What’s LXC?

LXC is a userspace interface for the Linux kernel containment features. Through a powerful API and simple tools, it lets Linux users easily create and manage system or application containers. LXC is something in the middle between chroot on stereoids and a full fledged virtual machine. LXC create an environment as close as possible to standard linux installation without the need for a separate kernel.

What is Docker?

The idea behind docker is to create a lightweight portable self sufficient application container that will run virtually anywhere. Docker is an open source engine that automates the deployment of any application as lightweight portable self sufficient container that will run virtually anywhere on developer laptops, on virtual machines, bare-metal servers, OpenStack clusters, CloudStack clusters, public instances, or combination of the above.

Docker containers can encapsulate any payload, and will run consistently on and between virtually any server. Common use cases for Docker include:

• Automating the packaging and deployment of applications
• Creation of lightweight, private PAAS environments
• Automated testing and continuous integration/deployment
• Deploying and scaling web apps, databases and backend services

Docker is developed using Go language and relies on LXC (Linux Containers). There is a roadmap to extend it to other container technology like Solaris, BSD Zones and Windows by implementing various containers technology as drivers. But for the near future docker only works on Linux.

LXC Release 1.0

On 20 ^th February 2014 LXC 1.0 version was released and set a significant milestone for Linux Containers and considered as production ready. This release features wide variety of improvenments to container security, a consistent set of tools, updated documentation and an API with multiple language bindings.

LXC 1.0 and Dockers

Current version of docker is 0.8 and due to heavy pace of development, it is still advised not be used in production (yet). Although some companies are already using docker in production and offering services based on it. Since docker stands on the giant shoulders of LXC, production release 1.0 of LXC might result in following changes in dockers.

• Its most likely docker will incorporate all the changes from upstream LXC 1.0 in coming months andbe production ready. Docker version 1.0 is just round the corner.
• Better security for docker containers - Support for fully unprivileged containers will help docker to offer better isolation of running containers and secure resource access from running containers to the host and other containers.
• API bindings for GO will help Docker to implement LXC driver code using the API, instead of using userspace command line tools provided by LXC (currently Docker uses tools like lxc-start, lxc-attach instead fo relying on API’s provided by LXC).
• Flexible backingstore systems support will help docker in supporting btrfs, zfs, lvm by leveraging LXC code directly instead of separate implementation. In docker 0.8 btrfs support is experimental and zfs support is in development, with LXC 1.0 docker will be able to support btrfs and zfs in next release.
• Reduction of command set will help reduce complexity of docker.
• Centrally generated images will help implementing better docker registry and index.

All of the above are just my personal views on the evolution of docker based on LXC 1.0 and does not represent the views and roadmap of docker or LXC projects. If you need more details on future development of docker please check the projects blog post Getting to Docker 1.0 [3]. I wrote this just out of my personal interest in docker and LXC.

Footnotes

Libvirt Goals & Architecture

Libvirt defines following terms for its goals [2]:

• Node is a single physical machine.
• Hypervisor is a layer of softeare allowing to virtualize a node in a set of physical machines with possible different configurations that the node itself.
• Domain is an instance of an operating system (or subsystem in case of container virtualization like OpenVZ and lxc) running on a virtualized machine provided by the hypervisor.

Based on above terms “The goal of libvirt is to provide a common and stable layer sufficient to securely manage domains on a node, possibly remote”. So libvirt should provide all APIs needed to do the management, such as: provision, create, modify, monitor, control, migrate and stop the domains - within the limits of the support of the hypervisor for those operations. This implies following sub-goals:

• All API can be carried remotely though secure APIs
• While most API will be generic in term of hypervisor or Host OS, some API may be targeted to a single virtualization environment as long as the semantic for the operations from a domain management perspective is clear
• the API should allow to do efficiently and cleanly all the operations needed to manage domains on a node, including resource provisioning and setup
• the API will not try to provide high level virtualization policies or multi-nodes management features like load balancing, but the API should be sufficient so they can be implemented on top of libvirt
• stability of the API is a big concern, libvirt should isolate applications from the frequent changes expected at the lower level of the virtualization framework
• the node being managed may be on a different physical machine than the management program using libvirt, to this effect libvirt supports remote access, but should only do so by using secure protocols.
• libvirt will provide APIs to enumerate, monitor and use the resources available on the managed node, including CPUs, memory, storage, networking, and NUMA partitions.

So libvirt is intended to be a building block for higher level management tools and for applications focusing on virtualization of a single node (the only exception being domain migration between node capabilities which involves more than one node).

Libvirt Driver Based Architecture

Libvirt to support wide variety of hypervisor implements a driver-based architecture. Based on car analogy, it means delegating the actual implementation of control of different cars to the drivers specifically designed for make, model and engine of the specific car. Libvirt currently supports:

Hypervisor

• LXC - Linux Containers
• OpenVZ
• QEMU
• Test - Used for testing
• UML - User Mode Linux
• VirtualBox
• VMware ESX
• VMware Workstation/Player
• Xen
• Microsoft Hyper-V
• IBM PowerVM (phyp)
• Parallels
• Remote - Accessing libvirt on remote node through libvirtd (libvirt daemon)

Storage

• Directory backend
• Local filesystem backend
• Network filesystem backend
• Logical Volume Manager (LVM) backend
• Disk backend
• iSCSI backend
• SCSI backend
• Multipath backend
• RBD (RADOS Block Device) backend
• Sheepdog backend

Virtual Networks

• Bridging
• NAT
• VEPA (Virtual Ethernet Port Aggregator)
• VN-LINK

Libvirt API structure [3]

The figure above shows the five main objects exported by the API:

virConnectPtr

Represents the connection to a hypervisor. Use one of the virConnectOpen functions to obtain connection to the hypervisor which is then used as a parameter to other connection API’s.

virDomainPtr

Represents one domain either active or defined (i.e. existing as permanent config file and storage but not currently running on that node). The function virConnectListAllDomains lists all the domains for the hypervisor.

virNetworkPtr

Represents one network either active or defined (i.e. existing as permanent config file and storage but not currently activated). The function virConnectListAllNetworks lists all the virtualization networks for the hypervisor.

virStorageVolPtr

Represents one storage volume generally used as a block device available to one of the domains. The function virStorageVolLookupByPath finds the storage volume object based on its path on the node.

virStoragePoolPtr

Represents a storage pool, which is a logical area used to allocate and store storage volumes. The function virConnectListAllStoragePools lists all of the virtualization storage pools on the hypervisor. The function virStoragePoolLookupByVolume finds the storage pool containing a given storage volume.

These names follow C conventions, but developers of cloud computing platforms and applications do not need to use C directly, there are language bindings available for major languages. Currently libvirt API language bindings [4] are available for C#, Java, OCaml, Perl, PHP, Python, Ruby.

Domain Management Architecture

There are two distinct means for domain management using libvirt API.

1. Single node domain management

As illustrated in the figure above in this mode applications (cloud management platform i.e. CMP applications) and domains exist on the same node. In this scenario applications directly works through the libvirt api on the host operating system (os) to control and manage the local domains.

2. Multi node domain management

As shown in the figure above, applications (CMP applications) using libvirt API and the domains to manage or control are on separate nodes. In this mode a special domain called libvirtd (libvirt daemon) needs to run on remote nodes. The management application nodes use the nodes underlying network communicattion to communicate with remote libvirtd through the local libvirt using custom protocol. Actually libvirt uses Remote [6] driver for communicating with remote node and remote API calls are handled synchronously. Remote driver for libvirt supports a range of transports like:

tls

TLS 1.0 (SSL 3.1) authenticated and encrypted TCP/IP socket, usually listening on a public port number. To use this you will need to generate client and server certificates. The standard port is 16514. This is the default transport, if no other is specified.

unix

nix domain socket. Since this is only accessible on the local machine, it is not encrypted, and uses Unix permissions or SELinux for authentication. The standard socket names are /var/run/libvirt/libvirt-sock and /var/run/libvirt/libvirt-sock-ro (the latter for read-only connections).

ssh

Transported over an ordinary ssh (secure shell) connection. Requires Netcat (nc) installed and libvirtd should be running on the remote machine. You should use some sort of ssh key management (eg. ssh-agent) otherwise programs which use this transport will stop to ask for a password.

ext

Any external program which can make a connection to the remote machine by means outside the scope of libvirt.

tcp

nencrypted TCP/IP socket. Not recommended for production use, this is normally disabled, but an administrator can enable it for testing or use over a trusted network. The standard port is 16509.

libssh2

Transport over the SSH protocol using libssh2 instead of the OpenSSH binary. This transport uses the libvirt authentication callback for all ssh authentication calls and therefore supports keyboard-interactive authentication even with graphical management applications. As with the classic ssh transport netcat is required on the remote side.

