China Cloud storage In-depth survey

China Cloud Storage Market dynamics

China's current corporate cio/cto on the public cloud is still skeptical, at present, about 70% of enterprises are unwilling to put the enterprise data on the public cloud, mainly from the perspective of security, but, the data for business investment more and more, Will force enterprises, especially for data security is very high financial industry also gradually towards the public cloud.

Domestic enterprise concept is more accustomed to spend money to buy hardware products rather than services, as the largest storage demand for the government, telecommunications, finance three industries, in China have their own data centers.

In the SME market, the current 90% of SMEs in storage procurement only 10% of the level, so the procurement of cloud storage capacity is very low.

The security of cloud storage requires a legal constraint on the security and reliability of cloud storage and the need for third-party organizations to help enterprises to audit cloud storage,

At the same time, China's relatively backward network construction restricts the development of cloud storage.

Market drivers for cloud storage in China

1. Network Broadband

Cloud storage System will be a multiple-region distribution, the vast global system, the current user is mostly through ADSL, DDN and other broadband access devices to connect cloud storage, so the cloud storage users need to use broadband network and storage system connectivity, only broadband network has been fully developed, Users can obtain large enough data transmission bandwidth, achieve large capacity data transmission, enjoy the convenience of cloud storage.

2.WEB 2.0 Technology

Through the WEB2.0 technology can make the user's application way and the service which can obtain is more flexible and diverse, the current cloud storage user only through WEB2.0 technology, will the PC, the handset, the mobile terminal and so on many kinds of equipments, realizes the data, the document, the picture, the audio, the video and so on the

3. Application Storage

When applying storage, a storage device that integrates the functionality of the application software in a storage device, it not only has the data storage function, but also includes the partial application software function, just like the data center contains the program application server and the database server, the application storage and the cloud storage union, may reduce the cloud storage server's forest, Reduce the cost of enterprise system construction, less system due to the single point of failure and performance bottlenecks caused by the server, improve system performance and efficiency.

4. Cluster technology, distributed file system and Network Computing technology development

Cloud storage System is a collection of multiple storage devices, multiple applications and multiple services, which need to realize the cooperative work of each storage device through clustering technology, distributed file system and Network computing.

5. Network Storage security Technology development

As the data transmission of cloud storage is transmitted through ordinary broadband, the security of data transmission must be ensured, and cloud storage can ensure that the protective devices in cloud storage are not accessed by unauthorized users through CDN content publishing system and data encryption technology. Each data backup and disaster-tolerant technology ensures that the data in the cloud storage is not lost, thus ensuring the security and stability of the cloud storage itself.

6. Development of storage management technology

Cloud storage involves the transfer management between multiple geographies, multiple vendors, and multiple hardware devices, so storage virtualization management and Multilink redundancy management are particularly important, as well as creating a performance bottleneck for storage systems like the network, which also leads to structural instability.

Due to the large number of storage devices in cloud storage, it is a difficult problem for storage equipment operators to solve the problems of centralized management, condition monitoring, fault maintenance and human cost.

Challenges facing China's cloud storage market

1. Data Centre Building

Data center construction needs a lot of input, China's domestic enterprises rarely have such a large strength, and if the domestic enterprises to build cloud storage, it needs a very large financial pressure. If the government deploys, many laws and regulations may limit its deployment.

2. Users of domestic cloud storage

Domestic has set up some cloud computing centers, but most customers are government or large enterprises to meet the user's computing needs, customer base is relatively limited, profitability is weak. Similarly, the establishment of cloud storage services, in the absence of a certain number of domestic users, how to maintain a cloud storage data center is a very big problem.

Huawei's internal data center (images from the Internet)

3. Reliability

Currently, the reliability of the cloud storage services is not yet up to the enterprise-class array requirements.

4. Large Enterprises

Large enterprises already have their own it facilities, whether willing to abandon the original IT facilities, the enterprise's information systems, storage systems to rearrange, the need for a huge investment.

The future development trend of cloud storage

Cloud storage has become a trend for future storage development, at present, cloud storage vendors are combining various kinds of search, application technology and cloud storage to provide a series of data services to enterprises, but the development trend of future cloud storage is mainly from the aspects of security, portability and data access.

