Policy-SDN.
FusionSphere is often deployed for Huawei carrier customers' NFV projects to implement one or more virtualized network functions or VNFs. FusionSphere is the Cloud OS for NFV Infrastructure (NFVI). It offers the following benefits:
• Reduced CapEx and OpEx through consolidated hardware resources and benefit from the economy of scale.
• Shortened time-to-market by minimizing the typical network operator cycle of innovation.
• Increased resource sharing across services and customer base.
• Network services can be launched and scaled up or down dynamically and quickly
• An open platform that facilitates more and faster innovation and total network function transformation.
• A dynamic cloud platform to enable Network-as-a-Service for customers
Huawei SD-DC2 is Huawei's next generation ICT converged data center reference architecture designed to consolidate and to virtualiz physical resources such as computing, storage, and networking resources from one data center, or across multiple distributed data centers with orchestration and unified management creating a super large data center housing millions of virtual hosts.
The Cloud OS that powers Huawei SD-DC2 is FusionSphere which delivers the following benefits in a data center :
• Openness - FusionSphere supports OpenStack APIs, thus Huawei SD-DC2 is an open data center architecture that is compatible with heterogeneous physical resources and virtualized platforms supported in the OpenStack ecosystem.
• Agility - The time to start a virtual data center (VDC) for a new business initiative can be decreased from years or months to hours; the time to start a new service can be decreased from months or days to minutes. The resources allocated for services can be dynamically resized due to current workload, and be over-provisioned for high availability. SD-DC2 is a data center game changer that helps enable fast service innovation with reliability.
• Efficiency - All resources, physical or virtual, from one vendor or multiple vendors, cloud or bare metal, can be pooled and standardized under one unified management. SD-DC2 significantly improves the overall operation and management, which leads to improved data center efficiency and significant cost reduction.
Huawei SD-DC2 has been successfully deployed in the various development stages of data center transformation: Virtualization, Private Cloud Building, Data Center Consolidation, and Hybrid Cloud Enablement; spanning across various industries such as government & public service agencies, transportation, energy & electricity, financial services, media & entertainment, large multi-national enterprises, service providers, and carriers.
OpenStack was launched 3 years ago and has achieved tremendous momentum in developer enthusiasm, user support, enterprise adoption, and industry recognition. OpenStack will continue to evolve and to introduce new services and capabilities with more reliability and maturity as each new release becomes available. For the last two years, there have been many solid OpenStack use cases emerging from notable Fortune 100 enterprises (Disney, BMW, Wal-Mart, Best Buy, etc), large carriers (AT&T, Orange, China Mobile, China Unicom, etc), and service providers (RackSpace, Comcast, Huawei Public Cloud, etc) which claim to be using OpenStack in their production environments. All of these use cases are the true testimonials of OpenStack's validity and maturity as an "enterprise ready" cloud platform. Moving forward, we are expecting to see an acceleration of OpenStack adoption as large enterprises, carriers, and service providers are incorporating OpenStack cloud platform into their data centers that are geared for the future growth of Cloud Services, Mobility, Social, Big Data, IOT, and etc.
More and more enterprises, carriers, and service providers are seeking OpenStack to launch their data center transformation journey. The main drivers of OpenStack adoption in the data centers are to avoid vendor lock-in, faster speed to innovate, more scalability, more cost effectiveness via the promise of the portability and interoperability of services designed once, and can be run anywhere in an OpenStack based architecture. <
OpenStack is playing, and will continue to play a significant role in driving data center transformation, and enabling IT to be an effective partner of Lines of Business.