Libvirt Project [5]

According to statistics on ohloh libvirt in a nutshell:

• 15,188 commits made by 331 contributors representing 481,506 lines of code
• Mostly written in C

• Established, mature codebase maintained by a very large development team with increasing year on year commits.
• Estimated 128 years of efforts (COCOMO model)

Conclusion

Libvirt is one very important library on whose giant shoulders cloud computing services and platforms like Amazon AWS, Google Compute Engine, OpenStack, CloudStack, Eucalyptus and numberous others are standing. Also this API enables developers and companies to build new and innovative cloud computing services or platforms and build awesome applications or services on top of it. Libvirt started in 2005 and with growing popularity of cloud computing, this project will continue to grow. But in most of the conferences, talks and papers related to Cloud Computing I did not find much coverage of libvirt, so while researching Cloud Management platform thought of writing and article on it. Kudos to all the libvirt code contributors for building a beautiful abstraction layer and making life easier for cloud computing services and platform developers. In spite of not getting as much press and coverage as mainstream cloud computing platforms like OpenStack, CloudStack or Eucalyptus the growth of the project, community and code commits are heartening.

Footnotes

December 4 ^th 2013 - Liberalized Financial Measures

According to the head of People’s Bank of China (PBOC) [3], central banking regulator in China, Shanghai will implement liberalized financial measures in Free Trade Zone in a year and will support openness. PBOC outlined a set of 30 measures for Shanghai FTZ covering:

• Cross-border investment and transactions.
• Expanding yuan cross-border usage.
• Promoting interest rate liberalization.
• Simplyfying foreign exchange management.
• Allow free fund transfer between free trade zone and offshore bank accounts.
• Qualified foreigners working in Shanghai FTZ to trade domestic equities.
• Enable qualified residents in the Shanghai FTZ to invest in the foreign securities market.
• Allow foreign companies with branches in SHanghai FTZ to issue yuan bonds onshore.

November 18 ^th 2013 - Procedures and Future Plan for Shanghai FTZ

During November various departments and government authorities in China unveiled their respective plans for Shanghai FTZ. Below are some of the highlights of those plans:

November 13 ^th 2013 - Foreign Owned Foreign Medical Institutions in Shanghai FTZ

Shanghai Government released a circular on 13 ^th November 2013 Provisional Measures on Administration of Wholly Owned Foreign Medical Institution in Shanghai Free Trade Zone, China (沪府办发〔2013〕63号) [2]. Some of the basic provisions in the circular are:

October 15 ^th 2013 - Import Tax Policies for Shanghai FTZ

China’s ministry of Finance, The General Administration of Customs, and the State Administration of Taxation jointly release the Circular on Relevant Import Tax Policies for Shanghai Free Trade Zone [4]. Following are some of the highlights of the measures:

• Domestic leasing companies in Shanghai FTZ and their subsudiary are entitled to 5% import value-added tax (VAT) rate for purchase of overseas aircraft after approval from relevant administrative departments. The inert weight of such aircraft should be more than 25 tons and are to be leased to domestic airlines.
• Goods produced and processed by company established in Shanghai FTZ and sold to the domestic market via the gate between Shanghai FTZ and other areas are still subject to import VAT and consumption tax. Also at the request of the enterprise, a pilot tariff policies will be implemented for goods sold to domestic market based on their corresponding imported materials or actual declaration status.
• The import of machines, equipment and other goods by manufacturing companies and manufacturing service companies in Shanghai FTZ are entitled to tax exemption.
• Establishment of bonded display trading platform is allowd in Shanghai FTZ.
• This circulars provisions are not applicable for goods imported by consumer service companies and for other goods that are not entitled for tax exemptions as per laws, administrative regulations and relevant rules.

However its not yet clear what will be the tax benefits of establishing an enterprise in Shanghai FTZ. Based on the information available there are no special tax incentive for companies in Shanghai FTZ.

Conclusion

So far this are the details available on Shanghai FTZ. If there are somethings missed or need to be updated please let me know, will update it accordingly. Hopefully this is useful for people looking for consolidated information on Shanghai FTZ so far.

Footnotes

Cloud Computing Definition in China

There is no generally accepted definition in China and among multile version the most prominent version according to The Ministry of Industry and Information Technology (MIIT) [2] is:

Cloud computing is a method for achieving large-scale computing information processing, which unifies, organizes, and flexibly draws upon various Information and Communication Technology (ICT) information resources through the Internet. Cloud computing utilizes distributed computing and virtual resource management technologies, among others. Using the Internet, it takes spread-out ICT resources (including computing, storage, application platforms, and software, among others) and brings them together to form a shared resource pool. Furthermore, it uses dynamic, on-demand, and scalable methods to provide services to users. Users can use various types of terminals (such as personal computers (PCs), tablet computers, smart phones, even smart televisions, among others) to access ICT resource services through the Internet.

If you read through the definition, it mostly matches the generally accepted definition except “on-demand self service”. This part is left out due to requirement of ICP (Internet Content Provider) license in China. Since ICP license is linked to computing resources like domain name, IP addresses (external accessible through internet); it makes it challenging if not impossible to provide elasticity, auto-scaling, movement of computing resources based on application and user demands (since it needs some human interaction and not an automated process).

This is not the only definition in China and there are some other definitions as well. Some of them are same as NIST like this one from Advisory Committee for State Informatization - ACSI (国家信息化专家咨询委员会) which defines: “Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.”, in one of their presentation.

ICP License [13]

ICP license application process and approval is manual multi-step process involving regulatory authorities and is described briefly here. This license is required by both local foreign enterprises in China for Internet based application and services.

• You need to create an account at Ministry of Industry & Information Technology by entering your e-mail address and cell phone number .
• After the approval by the MII, you will receive a confirmation code on your cell phone and e-mail account.
• After receiving the code you have to login into the MII website, enter the code, and access your account.
• After login into the MII Web site need to fill an on-line form with name, mobile phone number, home phone number, home address, domain name, host provider, company name, and description of website.

Usually this process can take from 20 days to 90 days and if you change the cloud provider you need to go throught the process again for approval.

China Cloud Computing Market Analysis

In China like other countries its difficult to find reliable market report for Cloud Computing Services. Also given the way its defined many reports add server virtualization market into Cloud Computing (which inflates the figure and give a wrong impression of the Cloud Computing Market in China). Still based on some estimates following are some of the findings for the China’s Cloud Computing market.

In 2012 the market size for Cloud Computing in China is approximately 3.5 billion yuan (US$ 573 million) which compose of:

• Software as a Service (SaaS) - 2800 million yuan (US$ 459)
• Infrastructure as a Service (IaaS) - 511 million yuan (US$ 83.77 million)
• Platform as a Service (PaaS) - 184 million yuan (US$ 30.16 million)

Among all the cloud providers SaaS only covers 16% and remaining 84% are primarily IaaS providers with few PaaS providers. This means majority of the revenue for Cloud Computing comes from SaaS providers, which is inline with the market reality, since majority of the revenue is generated by companies like Alibaba with its marketplace (Tmall and Taobao), Tencent with its Weixin (WeChat) platform and Sina with its Weibo platform. Some might question that Taobao, Tencent and Sina generate much larger revenue, but if they look closely at revenue the portion of revenue from pure SaaS service is very small and majority comes from other means like advertising, payment and games.

Since Cloud Computing depends on the underlying data center, the above figure shows market size and trends for internet data center in China. Since Internet Data Center license is prohibited for local and foreign enterprises since 2008 in China. the growth has been slow. Please take above figures for reference only, like Cloud Computing, I cannot verify the validity of figures from reliable data sources. It seems most of this figures are derived based on a limited survey.

There are few counts and reports available from Netcraft, but I did not include it in this article since they do not accurately depict the market and tend to skew towards large telecom and internet companies in China, which does not reflect the normal Cloud Computing market.

