Shulou(Shulou.com)06/02 Report--

Super converged Infrastructure (Hyperconverged Infrastructure), as one of the most important transformational technologies in the field of IT infrastructure in recent years, has been known to more and more customers. Some enterprises have tried to apply the super-convergence architecture to various business scenarios, but due to the great differences between the super-convergence and the original architecture, there are various views and voices in the market, resulting in the following questions and concerns for many users:

1. Why do you want to go to super-converged infrastructure? What is the core difference between super-integration and traditional architecture?

two。 The value of super-integration is to integrate all the functions together?

3. Super convergence is built on servers. Is the performance and reliability sufficient to support the core business?

4. I heard that the advantage of super integration is that it is cheap.

5. With the super-integration architecture, there are great changes in the use, maintenance and even procurement mode of IT. Is it really worth "messing around"?

The following is a detailed comparison and analysis to let readers understand the differences between the super-converged architecture and the traditional architecture, as well as the important reasons for the improvement in reliability, performance, scalability and manageability of this new architecture.

I. comparison of architecture and resource management models

Taking the SmartX super-fusion product as an example, the differences of deployment and resource management modes between the lower super-fusion architecture and the traditional architecture are given below.

Figure 1 deployment difference between traditional architecture and super-converged architecture

Figure 2 difference between deployment and resource management model of traditional architecture and super-integrated architecture

Compared with the traditional FC SAN architecture, the super-fusion architecture has the following significant changes:

1. Instead of proprietary storage hardware and networks, use standard and easy-to-maintain x86 servers and 10 Gigabit Ethernet switches

two。 The core is distributed storage, and each server is a storage controller. It should be noted that each node needs to configure an SSD disk as a cache, which is much larger than the volatile memory of traditional storage, and multi-node concurrency brings stronger aggregation performance.

3. Logically, storage is no longer a different storage space composed of RAID, but a unified and scalable storage pool with stronger scalability.

4. Another benefit of distributed storage is that performance and capacity can be configured on demand within a unified resource pool, and data can be automatically balanced.

5. Compute virtualization and storage are deployed on the same server node.

Second, distributed storage + virtualized converged deployment is the essence of super-converged architecture

For the super-fusion architecture, because of its unique name, many users have a misunderstanding: the core of the super-fusion architecture is integration, so the more functions of fusion, the better. However, through the comparison of the above architectures, we can see that the change of the super-converged architecture is first of all the replacement of distributed storage for traditional storage. Other more advantages (such as building based on x86 servers, concurrency and easy scalability) are based on this substitution.

Of course, the unique deployment mode of distributed storage and virtualization further simplifies the IT architecture of users, reduces the cost and difficulty of operation and maintenance, and these values also greatly accelerate the acceptance of distributed storage mode.

Third, what improvements will super-converged architecture bring to IT infrastructure? Why?

The local upgrades of different products and the improvements brought by the super-converged architecture to the IT infrastructure are comprehensive, which will be explained one by one below.

1. Effectively improve reliability

Many enterprise customers have a common misconception about software-defined data center architecture: is the system built by x86 servers reliable? Although the x86 server itself has a single point of failure, superconvergence itself works in a cluster mode, and the first problem to be solved in its core-distributed storage system is to use technologies such as multiple replicas to build large-scale systems with higher reliability. these technologies are the core of building software-defined data centers.

Not only that, distributed storage has the following advantages in terms of reliability over traditional centralized storage, which is compared below:

In addition, SmartX, Nutanix, VMware and other vendors provide enterprise-class storage services ranging from data checksum to rack-aware, dual-active, remote backup and other features, as well as enterprise-class high availability and data protection solutions integrated with third-party products.

2. Concurrency performance is greatly improved and IO latency is reduced.

The following still uses SMTX OS to explain why the super-converged architecture can improve performance and reduce access latency.

Among them, the distributed architecture improves the aggregation performance of the system as a whole; the SSD cache mechanism improves the single-node access performance; and the Icano localization mechanism is the advantage of the super-fusion deployment mode, which can further reduce the access latency. It should be noted that SmartX O localization is mainly supported by SmartX and Nutanix, and super fusion based on open source products such as Ceph cannot be supported.

To take a practical example, because dual control has become a bottleneck, HP 3PAR 8440 has basically reached its maximum performance in eight SSD blocks, but this read-write performance can be achieved by using only four SmartX superfusion nodes.

3. Scalability is greatly improved.

As mentioned earlier, the core distributed storage of the super-converged architecture has a substantial improvement in scalability compared to traditional storage, including the following features:

Among them, heterogeneous node support needs to be confirmed with vendors, similar to SmartX can support, but many vendors do not provide such support.

4. The difficulty of operation and maintenance is greatly reduced.

In view of the difficulty of operation and maintenance, we can compare the two architectures for the entire operation and maintenance cycle:

As can be seen from the above figure, the super-fusion architecture not only has a large number of automated operations, simple operation and maintenance, but also has a short time and high efficiency in the whole product operation and maintenance cycle. It can effectively reduce personnel requirements and liberate IT personnel for more innovative activities.

5. Reduction of procurement cost and total cost of ownership

In terms of the cost that customers are most concerned about, the procurement cost of server + super-converged software (or super-converged all-in-one machine) has been greatly reduced compared with the server plus traditional mid-and high-end storage. But in addition to the procurement cost, super-integration has a greater advantage in the total cost of ownership.

For a comparison of the cost advantages of super integration, please click:

What is the cost advantage of super-convergence over traditional FC SAN architecture?

As you can see, superconvergence makes the IT infrastructure more agile and capable at a better cost through innovative architecture, and these features will lay a solid foundation for users to accelerate their digital transformation.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.