Security。 Since the advent of cloud computing, security has been one of the primary considerations in the implementation of cloud computing, as in the case of cloud storage, security is still a concern in the home page, and security is often the primary business consideration and technical consideration for customers who want to cloud storage. But many users have more security requirements for cloud storage than they can provide for their own architectures. Even so, faced with such high unrealistic security requirements, many large, reliable cloud storage vendors are also struggling to meet their requirements by building data centers that are much more secure than most enterprise data centers and protecting data in storage with encryption layers and a protective layer comparable to the NSA. Today, users often find that cloud storage has less security vulnerabilities and that cloud storage provides a higher level of security than a user's own data center can provide.

Portability. Some users also consider the portability of data when considering managed storage. This is generally guaranteed, and some large service providers offer solutions that promise their data portability is comparable to the best traditional local storage. Some cloud storage combines powerful portable capabilities to transfer the entire dataset to any medium you choose, even to a dedicated storage device.

Performance and availability. Some of the past managed storage and Remote Storage always had a long latency problem. Similarly, the nature of the internet itself poses a serious threat to service availability. The latest generation of cloud storage has a breakthrough achievement, reflected in the client or local device cache, keeping the most frequently used data locally, effectively easing internet latency issues. With local caching, these devices can mitigate latency problems even when the most severe network outages are encountered. These devices also make it possible for frequently used data to appear as fast as local storage. With a local NAS gateway, cloud storage can even mimic the availability, performance, and visibility of midrange NAS devices while remotely protecting the data to a small number of businesses. Another problem with performance is that these solutions can consume too much internet bandwidth when the rate of data change is too high, and this may cause hidden costs to your cloud storage solution. Even so, vendors will continue to strive for capacity optimization and WAN (WAN) optimizations to minimize data latency.

Data access. Another common concern is whether cloud storage can provide sufficient accessibility if large-scale data requests or data recovery operations are performed. As we discussed earlier, some vendors can transfer large amounts of data to any type of media, sending data directly to the enterprise, and at a faster rate than a copy-and-paste operation. In addition, some vendors can provide a set of components that mimic cloud addresses on fully localized systems, allowing local NAS gateway devices to continue to function without having to be reset. If large manufacturers build more regional facilities, data transfer time will be shortened. More importantly, even if you have a catastrophic loss of local data, vendors can transfer the data back to you.

General structure of cloud storage

The concept of cloud storage began with a service provided by Amazon, along with its cloud computing offerings. Behind Amazon's S3 service, it manages a number of commodity hardware devices and bundles the appropriate software to create a storage pool. Emerging web companies have embraced the product and proposed the term cloud storage and its corresponding concepts.

Cloud storage is a schema, not a service. Whether you own or lease this architecture is a secondary issue. Fundamentally, cloud storage can easily extend cloud capacity and performance by adding standard hardware and access to shared standard networks. It turns out that managing hundreds of servers makes it feel like a single, large storage pool device is a fairly challenging job. Early suppliers took on this responsibility and made a profit in the form of online rentals. Other vendors, such as Google, employ a large number of engineers to implement this management within their firewalls and customize storage nodes to run applications on them. Cloud storage has become a highly disruptive technology in the data center as Moore's law has driven down the prices of disk and CPU goods.

The cluster NAS system has improved over the past decade. This article reviews various architectural approaches to building a cloud storage or large-scale scalable NAS system, for enterprise IT managers seeking to build private cloud storage to meet their consumption, or for service providers seeking to build public cloud storage products to provide storage in the form of services, These methods are closely related to them. The architecture approach is divided into two categories: one is the architecture through the service, the other is the architecture through software or hardware devices.