Regulatory Framework and factors affecting Cloud Computing in China

After detailed analysis of China Cloud Computing market, the factors inhibiting its growth can be summarized with one word Regulations. In China there is no specific regulation or license which directly address Cloud Computing in China. Since Cloud Computing uses Internet network to access services, it is regulated based on the China Telecommmunication Regulations [3] and ususally classified as Value Added Telecommunication Services (VATS). So it creates a high regulatory barrier to entry even for local innovative companies and startups (since for foreign companies if not prohibited mostly limited to non-controlling stakes). So in my view in China Cloud Computing is not hampered by technology related issues, but more due to regulatory issues. During my discussion and talks within local technology companies and startups I find many are technically at par with leading cloud computing companies globally but cannot execute it locally due to high barrier to entry set by current regulatory framework.

Following are some of the key regulatory framework covering Cloud Computing in China:

• Cloud Computing is governed by Circular of the Ministry of Information Industry on the Readjustment of the Classification Catalogue of Telecommunication Services (信息产业部关于重新调整-电信业务分类目录-的通告) - 2003 Classification Catalogue [4] and usually classified as VATS.
• Cloud Computing Service providers which needs to provision Internet Data Center (IDC) services or built an IDC to deliver Cloud Computing Services needs an IDC Service license. IDC license are currently not granted to either local or foreign enterprises since 2008. Recently IDC license are accepted from local enterprises in 2013. Also foreign enterprises in China cannot get IDC license and normally use Variable Interest Entity (VIE) structure or partnership with local firms.
• The services which do not fall within 2003 Classification Catalogue will only need to file and record case with the provincial counterparts of the MIIT before commercially operating the services. But in practice at present, no specific procedures are in place to fulfill such record-filing requirements and information about successful cases of record-filing are not available publicly.

• Foreign company must establish a Foreign Invested Telecommunication Enterprise (FITE) in China to provide Cloud Computing Services to local companies and consumers. They need to follow:

  • Catalogue of Industries for Guiding Foreign Investment (外商投资产业指导目录) [5]
  • Telecommunication Regulations, the Provisions on Administration of Foreign-Invested Telecommunications Enterprises (外商投资电信企业管理规定) [6]
  • Portion of foreign investment in Basic Telecom Service (BTS) (applicable if providing IDC services) should not exceed 49%. This means FITE needs to form a joint venture with local Chinese entity having the majority controlling stake.
  • Portion of foreign investment in VATS type FITE should not exceed 50%, which means a joint venture with equal stack for FITE and local Chinese entity.

In most cases foreign enterprise will most likely provide Cloud Computing services under VATS, the followin additional requirements needs to be considered (these are same for local Chinese entity applying for VATS license):

• The proportion of capital invested by the foreign company shall not exceed a total of 50%.
• The registered capital should be either RMB 10 million or more for nationwide service in China.
• The foreign investor shall have a record of good performance and operating experience in managing value-added telecommunications business.
• FITE employs minimum of 10 employees at the time of application.

Obviously these regulationa creates a big barrier for most of the startups or small companies to enter Cloud Computing Services market and make it extremely hard for large enterprises to enter. Moreover based on information available only 28 FITE’s managed to get a license since 2003 in China. So based on this information majority of the Chinese small and medium sized Cloud Computing Service providers and most of the foreign Cloud Computing Services Providers operate under grey zone or without license in the hope regulations will evolve.

Future of Cloud Computing in China

China recognize the importance of Cloud Computing for Information Technology and identified it as key strategic industry for technological progress of China. According to some industry analysts China’s Cloud Computing industry value chain will be between 750 billion yuan to 1 trillion yuan (US$ 123 billion to US$ 164 billion). China’s central state council and various ministries provided support for development and investment in Cloud Computing technology through following major planning documents:

1. China’s National Medium and Long-Term Plan for Science and Technology Development [7]
2. China’s 12 ^th Five Year Plan (FYP) for Economic and Social Development [8]
3. Ministry of Science and Technology 12 ^th FYP [9]
4. Software and Information Technology Service Industry Development Plan for the 12 ^th FYP (软件和信息技术服务业“十二五”发展规划) [10]
5. Support Project for Next-Generation Information Technology Application Security (新一代信息技术应用安全支撑工程) [11]

Besides government planning document and investment, China is also working on revising regulatory framework to make it inline with market needs and create an environment necessary for proliferation of Cloud Computing technology and services.

• Amendments to 2003 Classification Catalogue

  MIIT circulated an updated version 2013 Draft Classification Catalogue to solicit public comments on 25 ^th May 2013. This new catalogue did not add Cloud Computing Service category, but introduced a new category named Internet Resources Collaboration Services which refers to the use of equipment and resources constructed on data centers, and through the internet or other networks, to provide customers with services, including, data storage, development environment for internet applications, deployment of internet applications and operation management to users by way of easily accessible, use on-demand, easily expanded and/or collaborative sharing which covers the Cloud Computing Services.

• Above amendment will enable foreign entities to offer Cloud Computing Services in China.
• MIIT lifted ban on IDC by a notice [12] on 30 ^th November 2012 and recently granted license to Huawei and Alibaba.

Also due to inherent requirements for serving over 500 million internet users in China, most of the internet companies, enterprises and startups are embracing and developing cloud computing technology for internal use.

Conclusion

In essence Cloud Computing Services in China despite all the hurdles is poised to take off in a big way. Since it is fundamental for serving millions of Internet users cost effectively and flexibly. Also since the regulatory framework is evolving I expect more Cloud Computing service providers in market with innovative and mature compute, storage and network related Cloud Computing services at attractive prices. Also more foreign players will be available in China, like Microsoft is doing a beta test with its Azure services in China. In this article I intentionally left the current Cloud Computing Service providers in China since many of them are still operating in grey zone or enjoying a monopoly due to regulations and not due to innovation in technology or Cloud Computing service offerings. Moreover I only have first hand experience of using Cloud Computing Services from two large players which is not representative of the overall market. So I will do bit more research and work with local Cloud Computing providers and cover them in next article. If you are a Cloud Computing Service provider in China and would like to be included in my next article, please send me mail with the details of your Cloud Computing Services. Hopefully information in this article is useful for some people, and if you notice some issues please notify me for corrections.

Footnotes

1) Lack of Clarity about what OpenStack does and does not

In this point the researcher contends that OpenStack is positioned as an alternative to Commercial Cloud Management Platform (CMP), as they are called in Gartner. This is indeed untrue, if you look at the figure above and read the definition of OpenStack [2], it gives you a clear idea of What is OpenStack. In the same paragraph Gartner researcher contends it needs to be augmented by a commercial CMP for being useful in enterprise private cloud and here I smell a rat.

Indeed instead of taking a narrow CMP view and definition of Gartner, I use a more neutral cloud computing definition from NIST and try to explain how OpenStack platform can help enable true private cloud with or without commercial CMP. This will directly address the first point regarding what OpenStack does and doesn’t.

2) Lack of transparency about the business model around OpenStack.

OpenStack is a very young project and business models around it are building and some of them will bear fruits in time to come. This does not mean its not ready for enterprises today, as I covered in earlier section OpenStack is ready for enterprise and vendors can sell it today. Indeed in OpenStack Summit there was a video presentation from Redhat Red Hat Keynote - Truth Happens (For China users Red Hat Keynote - Truth Happens - China), that succintly can answer this concern. Readers can find more success stories at Open Source Stories.

Also one thing which every vendor needs to understand is that private cloud implementation is not simply install, configure and manage a piece of software. Its much more and requires proper understanding of Cloud Computing and how it will generate business value for enterprises. Since cloud computing is so new it will take some time for vendors to digest and understand, as it is disrupting their traditional models with which they worked for decades. Being part of the open source eco-systems and projects within OpenStack, it helps them to prepare for demonstrating business value and at the same time improving OpenStack to deliver that business value. Vendor participation in OpenStack is not philonthropy, every vendor participating in the process is not just contributing, but learning, preparing, training and improving their own team to be ready for next disruption and in directly helping the overall cloud computing eco-system. This investment is far less then the money they will spend on consulting and market research firm to tell them what to do, without any tangible outcome. Participating in open source cloud computing platform helps them with tangible outcomes, not just with a piece of powerpoint, pdf or a report.

As for low adoption rate and few deployments, it can directly be attributed to lack of knowledge and awareness about Cloud Computing within IT organizations of traditional enterprises. I will be very happy if market research and analysis firm (like Gartner, Forrester) can take lead in this area and promote awareness about Cloud Computing within IT organization of traditional enteprprises. Hopefully they can steer them away from narrow cloud computing definitions and visions promoted by Commercial CMP’s. I am sure once the IT Organization understand the basics of Cloud Computing, they will adopt and launch multi-million dollar projects with OpenStack or its other siblings for implementing private clouds and integrate them with public clouds in ways unseen before, its just a matter of time.

