New developments in networking can enable data center operators to go all-Ethernet- even for storage area networking

It’s been long since the evolution of storage solutions had begun. The shift from hard-disk drives to SSDs has already started in earnest but it will gather momentum as data center operators seek to wring greater energy savings from their infrastructure, increase their service reliability and uptime still further, and pack more computing power onto valuable real estate. In the new-age data center, instead of spinning rust, storage will be all-flash and data center networks will be consolidated on industry-standard Ethernet, from ‘cold storage’ all the way up to the server.

With the usage of SSDs, power consumption can be reduced by 70 percent when compared to conventional hard-disk storage, and cut the space devoted to them by half. Furthermore, as a result of the introduction of deduplication and compression technology, economics has shifted in favour of flash storage.

The all-Ethernet data center

Due to packet loss, the use of Ethernet in storage area networks has been constrained resulting in a radical affect on networking performance.  This is generally due to multiple servers sending a large volume of packets to a server at the same time, exceeding the processing capability of the network infrastructure. Moreover, Ethernet was never designed to work over the kind of distances that data center storage area networks operate.

However, some of the recent technology solutions such as Huawei’s iLossless-DCI algorithm has been disrupting this –space. The latest CloudFabric 3.0 Hyper-Converged Data Center Network solution – can extend the range of Ethernet to 70 km. These algorithms use artificial intelligence trained over millions of real-service samples, and millions of random samples.

The advancements of such algorithms will establish a new age of data center network consolidation on Ethernet as a precursor to exceptional data center automation.

Automation challenges

In the ear of cloud, the conventional approaches to designing, deploying, provisioning, and operating networks have become invalid. Applications have become the lifeblood of today’s enterprises, and networks acts as the digital nervous systems that support those applications.

Today’s organizations need more than a fragmented and piecemeal approach represented by a mix of automated scripts and configuration automation tools. A comprehensive approach to intelligent network automation that provides a flexible architectural foundation and optimal operational simplicity is the need of the time. However, this remains a long way off. Keeping pace with compute and storage automation is one of the major challenges faced by organizations even today. Additionally, supporting mixed workloads and environments, as well as managing multi-vendor infrastructure, including cloud architectures and applications, has also been proving onerous.

Keys to the future

New solutions such as Huawei’s CloudFabric 3.0 provide device replacement impact analysis, interface capacity prediction, and RoCE (RDMA over Converged Ethernet) traffic optimization to ensure zero [Ethernet] packet loss. These systems provide recommendations to assist manual decision-making to implement basic closed-loop management of dynamic policies

The development of lossless long-distance Ethernet communications opens up many possibilities including the implementation of networking-as-a-service and second-level provisioning. The OPEX [operational expenditure] can be reduced by 30 percent by providing end-to-end intelligent deployment and operations. These solutions can also build all-Ethernet computing and storage networks with zero packet loss, helping to unleash computing performance by 100 percent.

It has taken 40 years or so for the first lossless Ethernet implementation to become commercially available and it might take years before data center operators realize the potential of a transition to all-Ethernet networking. However, the combination of greater automation and radically lower operational expenditure ought to encourage many to experiment if their data centers are ready for Huawei’s Level 3 autonomous driving network technology.

Authored by:- Mr. Fisher, Head- Marketing & Solution Sales, Huawei India

(The views expressed in this article are by – Mr. Fisher, Head- Marketing & Solution Sales, Huawei India. doesn’t own any responsibility for it.)

Leave a Reply

Your email address will not be published. Required fields are marked *