The traditional system utilizes tightly coupled symmetric architectures designed to solve the problem of HPC (High performance computing, super Operations) and is now expanding outwards into cloud storage to meet the fast-presenting market demands. The next-generation architecture has adopted a loosely coupled asymmetric architecture that centralizes metadata and control operations, which are not well-suited for high-performance HPC, but are designed to address the bulk storage requirements of cloud deployments. The summary information for the various schemas is as follows:

Tightly coupled symmetric (TCS) Architecture:

TCS is built to address the challenges of single file performance, which limits the development of traditional NAS systems. The advantages of HPC systems quickly overwhelm storage because they require a much greater number of single file I/O operations than single device I/O operations. The industry's response to this is to create products that utilize the TCS architecture, many of which are accompanied by distributed lock management (which locks the write operations of different parts of the file) and cache consistency. This solution works well for single file throughput issues, and many HPC customers in several different industries have already adopted this solution. This solution is advanced and requires a certain degree of technical experience to install and use.

Loosely coupled asymmetric (LCA) Architecture:

The LCA system uses different methods to extend outward. Instead of executing a policy to make each node aware of what each action is doing, it uses a central metadata control server outside of a data path. Centralized control offers many benefits, allowing for new levels of expansion:

Storage nodes can focus on the requirement to provide read and write services without the need for acknowledgement from network nodes.

Nodes can take advantage of different hardware CPUs and storage configurations, and still play a role in cloud storage.

Users can adjust cloud storage by taking advantage of hardware performance or virtualization instances.

Eliminating the large amount of state overhead shared between nodes can also eliminate the need for user computers to interconnect, such as Fibre Channel or infiniband, to further reduce costs.

The blending and matching of heterogeneous hardware enables users to expand storage on the scale of the current economy when needed, while providing permanent data availability.

Having centralized metadata means that storage nodes can be rotated for deep application archiving, and metadata is often available on control nodes.

Cloud storage Selection

Although there are many options on scalable NAS platforms, but generally, they represent a service, a hardware device, or a software solution, each with its own strengths and weaknesses (Pcpop Enterprise channel recommends "Seven tips to help the enterprise successfully deploy cloud storage"):

Structure model of cloud storage System

Service mode

Most commonly, when you think about cloud storage, you think of the service offerings it offers. This pattern is easy to start, and its extensibility is almost instantaneous. By definition, you have a backup of offsite data. However, bandwidth is limited, so consider your recovery model. You have to meet the needs of your data outside your network.

HW Mode

This deployment is behind the firewall and provides better throughput than the public internal network. It is convenient to purchase integrated hardware storage solutions and, if the vendor is doing well on installation/management, it is often accompanied by an organic rack and stack model. However, you will give up some of the advantages of Moore's law because you will be limited by hardware devices.

SW mode

SW mode has the advantage of HW mode. In addition, it has a price competitive advantage that HW does not have. However, its installation/management procedures should be carefully focused, because it is really difficult to install some SW, or may require other conditions to limit the choice of HW, and choose SW.

With the advent of a massive digital data era, in which companies use YouTube to distribute training videos, there is no need to place these numbers "data" everywhere. Like these companies are committed to the creation and distribution of content, genome research, medical imaging requirements will be more rigorous and accurate. The LCS architecture's cloud storage is ideal for this type of workload, and provides a huge cost, performance, and management advantage.

Cloud storage products for mainstream manufacturers

At present, many manufacturers have launched a cloud storage products, more famous is EMC's Atmos cloud storage infrastructure solution, Atmos is a policy-based management system that provides the ability to build different types of cloud storage, for example, to create two copies of files for non-paying users. and stored in different locations around the world, and for paid users to create 5~10 backups for storage, and to provide their access to files around the world with higher reliability and faster access. In software systems, Atmos includes data services, such as replication, data compression, and data de-duplication, to obtain hundreds of TB of hard disk storage space through inexpensive standard x86 servers. EMC promises that it has the ability to automatically configure new storage space and adapt to hardware failures. Also allows users to manage and read using Web service protocols. Currently there are three versions of Atmos, with system capacity of 120TB, 240TB, and 360TB, all based on x86 servers and support gigabit or 10GbE Ethernet connections.