Cloud Computing Definition [1]

After evaluating many definition, I settled on one from National Institute of Standards and Technology [1] (hereafter called NIST), which in my view is the most complete and logical definition of cloud computing.

Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. This cloud model is composed of five essential characteristics, three service models, and four deployment models.

Essential Characteristics

1. On-Demand self service

   A consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with each service provider.

2. Broad network access

   Capabilities are available over the network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g., mobile phones, tablets, laptops, and workstations).

3. Resource pooling

   The provider’s computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand. There is a sense of location independence in that the customer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of abstraction (e.g., country, state, or datacenter). Examples of resources include storage, processing, memory, and network bandwidth.

4. Rapid elasticity

   Capabilities can be elastically provisioned and released, in some cases automatically, to scale rapidly outward and inward commensurate with demand. To the consumer, the capabilities available for provisioning often appear to be unlimited and can be appropriated in any quantity at any time.

5. Measured service

   Cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, active user accounts). Resource usage can be monitored, controlled, audited, and reported, providing transparency for both the provider and consumer of the utilized service.

So in essence if any company calls its products, solutions and service as cloud computing it should be able to fulfill the above five criterias.

Service Models

1. Software as a Service (SaaS)

   The capability provided to the consumer is to use the provider’s applications running on a cloud infrastructure. The applications are accessible from various client devices through either a thin client interface, such as a web browser (e.g., web-based email) or a program interface. The consumer does not manage or control the underlying cloud infrastructure including network, servers, operating systems, storage, or even individual application capabilities, with the possible exception of limited user- specific application configuration settings.

2. Platform as a Service (PaaS)

   The capability provided to the consumer is to deploy onto the cloud infrastructure consumer-created or acquired applications created using programming languages, libraries, services, and tools supported by the provider.12 The consumer does not manage or control the underlying cloud infrastructure including network, servers, operating systems, or storage, but has control over the deployed applications and possibly configuration settings for the application-hosting environment.

3. Infrastructure as a Service (IaaS)

   The capability provided to the consumer is to provision processing, storage, networks, and other fundamental computing resources where the consumer is able to deploy and run arbitrary software, which can include operating systems and applications. The consumer does not manage or control the underlying cloud infrastructure but has control over operating systems, storage, and deployed applications, and possibly limited control of select networking components (e.g., host firewalls).

So Cloud Computing can be delivered in any one or combination of these models, the only pre-requisite is to have the five essential characteristics explained earlier.

Deployment Models

1. Community cloud

   The cloud infrastructure is provisioned for exclusive use by a specific community of consumers from organizations that have shared concerns (e.g., mission, security requirements, policy, and compliance considerations). It may be owned, managed, and operated by one or more of the organizations in the community, a third party, or some combination of them, and it may exist on or off premises.

2. Public cloud

   The cloud infrastructure is provisioned for open use by the general public. It may be owned, managed, and operated by a business, academic, or government organization, or some combination of them. It exists on the premises of the cloud provider.

3. Private cloud

   The cloud infrastructure is provisioned for exclusive use by a single organization comprising multiple consumers (e.g., business units). It may be owned, managed, and operated by the organization, a third party, or some combination of them, and it may exist on or off premises.

4. Hybrid cloud

   The cloud infrastructure is a composition of two or more distinct cloud infrastructures (private, community, or public) that remain unique entities, but are bound together by standardized or proprietary technology that enables data and application portability (e.g., cloud bursting for load balancing between clouds).

Again the litmus test for calling any of the deployments as Cloud Computing are the five essential characteristics in earlier section.

Cloud Computing Reference Architecture [2]

Cloud Computing definition provides a clear understanding of what it is and to further clarify details NIST also provide a referene architecture. This reference architecture is a generic high-level conceptual model for discussing the requirements, structures, and operations of cloud computing. The model is not tied to any specific vendor products, services, or reference implementation, nor does it define prescriptive solutions that inhibit innovation. It defines a set of actors, activities, and functions that can be used in the process of developing cloud computing architectures, and relates to a companion cloud computing taxonomy. It contains a set of views and descriptions that are the basis for discussing the characteristics, uses, and standards for cloud computing.

Above cloud computing reference architecture defines five major actors: cloud consumer, cloud provider, cloud auditor, cloud broker and cloud carrier where each actor is an entity (a person or an organization that participate in a transaction or process and/or performs tasks in cloud computing. It also includes architectural elements of cloud computing.

Conclusion

So in essence Cloud Computing refers to the delivery of scalable IT resources over Internet, such as storage, compute, networking with security and privacy. Cloud computing enables end users to “rent” and remotely access resources from cloud providers on a pay-per-use basis, resulting in both improved system scalability and cost savings. This transition to the cloud may be self-reinforcing; as businesses discover cost benefits in moving some or their entire IT infrastructure to cloud platforms, economies of scale should allow cloud computing prices to decline further still, which in turn will attract new providers and cloud computing solutions into the market. Beyond the financial incentive to use cloud services to address current IT needs, the availability of low-cost distributed High Performance Computing (HPC) may open up new avenues of business, enabling companies that would not be able to afford traditional physical supercomputing infrastructure to tap into those services whenever necessary. Also cloud computing by automating many tasks and reducing costs can bring efficiency in building scalable applications, services and products which will enable Startups to disrupt the traditional business models.

This article serves as foundation for my next article for covering next topic from the talk on Current State and Future of Cloud Computing in China.

Footnotes

Administrative Measures for Shanghai Free Trade Zone (FTZ)

Shanghai Government has release administrative measures for pilot FTZ in the following areas:

1. Investment Management
2. Trade Development and Facilitation
3. Financial Innovation and Risk Management
4. Supporting Management and Services

Following are the details of the Administrative Measures

Conclusion

If Shanghai Government can implement reforms and policies in Shanghai FTZ, it will create substantial financial, tax, supply chain and operational incentives for companies to do business. Also it will help propel Shanghai as one of the leading business centers in the world not just for China, but for Asia Pacific. Only time will tell the difference, currently both investors and government are treading carefully and adopted a wait and watch before act plan.

Footnotes

Draft Regulation [1] [9]

Shanghai Pilot Free Trade Zone draft regulation was published on 29th September 2013 with following key provisions:

Negative List [2]

Shanghai FTZ is departure from the earlier mechanism where foreign investors can only invest according to the sectors open for investment as defined by MOFCOM [3]. Shanghai FTZ will follow a negative list approach where investors can invest in any sectors not covered in this list. This means all the sectors except in negative list are open for foreign investment in Shanghai FTZ. The list covers 1,069 sectors across 18 broad categories of industry and contains 190 special regulatory measures. Some of the sectors are highlighted below:

1. Agriculture, Forestry, Animal Husbandry and Fishery Industries
2. Mining Industry
3. Manufacturing Industry
4. Production and Supply Industries for Power, Gas and Water
5. Construction Industry
6. Wholesale and Retail Industries
7. Transportation, Warehousing and Postal Service Industries
8. Information Transmission, Computer Service and Software Industries
9. Finance Industry
10. Real Estate Industry
11. Leasing and Commercial Service Industries
12. Scientific Research and Technical Service Industries
13. Water Conservancy,
14. Environmental, and Public Facilities Management Industries
15. Education Industry Health and Social Industries
16. Cultural, Sports and Entertainment Industries

Following provisions make it bit difficult for technology Startups in Shanghai FTZ compared to Hong Kong, Singapore are:

1. Investment is prohibited in Internet Data Center Services Industry.
2. Prohibited Online Game Services
3. Foreign Investment in eCommerce shall not exceed 55%
4. Foreign Investment in Data Processing and Online Transaction processing shall not exceed 50%.
5. Foreign Investment in IP-VPN services shall not exceed 50%.
6. Foreign Investment in Mobile App Store shall not exceed 50%

Shanghai FTZ in Business

Shanghai FTZ opened enterprise registration on Tuesday 9 ^th October 2013. Following are the updates so far:

• On the opening day of registration in Shanghai FTZ 577 people applied for business registration and 1480 people ask for advice and additional information at the service lobby. [4]
• HSBC and Bank of East Asia (BEA) received regulatory approval to open a sub-branch in Shanghai’s pilot free trade zone, joining Citigroup and DBS Bank who already got the approval. [5]
• buyeasi.com, operated by payment service provider Easipay will launch e-commerce platform in the Shanghai FTZ to allow consumers from China’s mainland to buy foreign brands at lower prices. [6]
• Since the opening 600 people visit Shanghai FTZ service desk daily with almost 2000 people seeking advice and information every day. [7]
• Shanghai FTZ mau allow foreign companies to raie capital by selling shares and try to tap the world’s fourth largest stock market. [8]

Special Provision for Company Registration in FTZ

Shanghai FTZ offers many reforms and one of the key reforms is related to Company Regitrations. Based on opinions published for Shanghai Free Trade Zone following reforms will be tested:

Brief Overview of CloudStack

Apache CloudStack is a top-level project of the Apache Software Foundation (ASF). The project develops open source software for deploying public and private Infrastructure-as-a-Service (IaaS) clouds. CloudStack provides an open and flexible cloud orchestration platform to deliver reliable and scalable private and public clouds.

