Why 400G Ethernet? In one sentence, because the easiest way to go faster is to go faster.
Over time, Ethernet speed transitions have been the primary driver for improving both the throughput and price-performance of datacenter networks. 400G Ethernet is the next major transition on this journey. While 100G Ethernet is still ramping up rapidly this year and next, it is projected that by the end of 2021 400G Ethernet will represent the majority of Ethernet bandwidth shipped.
When I joined Arista ten years ago, we were in the midst of developing a novel purpose-built software architecture from a clean sheet of paper for networking. The financial services industry was in crisis, with the collapse of major banks like Lehman Brothers. In parallel, emerging slowly but surely, was a new breed of hyper-scale cloud operators. Amazon AWS, Microsoft Azure and Google GCP were in the early stages of shaping what was to become the public cloud. The requirements of these new titans provided a source of inspiration for the Arista founders and me. We couldn’t have predicted the pace nor the impact of this cloud fury; it came strongly and rapidly. As I reminisce over the past decade, it is clear that the cloud inflection has forever changed the face of enterprise IT for the better. Yet it is just the beginning, and there is much ahead.
Every CXO worries about security because the perimeter is changing; in fact, there are no walls for protection. The lines between cloud, workloads, applications, enterprise networks and hosts are blurring and the challenges are getting exponentially greater. The true security architect must rapidly address the reality of a more holistic network-wide security strategy. It must be one that goes beyond the cyber threat of the day to address the risk, scale and mitigation of persistent security issues. The state of cyber security needs urgent resolution because:
At Arista Networks, the status quo inspires us to innovate and continue our mission to reinvent the network – from cloud to client. Today, we’re continuing that journey – into the campus network. Let’s face it; the legacy three-tier architecture of access-aggregation-core is wasteful and oversubscribed – creating a perfect storm for market transitions and Arista innovation.
The last 40 years have seen tremendous growth and progress in the data networking industry. Ethernet, IP, MPLS, GRE, IPsec, MACsec, and VXLAN enable operators to build secure, multiservice, high-performance data planes that interoperate across multiple vendors, multiple operators, and multiple administrative domains. Likewise, BGP, OSPF, IS-IS, LDP, RSVP, BFD, LACP, L3VPN, VPLS, and EVPN enable operators to build scalable multi-vendor control planes that federate across organizational boundaries, supporting mission-critical networks with global reach.
At Arista we have always embraced open networking trends by designing our hardware and software to be as programmable as possible, driving the use of merchant silicon and diversity for the broader industry. It has allowed our customers to select their favorite silicon architectures for the switch pipeline and choose the suite of software and hardware they want to form their cognitive network systems.
Artificial Intelligence, machine and deep learning have to be among the most popular tech-words of the past few years, and I was hoping that I wouldn’t get swept away by it. But when I heard a panel on this topic at our customer event this month on the state of AI networks, I found it incredibly fascinating and it piqued my curiosity! Let me start with a note of disclaimer for readers who are expecting a deep tutorial from me. There is a vast amount of research behind models and algorithms on this topic that I will not even attempt to cover. Instead I will try to share some thoughts on the practical relevance of this promising field.
The modern cloud networking world is vastly different from traditional enterprise IT, and the gap is getting wider everyday. How does one truly scale across millions of machines and workloads globally? A decade ago, Arista pioneered the new software driven networking era and the same challenges now exist not only across the LAN intra-datacenters but also inter-datacenters via wide area networks (WANS). Although we have never promoted “SD-LAN” nor understood the “SD-WAN” hype, Arista has redefined software driven networking and pioneered the convergence between LANs and WANS.
We all know the future of enterprises is in flux with the consolidation or demise of Avaya/Nortel, Brocade/Foundry, Force 10, H3C and BNT assets. This speaks to the major trend of enterprise applications migrating to the cloud, any cloud be it public, private or hybrid. As this rapid transition prevails with efficiencies and scale, traditional applications and tools cannot meet the needs of exponential cloud growth.
Today’s applications are connected both to users and other applications, increasing traffic and profoundly affecting performance. Recently, there has been much discussion around the Application Performance Management (APM) category. This got me thinking about new cloud migration versus legacy application tools. Is the industry being myopic and dealing with silos?
Networking is running blind. It is akin to driving a car in the fog at night without street lights, signs or a navigation system. Simply put, it’s a scary visibility problem, and it impacts the security and availability of the network. When coupled with massive shifts to virtualization, containerization, cloud-native applications and unstructured data, the insatiable telemetry demand is exponential. Every architect is looking for modern analytical methods of networking to gain visibility for millions of devices, data or events efficiently and consistently.
In the 1950s, the introduction of jet engine technology revolutionized air transport, allowing for the transport of goods and people across oceans far more quickly than with traditional propeller-based aircraft. A new airline industry on a global basis arose from the jet engine revolution.
In the early 2000’s a new generation of smartphones revolutionized the cell phone industry, eliminating the market for “flip phones,” introducing new tools and completely redefining “phones” to universal internet devices. New companies rose and old ones adapted or failed. In 2015, a new generation of electric cars (Tesla being the most well-known), were introduced and drove the next wave of technology transitions, replacing gas-guzzling cars in many households. Consumers and traditional car companies worldwide are now embracing the new technology.
We have witnessed a massive shift in the way applications are built and deployed, moving away from siloed IT to seamless workloads, workflows and work-streams. This revolution has changed the fundamental way that networks are architected to enable support of cloud native applications. With these new architectures, one can now automate and provision the entire network with real time agility, ushering in a new class of cloud networks.
The rapid migration from enterprise to cloud, driven by the economics of scale, the convergence of local and wide-area networking (LAN-WAN), the migration from Fibre Channel to IP storage, the rise of analytics and the emergence of new cloud applications is dramatic. In the past two years, we have witnessed a massive shift in the way applications are built and deployed, moving away from legacy siloed infrastructure to seamless workload mobility. The demands of these new workloads change the way spine networks are reconstructed for cloud networking. As physical compute or storage silos evolve to support cloud applications, one can automate and provision the entire network to handle any workload, workflow or workstream, with real time agility.
It's tough times on Tasman Drive. Struggling to apply old technology to the new world of cloud computing, Cisco is potentially facing the largest loss of data center market share in its history. We can understand why Cisco would take the battle from the marketplace to the courtroom. What surprises us is the length that Cisco has gone to misrepresent our actions and the nature of the litigation in order to justify their assault.
As we celebrate 2016, and reflect upon Arista’s decade of innovation, the disparity between legacy technology and modern technology is clear, and the gap is widening. Arista is unique in its role as a pioneer in cloud networking. Let us review.
We are very pleased to report a major milestone for Arista Networks. According to the just released Q4’2014 Crehan Data Center Switch Market report, in the high-speed Ethernet switch market (defined as 10G Ethernet and above excluding blade switches) Arista’s port market share in 2014 rose to 9.3%, up from 6.7% in 2013. Our revenue market share for 2014 rose to 7.7% from 5.4% in 2013. All of this speaks volumes to the rapidly growing acceptance of our products by customers around the globe.
In the meantime, according to the same Crehan research report, Cisco’s port market share fell from 70.8% in 2013 to 66.1% in 2014, a loss of 4.7%. Cisco’s revenue market share declined from 74.2% in 2013 to 69.3% in 2014, a loss of 4.9% YoY. Again this speaks volumes to the decisions customers are making.
What is going on here? At the high level, it is very simple. Customers are transitioning away from legacy proprietary networking solutions and are embracing Arista’s cloud network architecture, a fully standards based solution that offer unparalleled scale, availability and open programmability.