IBM side

XIV is a new generation of storage products provided by IBM. It uses grid technology, which greatly improves the reliability of data, the scalability of capacity, and the manageability of the system. XIV is upgraded over traditional storage devices. It has a mass storage device + Large capacity file system + High throughput Internet Data access interface + Management system design features. XIV because of its unique design, so that it is born with a huge amount of storage capacity and strong scalability, to meet the needs of a variety of Web2.0 applications, is an ideal product to achieve cloud storage.

"XIV products with IBM Information Management, protection, archiving and other important functions, is the IBM information infrastructure and storage key components, but also IBM can redefine the concept of storage products." "IBM Systems and Technology division of Greater China Product Department general manager Hou said."

The XIV architecture combines the features of midrange and high-end storage. When a user has a new business, or the data grows rapidly, and is able to anticipate a high rate of growth in the future and a complex data type, XIV is a reasonable choice for the user.

The virtualization technology built into the XIV storage system dramatically simplifies management and configuration tasks, and thin provisioning improves IT operations, almost unlimited snapshots, and instant cloning of data volumes, significantly increasing the speed of testing and accessing database operations. Its aim is to provide highly consistent performance by eliminating the full footprint of hotspots and system resources. IBM? The XIV storage System enables users to deploy reliable, multipurpose, and usable information infrastructures, while improving storage management, configuration, and improved asset utilization.

hp

ExDS9100 (StorageWorks 9100 extremedatastorage) is a massive scalable storage system for file content that combines HP PolyServe software, BladeSystem chassis, and blade servers to improve performance. Also used is the storage called "block". These blocks contain 82 1TB SAS drives in the same container.

Designed to simplify PB-level data management, the new business services offered to Web2.0 and digital media companies, including picture sharing, streaming media, video ExDS9100, and social networking, bring a lot of document-based data to fully meet the immediate storage and management needs. At the same time can meet the oil and gas production, safety monitoring and genetic research and other large enterprises similar needs.

EXDS9100 is a unified system with the following three main components:

Performanceblock: Energy efficient Hpbladesystem chassis equipped with Blade server to meet the needs of a large and high-performance operation. The solution's basic features include four blades, which can be extended to 16 blade configurations, each with up to 12.8 cores, with performance reaching 3.2GB per second.

Capacityblock: Basic configuration provides three highly available storage blocks and up to 246TB of storage capacity. The maximum configuration supports up to 10 storage blocks, providing 820TB of storage capacity.

Software: The system uses HP's file clustering technology to meet the stringent requirements of Web2.0 and digital environments. To reduce the complexity and cost of the system, the application can run directly on the server module, removing unnecessary software layers. Through a single image management interface, users can easily manage more storage products and devices.

EMC aspects

Atmos is a software and hardware combination kit, code-named Maui and Hulk respectively. Mike Feinberg, senior vice president of the EMC Cloud Infrastructure Department, said that it was originally designed to help users manage data at dozens of or hundreds of different geographical locations at gigabit levels. EMC says Atmos features automated architecture, autonomous repair, and cloud storage, primarily for media and entertainment companies, telecoms companies, and Web 2.0 Web service providers designed to help them build external cloud storage services or build in-house storage cloud based on cloud storage concepts, Storage capacity can scale to petabytes, support billions of of files and objects, and provide the ability to access the world wide.

EMC offers three different configurations for Atmos, all based on x86 servers and supports gigabit or 10Gb Ethernet, 120TB, 240TB, and 360tb,martin, respectively. We run 15 1TB drives in a 3U rack, and you can disassemble any part. "Jon Martin, director of product management at EMC's cloud infrastructure division, says Atmos is" fundamentally different "from existing storage-system software, with most of its management and core functionality built into storage systems.

Atmos mainly provides a cloud storage service, store a copy of the file separately in different locations, for example, it can create two copies of files for users free of charge, and store them in different data centers around the world, providing 5 to 10 backups for paying users to provide faster access speed and higher security stability to users around the world. In its software component, atoms includes a wide range of data services, such as replication, data compression, and de-duplication, resulting in hundreds of terabytes of hard disk storage through inexpensive standard x86 servers. EMC says atoms has the ability to automatically configure new storage space and automatically adjust hardware failures, and allows users to manage and read various types of data using Web service protocols.