Release 4.2.0 New Notable Features [1]

The 4.2.0 release of CloudStack, the second major release from the Apache CloudStack project since its graduation from the Apache Incubator. CloudStack 4.2 includes more than 70 new features and enhancements. The focus of the release is on three major areas:

• Improved support for both legacy-style and cloud-style workloads
• New third-party plug-in architecture
• Networking enhancements

Besides this the other 2 areas of enhancement are:

• Host and Virtual Machine Enhancements
• Monitoring, Maintenance, and Operations Enhancements

For me personally the major enhancements which stands out are in following specific areas and I will also point out one area where its not being enhanced and hopefully someone will work on them in future release soon:

Interesting Enhancement Not Done or Not in Future Roadmap

• Support for LXC (Linux Containers), OpenVZ and Solaris containers I was hoping 4.2.0 will contain some significant enhancement in virtual containers and integration. I was specially interested in LXC containers and integrating Docker with CloudStack. But since this was not a priority no major enhancement happened in this area. Hopefully if one day I have a big enough budget or financially capable will sponsor the development of those features or someone will have similar needs as me and work on it.
• Integration with Ubuntu Juju and MAAS It will be interesting if Ubuntu Juju and MAAS can be integrated seemlessly with CloudStack.
• Integration with The Foreman Hopefully someone can integrate The Foreman with CloudStack via plugin like its done in oVirt
• Integration with Shinken Monitoring Shinken Monitoring is a modern replacement of Nagios with scalability in mind. If someone can integrate it with Admin, Domain and Cloud User portal of CloudStack via plugins will be a great addition and besides simple monitoring all the nagios plugins can be used on hosts and clients to monitor. This is on my personal wishlist of enhancement for CloudStack.
• Some simple billing plugin Currently there is no opensource simple metering and billing plugin for CloudStack using cloud_usage data. Since I wanted to test the cloud-usage for CloudStack I have to write my own scripts. If someone can provide an opensource plugin which can integrate with CloudPortal will be great and make this platform very enticing for companies who wants to meter the usage and provide billing to internal departments or for cloud providers to provide simple billing to its clients.

This is my personal coverage of CloudStack release 4.2.0. Obviously you can check the full Release Notes for more details and enhancements and bug fixes interesting for you. Also if you haven’t tried or heard of CloudStack then now is the time to give it a try.

Footnotes

Restriction on Information Transmission, Computer Service and Software Industries

1. Foreign Investment in eCommerce shall not exceed 55%

   This is a very good news indeed compared to current regulation outside the Shanghai FTZ. Since outside FTZ its almost impossible for foreign firms to get ICP License (category 2) and start a legal eCommerce business in China, unless they go through a join venture or via VIE route or manufacturer of products for which they hold a retail license. Its also a very good news for local enterpreneurs since they can start eCommerce business from Shaghai FTZ without lengthy approval process and company formations. Also they can tap angel, private and venture capital investments.

2. Foreign Investment in Data Processing and Online Transaction processing shall not exceed 50%.

   This is also a very good news since currently online transaction processing is completely off-limits to foriegn investment outside FTZ. Also this means for local enterpreneurs it has similar benefits as above.

3. Foreign Investment in IP-VPN services shall not exceed 50%

   This is a good news for StartUps who target China for launching innovative IP-VPN services based on TCP/IP protocol. Since outside Shanghai FTZ its off-limits for foreign investment.

4. Foreign Investment in Mobile App Store shall not exceed 50%

   This is a good news as well for StartUps who wants to build a service similar to google play in China for Mobile and Table software applications. Outside FTZ its prohibited for foreign investment.

5. Investment is prohibited in Internet Data Center Services Industry.

   This is really bad news for Startups. This effectively means companies and startups who work in areas which requires large Internet Data Centers still needs to seek approval from the government. This is similar to existing regulation outside the FTZ. This includes web hosting, server rentals, telecommunication, ip network and other basic computing services necessary to build the basic foundation for innovative software and technology Startups. This is off-limits for foreign investment and for local requires a very lengthy approval process, which requires significant energy, upfront capital; unless you bypass or use guanxi or some clever arrangement. This effectively means Startups are at the mercy of the large telecom companies and need to find creative ways for a work around and waste energy rather then spending it on building something innovative helping society. Unless the target market is China, it doesn’t make any sense to build a StartUps in this area unless you are local enterpreneur.

6. Prohibited Online Game Services

   This is a bad news for StartUps who wants to build next Zygna or PopCap in China. So if you plan on doing anything with gaming there is no incentive in Shanghai FTZ, you are better off building it in Software Technology Parks which might offer you better incentives.

Conclusion

Although the negative list is rather long, its a mixed bag of good and bad news for Startups to setup in Shanghai FTZ. But definitely Shanghai FTZ is still the best location for StartUps in China, if the target market is China. Also it will take time for the state council and various ministry to clarify and fine tune the policies for FTZ. As it is said “Well begun is half done”, I still believe State Council under Li KeQiang took the right steps for the transition of China’s economy by launching a pilot free trade zone in Shanghai. So the work is already half done and we need to wait for few more years to see the next half. If some of the readers who stumble across this article have some opinion to share, make suggestions or improvement please let me know. Hope someone find this information useful.

Footnotes:

Shanghai Free Trade Zone (FTZ) - First Regulatory Draft [1]

Shanghai Government with support from China Central Government unveiled the first draft of regulation governing the Free Trade Zone (hereafter called FTZ) on 27th September 2013. This FTZ is considered as second biggest change in China after the opening up of Chinese economy by Deng Xiao Ping in 1990’s by establishing Shenzhen Special Economic Zone. It marks the trasition of Chinese economy from world’s factory into a developed economy driven by consumptions and services with free market economy. This will serve as a blueprint for Chinese government for transition of China to free market economy. Only time will tell how its going to shape China and its development, but definitely by bypassing the local ministry and their entrenched interest the Central Government of China has indicated its willingness to reform and change the course of Chinese economy.

Personally I have been waiting for this reforms eagerly, since some of its regulations takes away the barriers in China for doing Startups especially in information technology and internet area. If Shanghai Government can target innovative startups to come and setup in FTZ, it might as well become similar to silicon valley in US, a Mecca of startups in China. This will obviously needs some minor reforms on the China immigration, but at least for local startups already in China, these changes might bring them closer to the environment offered in South Korea, Hong Kong, Singapore. This offers 2 main advantages to early startups:

1. Instead of setting up shops in Hong Kong, Singapore for China, its better to come inside the second largest economy providing large user base.
2. Significant cost advantages with more preferential policies they might get in Hong Kong or Singapore.

Following are some of the key areas which will help startups in FTZ:

1. No approval required to start a business, just registration is enough. Takes away the burden of going through approval process. Its a key for young startups where founders are more interested in implementing innovative ideas instead of going through regulatory hoops.

   Starting from October 1, 2013, 11 administrative examination and approval items regulated under the following three laws will be suspended in the Shanghai FTZ on a three-year trial basis.

   • The Law of the People’s Republic of China on Wholly Foreign-Owned Enterprises.
   • The Law of the People’s Republic of China on Sino-Foreign Equity Joint Ventures.
   • The Law of the People’s Republic of China on Sino-Foreign Cooperative Joint Ventures.

   In addition, in order to better promote investment into the Shanghai FTZ, the Plan also allows enterprises or individual shareholders registered in the Shanghai FTZ to pay income tax in installments within a five-year period for value-added assets arising from asset restructuring.

2. Capital market reforms and free yuan conversion. This brings FTZ closer to developed economy like Hong Kong and Singapore where angel, private investors and venture capital firms can easily inject money in promising startups with legal protection. They do not need to adopt creative structures through contractual controls, like they did with alibaba and other local firms where they are not adequately protected and take significant risks.

3. Innovative Information Technology and Internet services business can be started in the FTZ without lengthy approval processes. Obviously as most startups use information technology as their backbone for scalalble innovative services its key that this sector is free from redtap. This is exactly what FTZ offers.

4. Negative list approach for foreign investment. This actually supports the angel, private and venture capital investment companies which will help spur the growth of promising startups.

Below is the translation of the first regulatory draft [1]:

General Guideline

In the first section besides discussing the General Requirement related to ideology, overall goal contains a scope which provides the geographical scope of the reforms:

Major Tasks and Measures

This sections defines the 18 services and sectors, the Shanghai FTZ will offer easier investment access to both foreign and domestic capital and further open up. Below are the details of the services and sectors covered in Shanghai FTZ:

Opening-up of the Financial Services Sector

Since Central Government wants Shanghai to be one of the leading financial center like New York, London and Hong Kong. According to the plan, Shanghai FTZ will implement the following measures for opening up of the financial services sectors:

• Shanghai FTZ will start trial programs of RMB convertibility under the capital account, market-oriented interest rates, and cross-border RMB transactions.
• Shanghai FTZ will facilitate market-oriented pricing of financial institution assets.
• Shanghai FTZ will pilot a new mechanism for the management of foreign exchange to facilitate trade investment.
• Shanghai FTZ will encourage enterprises to take advantage of both onshore and offshore markets.
• Shanghai FTZ will deepen foreign debt management reform to facilitate cross-border financing.
• Shanghai FTZ will encourage multinationals to establish regional or global capital management centers within the zone.

Gradually Shanghi FTZ will allow WOFE (Wholly Owned Foreign Enterprises) to do commodity futures trading and open up future delivery warehouse in FTZ.

Conclusion

Although the main purpose of Shanghai Free Trade Zone is to understand and create a blueprint for broader reforms in China’s financial and services sector and implementing free market economy. It also on the other side boost innovation by creating an environment for young enterpreneurs from China and around the world to do innovative startups in China. If Shanghai can play it cards carefully and remove the barriers as envisioned in the buzz around FTZ, it might as well become the Silicon Valley of Asia due to the size of Chinese economy.

Hopefully someone in Central or Shanghai government can take a notice and use this opportunity to build a sustainable ecosystems of angel, private investors and venture capital industry in China to boost the Startup eco-system. As proven by Silicon Valley by creating apple, google, hopefully Shanghai FTZ can create global success similar in scale to google.

The key here is to allow this Startup to get unfiltered access to information from around the world and support them with legal and financial policies to create next big thing.

Footnotes:

Introduction

Online retail is growing steadily with double digit growth rates worldwide with the ubiquity of internet and proliferation of computing devices like laptops, desktops, smartphones and tablets. In some cases online retailers start to cannibalize the sales from traditional offline retailers like the case of Amazon affecting BestBuy. Current internet trends and consumer behaviour is forcing traditional retail and consumer brands to embrace digital transformation and launch online retail. Originally most traditional retail and consumer brands decided to separate the operations of traditional and online channels. Now some of them are exploring integration strategies for their online and traditional offline channels (called multichannel retailing [2]). Retail and consumer brands are adopting multichannel retailing to:

1. Gain access to new markets - The market for store-based retailers is typically limited to the local trading areas of their stores. Thus, adding non-store channels (e.g. Interne, catalogs, mobile phones) enables retailers with limited locations to exploit economies of scope by expanding their markets without building additional stores. With the rapid growth ofonline retail, most consumers will become multichannel shoppers eventually, which means that segmentation based on channel choice may not be very useful in the future.
2. Increase Customer satisfaction and loyalty - Multichannel retailers can better satisfy their customers’ needs by exploiting the benefits and overcoming the deficiencies of each channel. Moreover multichannel customers on average spend more and have a higher lifetime value (LTV) than single-channel consumers [4].
3. Create a strategic advantage - By developing resources that are not easily detected or duplicated by competitors like propriety customer information and tacit knowledge for providing a seamless customer interface. Also this helps in reducing costs and increasing revenues.

In advanced economy like USA more then 80% of retail and consumer brands sell merchandise through multiple channels. In China still most traditional retail and consumer brands do not have or in process of having multichannel retailing. Integration between online and offline channel varies widely among retail and consumer brands. For example most traditional retail and consumer brands often show in-store inventory availability information online and most consumers can do research online and buy offline (ROPO). Advanced configuration includes shipping the product ordered from the store closest to its detination, or offering the option to buy products online and pick them up in the store, or offering to return a product bought online in the store, or order in store and ship to home.

In most of the advanced economy like USA, Western Europe many large traditional retail and consumer brands offer Buy Online Pick Up in Store (BOPS). The retailer during online shopping process shows the locations at which the item is available, and gives customers the option to close the transation online and then pick up the product at one of the locations. Usually the item is ready to be picked up at location within 2 hours, some retailers might take longer but this is the mean average for BOPS functionality to have an impact.

Impact of BOPS

Based on the research done on this subject [1]. Following is the impact of BOPS:

Introduction

China is one the fastest growing eCommerce market in the world [2] and for successfully running a B2C eCommerce operations in China, traditional retail and consumer brands (which already have experince in offline commerce in China) needs additional expertise in:

• Softwares and IT systems - which powers their ecommerce site, 3rd party market, digital marketing, customer services, logistics and supply chain management, online payment services, finances etc.
• Operational expertise in Supply Chain Management, Warehousing, Logistics and Reverse Logistics which helps in managing pick, pack, dispatch, delivery, returns management. Also optionally helps in Vendor Managed Inventory, Drop Shipments, Convenience Store, Corner Stores or Self Store for Pickups and returns. This is in additiona to their expertise in storage, bulk transport and cross-docking they use for their traditional offline store fullfilment.
• Digital Marketing - Using internet and digital media for marketing, campaigns and brand awareness. This expertise can also complement their offline retail operations. Additionally for eCommerce they will need expertise to drive traffic and sales using process and tools for Search Engine, SEO, 3rd Party Aggregators, Price Comparison sites etc.
• Finance and 3rd Party Payment Process - Since online commerce require additional financial process for revenue, payment, reconciliation, profits, returns etc. Like in offline commerce when customer purchase goods in store, they can directly be recognized as revenue, since the goods are already delivered to customer. But in case of eCommerce different companies have different process, some recognize revenue as soon as the goods are given for delivery, some recognize it as soon as the payment is received, some recognize it once the customer receives the goods and they have received the notification from delivery system (either run by themselves or 3rd party).
• Customer Services - eCommerce puts additional burden on managing a good customer services. Also good customer services is one of the key success factors for any decent eCommerce operations. Given the Internet users in China use mostly Instant Messaging, Weibo besides phone. Traditional retail and consumer brands either needs to build this capacity or use some third party provider for it.

Given the current state of development and availability of human resources in this areas of expertise in China, it might look daunting for traditional retail and consumer brands to start eCommerce operations in China. So many of these traditional retail and consumer brands started, by outsourcing part of these expertise to one or multiple Third Party Service providers in China. This led to proliferation of such service providers with varied level of expertise and service offerings to flourish in China. This in turn created a new business segment called “Third Party eCommerce Services” in China. One of the main driver for this market is also Alibaba with its Tmall and JinDong (formerly 360buy.com).

Explosive growth of eCommerce in 2010-2012 resulted in explosive growth of such providers in China. But as the market mature and more and more traditional retail and consumer brands starts building their expertise, such providers are under heavy margin pressure. In order to survive they need to constantly innovate and bring down the costs by building new softare and IT systems and add value to the eCommerce Services offered to retail and consumer brands.

Overview of eCommerce Services Market in China

Areas of eCommerce Services in China

Third Party eCommerce Services market consists of service offering in following areas:

• Software and IT Systems - Example provider includes:

  • Traditional Systems Integrator like Accenture, arvato systems and their software and technology partners like IBM, hybris, Magento etc.
  • Third Party marketplace provider like JinDong (formerly 360buy.com), Taobao, Tmall, eBay etc.
  • Specialized Software provider for eCommerce connecting specialized service providers and their services together like arvato systems aroma, ShopEx etc.
  • Most Traditional Retail and Consumer Brands outsource it in build phase and some during operations phase

• Digital Marketing - Examples provider includes:

  • Traditional Marketing Agencies like WPP, Ogilvy and others which have specialized division for digital marketing.
  • Companies providing specialized platform for digital marketing like Google, Baidu, Weibo, Tencent etc.
  • Companies providing specialized software for digital marketing like Adobe, IBM.

• Finace and Third Party Payment - Example provider include:

  • Traditional Payment Provider like AliPay, TenPay, YeePay, UnionPay, Banks like CMB, ICBCetc.
  • Fraud detection and payment reconciliation provider like arvato, Cybersource.

• Fulfillment (Logistics, SCM & Delivery) - Example provider include:

  • Companies like SF Express, Yunda, ST (ShenTong) Express, UPS, FedEx, DHL etc.
  • Companies like arvato, BaoZun, JinDong (formerly 360buy.com).

• Customer Services - Examples provider include:

  • Companies like arvato, BaoZun, PCCW, 360Buy.com, Accenture etc.
  • Companies providing specialized software for Customer Services like Live800, Oracle Siebel etc.

Due to intense competition in Third Party eCommerce Services business segment most providers try to increase their revenue streams by trying to provide integrated eCommerce services either by building their own integrated stack or partnering with specialized providers. For example arvato is trying to build their own integrated eCommerce Services stack and on the other hand Accenture is providing integrated eCommerce Services by partnering with JinDong with its customer services, logistics, payments and platform, combined with their own Software & IT and Digital Marketing Services.

Third Party eCommerce Service Market Size

Based on the above chart and the myriad of options and services under the umbrella of “eCommerce Services”, make it difficult to estimate the size. My methodology for calculation of Market size is based on projected B2C ecommerce market revenue. So to come up with the market size first analyzed the spending by Traditional Retail and Consumer Brands with following assumptions:

• Most of the traditional retail and consumer brands who are engaged or wants to engage go through buid phase and operations phase and for the market size determination its assumed that at any given time 10%-15% organization are going through build phase and remaining through operations phase. Since the percent weight for each area of expertise for service providers vary depending on the phase. So I take average 12.5% in build phase.
• Digital marketing in eCommerce is classified as Customer Acquisition Cost - In build phase retail and consumer brands are willing to spend 7.5% of order value and for operations phase pay approximately 2.5% of order value according to digital marketing by gartner [3]. According to Amazon 10-K [4] filings they spend net 3.8% of their revenue for marketing in 2012. So for our estimation purpose we take 7.5% for build phase and 3.15% for operations phase.
• Usually in China for third party payment providers charge between 0.5% to 1.2% and for reconciliation and other financial services it may additionally increase to 1.5%-2% of the order value. For the purpose of estimation we will use 1.75%.
• Fulfillment (Logistics, SCM and Delivery)- most efficient estimates is around 10.2% of the total order value, so for estimation I will take 10.2%.
• In order to provide a decent customer services and to retain the customers (since customer acquisition costs 2 to 3 times more then retention) the spending by retail and consumer brands should be 1.3-1.5% of the overall online B2C revenue. Since customer services is one of the key aspects of online retail, I tend to choose 1.5% of the total market size.
• Since eCommerce is technology driven and requires extensive software and IT systems the overall share of technology spending is very high. So during the build phase the overall spending on Software and IT systems is around 10-12% and during operation is around 6.8% (based on estimates from Amazon 10-K [4] filing). So for the purpose of market size estimation I will use 11.5% during build phase and 6.8% during operations phase.

Based on the above assumptions the table below shows the spending by online retailers in different areas of expertise to support B2C sales volume of 1260 billion yuan in 2012.

B2C Market in 2012 [2]		1260 billion ¥
Digital Marketing	(12.5%x7.5%x1260) + (87.5%x3.15%x1260)=	46.8 billion ¥
Finance & Payments	(1.75%x1260)=	22.1 billion ¥
Fulfillment	(10.2%x1260)=	128.5 billion ¥
Customer Services	(1.5%x1260)=	18.9 billion ¥
Software & IT Systems	(12.5%x11.5%x1260) + (87.5%x6.8%x1260)=	92.7 billiion ¥
Total Spending to Support B2C Market		308.8 billion ¥

So in order to manage a B2C market size of 1260 billion ¥, Traditional Retail and Consumer Goods needs to spend approximately 308.8 billion ¥. Based on this, following is the market for individual areas for eCommerce service providers:

• Digital Marketing Size (based on Gartner Digital Marketing Spend Report [3]):

  Total Digital Marketing Size 46.8 billion ¥
  Online advertising	(12.5%x46.8)=	6.1 billion ¥
  Content Creation and Marketing	(11.6%x46.8)=	5.6 billion ¥
  Search Marketing	(10.7%x46.8)=	5.2 billion ¥
  Design, Development and Maintenance of Corporate Website	(10.7%x46.8)=	5.2 billion ¥
  Email Marketing	(9.6%x46.8)=	4.7 billion ¥
  Analytics	(9.5%x46.8)=	4.6 billion ¥
  Social Networks	(9.4%x46.8)=	4.6 billion ¥
  Mobile Marketing	(7.4%x46.8)=	3.6 billion ¥
  Commerce Experiences	(7.2%x46.8)=	3.5 billion ¥
  Video Production	(5.9%x46.8)=	2.9 billion ¥
  Company Blog	(5.3%x46.8)=	2.6 billion ¥
  Others	(0.2%x46.8)=	0.1 billion ¥

• Third Party Payment and Finance Market Size

  Total Payment & Finance Market Size 22.1 billion ¥
  Third Party Payment	(95%x22.1)=	20.99 billion ¥
  Finance	(5%x22.1)=	1.11 billion ¥

• Fullfillment (Logistics, SCM and Delivery) Market Size (estimates based on [5]):

  Total Fulfillment Market Size 128.5 billion ¥
  Logistics & SCM	(42%x128.5)=	53.97 billiion ¥
  Delivery	(58%x128.5)=	74.53 billion ¥

• Customer Service Market Size is 18.1 billion ¥.

• Since the total spend in Software and IT Systems powering B2C commerce is estimated at 92.7 billiion ¥. Since many of the retail and consumer brands spend portion of it on their existing technology teams and process, so majority of the spend is internal. Besides 95% of the online retailers are small to medium size companies, so the overall Market Size:

  Total Market Size of Software and IT Systems is 21.3 billion ¥
  SME Market	(95%x20%x92.7)=	17.6 billiion ¥
  Traditional Large Retail and Consumer Brands Market	(5%x80%x92.7)=	3.71 billiion ¥

So based on this estimates the overall Third Party eCommerce Service Market Size is (46.8+22.1+128.5+18.1+22.3)= 237.8 billion ¥.

But take it with caution, since many of the revenue % share can be dramatically different based on industry and also can be changed by advances in Software and IT Systems. Also not all of this spending will translate into outsourcing to Third Party eCommerce Service Provider, so actual market might be smaller then this. But I feel this estimation can provide a logical framework for others to determine the market size. Funny enough the market size based on this estimation is 18.9% of the total B2C volume, and also if you look at any typical online eCommerce retail like Amazon, JinDong, YiHaoDian you will notice their operating expense is similar between 18-23% for the above eCommerce Service Areas. In hindsight I should have just mapped Amazon 10K reports for 5 years and by finding the average operating expense % of sales revenue would have given the estimation of spending to support a given B2C volume.

Footnotes

Introduction

Its 11:50 PM on 10th November 2013 and Lee MingSha, a 26 year old girl, is eagerly waiting for clock to turn 00:00. It will be a unique day in the year when both day and month are 11. This day is also called Double 11 in China and on this day most of the online retailers offer special discount and deals to celebrate singles day. As the clock turned 00:00, Lee MingSha went to the site of online retailers; where she already placed the items in the basket and start finishing the check-out process. Some went smooth and others need retry or updates. This day she spends more than she will spend in the entire year including Chinese New Year. Obviously most of the sales were driven by discounts and special offers, but it brings windfall of revenue not just for online retailers, but also for more then 217 Brands and traditional retailers participating in it.

Alibaba Double 11 Sales [1]

The combined sales of only Alibaba, which includes Tmall and Taobao, was RMB 19.2 billion (US$ 3.08 billion) more than the entire Cyber Monday in USA. Obviously it was not all cakewalk as there were periods of unstability on the online platforms due to unexpected surge of orders and traffic. This was a sudden wake up call for most of the traditional retailers and consumer brands, of changing digital and internet landscape in China. According to the latest report [2] from CNNIC:

• By the end of December 2012, China added 50.9 million new Internet users to reach a total of 564 million. This represent a growth rate of 3.8% compared to 2011 and Internet Penetration of 42.1%.
• By the end of December 2012, China added 64.2 million new Mobile Internet Users, reaching a total of 422 million Mobile Internet Users.
• By the end of 2012, China added 19.6 million Rural Internet Users to reach 156 million, representing 27.6% of the total Internet users in China.
• The ratio of Internet users (participating in Survey) using Desktop to access dropped to 70.3%, a drop of 3% compared to 2011. The ratio of Internet users using laptops dropped slightly to 45.9%, a drop of 0.9% compared to 2011. The ratio of Internet users using mobile phones rose to 74.5% from 69.3% in 2011.
• China had a total of 13.41 million domain names, including 7.51 million “.CN” domain names and 283,484 “.中国” domain names.
• The total number of websites, i.e. the number of websites registered by the registrants of website domain names, continue to grow reaching 2.68 million.
• By the end of December 2012, the ratio of the interviewed small and medium sized enterprises (SMEs) depending on computer-based offices reached 91.3%, and with access to Internet reached 78.5%.
• The penetration rate of fixed broadbandamong SMEs hit 71.0%.
• The proportions of online sale and online purchase among SMEs stands at 25.3% and 26.5% respectively.
• The ratio of SMEs employing internet to carry out marketing and promotion reached 23.0%.

Key Internet Trends in China

Internet Users

China has become a major internet market with increasingly sophisticated users. By 2015 it will add another 200 million users reaching over 700 million users double the number of users of US and Japan combined. On an average internet users spend 20.5 hours/week online. China also witnessed an explosive growth in mobile internet users and reached 422 million in 2012.

PC and Mobile Internet Users in China [2]

Internet eco-system and applications in China

China is a huge online market, but it’s not an easy one. The online eco-system in China is highly complex and due to regulatory and cultural differences there are several competing systems for every major global internet services . The big three Internet companies - Alibaba (with Taobao, Tmall and Alipay), Baidu and Tencent (with Weixin, QQ), have a firm hold on their corners of the market (eCommerce, search, and messaging, respectively). They are working hard to consolidate their positions and move into new fields. There are many new upcoming internet companies which are always challenging the dominant position of these companies with the recent addition of Sina (for its weibo service), JinDong (earlier known as 360buy.com) for e-Commerce services, Youku for online video. China’s online eco-system is highly dynamic and most global players’ market share is relatively small. The picture and the table below illustrate China’s dynamic internet online eco-system.

China’s Online eco-system [3]

Application	# of User	Usage Rate
IM	468 million	82.90%
Search	451 million	80.00%
Music	435 million	77.30%
Blogs	373 million	66.10%
Video	372 million	65.90%
Gaming	336 million	59.50%
Microblog	309 million	54.70%
SNS	275 million	48.80%
Email	251 million	44.50%
Shopping	242 million	42.90%
Travel Booking	112 million	19.80%
Group Buying	83 million	14.80%

Following are some of the highlights of key trends in China which will impact companies operating here.

• Internet has replaced newspaper and in some cases even television as the medium with the highest daily reach in urban areas. Indeed now television has dedicated shows which cover hot Weibo tweets, news and trends on Internet in China.
• Majority of the internet users spend most of their time on entertainment activities like IM, Music, Video, Gaming and Shopping. Internet is slowly replacing television as the primary entertainment medium among young Chinese between 10-40 years in urban areas.
• With 468 million users IM (instant messaging) is the most effective way for communitcating with potential customers for companies. This is a primary difference between western and Chinese market where instead of email, people prefer real-time instant messaging or mobile phone for customer support services.
• According to a survey half of the Internet users say that the Internet is their most trusted source of information, followed by television at 30 percent and 15 percent for newspapers.
• Although there is a shift within society towards Internet as primary medium for entertainment, retailers and brands still spend more money on traditional advertising.

Companies with ambitions in China should have a strong Internet presence and strategy. Internet and technology will no longer be supporting function, but will become part of the core functions similar to sales, marketing and operations.

Joyent Cloud Account Setup

While selecting a cloud service provider of choice after frustrating experience with Microsoft Windows Azure, due to its phone verification issues and with Amazon AWS due to its non acceptance of CVV2 verification (which by the way I feel is ridiculous) settled on Joyent Cloud. Following are the steps for setting up a Joyent Free Trial Account, which I intend to test for next 3 months:

• Step 1: Sign up for an account by filling up the registration form and later on providing the billing information and a valid credit card with cvv2.

  

• Step 2: Generate SSH key the instructions for Max OS X and Windows can be found in the link Generating an SSH Key

Joyent Cloud Ubuntu Server Setup

Once you have setup your account and logged on to Joyent Control Panel you can create instance for a virtual machine. Once the machine is running you can see the assigned ip address. You can log on to your machine using Putty on Windows or SSH on mac.

Installing Necessary Software

Normally when you setup Ubuntu from image repository of Joyent Cloud, you will get a bare-bones server install. Now we need to setup LAMP stack on the server. Usually I prefer to use Fabric for automatic setup for a virtual machine on the cloud. But since this was just one off setup so just used the command shell.

Installing ufw, apache2, mysql and php 5 on the server. I will also install Mercurial to get the code from the main repository where the code is regularly merged from upstream git repository of OpenCart. Since opencart repository is on github need to install hg-git mercurial plugin:

apt-get install ufw apache2 php5-mysql php5 php-pear mysql-server mysql-client mercurial php5-curl php5-gd php5-mcrypt
pip install hg-git

If everything goes smooth you will have apache2 webserver with PHP, MySQL and apache will be setup. You can test it by going to http://youripaddress/ and should see a page titled “It Works!”.

Download and Install OpenCart

Once the necessary software is installed now its time to setup OpenCart. Before beginning the download and install of opencart I would like to create separate accounts for future automatic deployment and merge of opencart:

hg clone git+https://github.com/opencart/opencart.git
hg clone opencart ecommerce_main

The above command will create two repository one in directory opencart and another which is a clone of upstream repository on the local machine. In future all the local changes and customization of the opencart will be managed in ecommerce_main and it will be used for deploying opencart on the servers.

Once opencart is downloaded its installed according to instructions on OpenCart Installation Guide. However there is a minor variation in the way we install the opencart as we are not using windows or filezilla.

Following steps were done on command line to install OpenCart on Ubuntu Server:

# Create mysql username and password for OpenCart
mysql -u root -p
mysql> CREATE DATABASE dbname CHARACTER SET utf8 COLLATE utf8_general_ci;
mysql> create user 'user_name'@'localhost' identified by 'user_pass';
mysql> grant all privileges on dbname.* to 'user_name'@'localhost';

# Try logging on to mysql using the given username and password
mysql -u user_name -p dbname
mysql>

# Clone the directory ecommerce_main to the public html. This structure is specific to Ubuntu.
hg clone ecommerce_main /var/www/vhosts/site_name

# Go to the directory of the cloned repository and update to release version 1.5.5.1
cd /var/www/vhosts/site_name
hg update v1.5.5.1

# Go to upload directory and copy config-dist to config.php and change owner for few files
cd /var/www/vhosts/site_name/upload
cp config-dist.php config.php
cp admin/config-dist.php admin/config.php
chown -R www-data:www-data /var/www/vhosts/site_name/upload/system/cache/
chown -R www-data:www-data /var/www/vhosts/site_name/upload/system/logs/
chown -R www-data:www-data /var/www/vhosts/site_name/upload/image/
chown -R www-data:www-data /var/www/vhosts/site_name/upload/image/cache/
chown -R www-data:www-data /var/www/vhosts/site_name/upload/image/data/
chown -R www-data:www-data /var/www/vhosts/site_name/upload/download/
chown www-data:www-data config.php admin/config.php

# Change the Default directory for apache server in Ubuntu
# Open file /etc/apache2/sites-available/default in any editor
# Change /var/www to  /var/www/vhosts/site_name/upload

Once the above steps are done you are ready to proceed with the installation point your browser to http://yourserver/ and you will be presented with the site for installation now you can follow the OpenCart installation guide to install. Usually for database prefix use site_name abbreviation but choice is yours.

After the installation run following commands to remove the install directory and also change permissions for config.php so that others cannot change it. Besides the changes are transferred back to the ecommerce_main repository so that it reflects the updated production version:

# Change to upload directory
cd /var/www/vhosts/site_name/upload
chown root:root config.php admin/config.php

# Remove install directory using mercurial and config.php and upload to server.
hg remove install
hg add config.php admin/config.php
hg ci -m "Production installed version"
hg push

# If you do not use mercurial then
rm -rf install

Now you can go to your store at http://yourserver/ and see the newly installed store. Go to http://yourserver/admin/ for managing it.