As a young Product Marketing Engineer at Intel, I found myself living by Moore’s Law dutifully bringing out new microprocessors that doubled compute power every two years.
Some years later I found myself at 3Com living by its corollary - Metcalfe’s Law - where we madly went about harnessing this raw compute power in client-server networks to increase the business value of collaborative work and communications.
Fast forward to 2016 and we see hyper scale giants like Amazon, Google and Facebook building out cloud scale data center networks that were only dreamed of in these early days of exponential compute and networking growth predictions. Here’s Delphina Eberly, Facebook VP Infrastructure Operations, discussing their state-of-the-art Facebook data center in Altoona, Iowa and Alexey Andreyev, a Facebook network engineer sharing his now famous data center fabric blog explaining the simplified standards-based IP Fabric design and DevOps style automation used to build the Altoona facility. This blueprint has become the model for enterprises and clouds service providers alike as they strive to implement their own modern data centers leveraging these best practices.
As for me, I am now at Brocade working on the latest data center building blocks for our customers embracing this digital transformation connecting video, IoT and cloud networked services. One such customer, Servers.com, explains here why a simplified IP Fabric design, similar to the one used at Facebook, was exactly what they needed to address their own customer demands for cloud scale and agility.
Most recently we just added a new highly scalable SLX 9850 Router to our Brocade IP Fabric portfolio of platforms. This future-ready platform brings new levels of performance, scale, automation and investment protection to Brocade customer data center designs preparing them for the unpredictable but surely exciting road ahead.
So how is your network scale-ability factor? Are you ready for the next wave of data, devices and users? We will be talking more about agile and cost-effective scale with fabrics and foundational building blocks in the weeks to come.
For my twenty or so years in computer networking, the promise of a network that is sensitive to the applications running on it has been put forth almost continuously. And to some extent, constructs such as Quality of Service (QoS), policy-based routing, and even simple logical domain separation through subnetworks and VLANs have provided the support to improve application performance.
As part of the digital transformation, however, applications are becoming much more complex, and traffic growth is increasing tremendously. Data center applications are composed of many components, which creates horizontal traffic (east-west traffic) within the data center. Furthermore, microservices have become more important across application development.
Even as data center networks scale to support this growth, operators of these networks need new tools to predict and adjust to traffic patterns with flows.
So questions as to how to:
Are more critical than ever.
What is Needed for Optimized DC Applications in the Digital Era?
A key requirement is visibility. A few weeks back, Sanjay Khanna wrote a very popular blog called I Can See (My Network) Clearly Now, which illustrated the value of visibility. Brocade executives Jason Nolet and Dan DeBacker have taken the discussion a step further in a new video called Optimize Application Performance with Advanced Brocade Network Visibility.
The latter video describes how Brocade worked with many accounts across network, security, and application teams to understand the direction that application development and architecture would take over time. We discovered that applications have largely become disaggregated with the advent of microservices and componentized, distributed development.
Accordingly, customers need to know about their traffic flows, and how much traffic traverses between application components, servers, and other networks. Customers typically do not know this and they can benefit hugely by finding out.
How Do We Gain This Visibility?
A visibility architecture, such as the SLX Insight Architecture (Figure 1), helps operators understand what the nature of the traffic between the components that make up an application.
Figure 1: SLX Insight Architecture
The key components and points are as follows:
The visibility data can then be sent to analytics engines of your choice (such as Splunk, New Relic, or Data Dog) for processing. These applications, and others, can also optionally be on-box in the Guest VM, although if a large scale-out is required it may typically make more sense to keep them physically separate.
How Can We Optimize Applications with This Information?
Now, if we take this a step further, we can look at Brocade Workflow Composer, powered by StackStorm technology, as a way to add DevOps-style automation to the mix (Figure 2).
Figure 2: Brocade Workflow Composer, Powered by StackStorm Technology
In addition to remediation and troubleshooting, Brocade Workflow Composer can used to reconfigure the network through Netconf interfaces to configure IP routing and thus redirect traffic flows.
The many approaches to data center design strategy can be confusing. But, essentially there are 2 approaches – a vertically integrated complete stack solution or build your own stack. There is a parallel here with bring-your-own-device and bring-your-own-app.Read more...
The Product Management team in Brocade’s Switching, Routing and Analytics Business Unit is shipping the 230Tbps-capable SLX 9850 router to NANOG 68 in Dallas (October 17-19). It will be presented at NANOG’s “Beer and Gear” event on Tuesday afternoon, October 18 at 6:00PM.
This high-end router is especially suited to IP fabric and core-aggregation applications in the spine and super-spine layers of large data centers. As discussed by the CTO of the Amsterdam Internet Exchange in this video, the SLX 9850 has industry-leading 100GE density.
Brocade will demonstrate the extensible architecture and unique 1.5 RU interface module design. The interface densities of the current modules are 288 × 100 GbE, 480 x 40GE, and 1920 x 10GE.
We will also illustrate the management modules and switch fabric cards, and even the front-to-back cooling system. Furthermore, we’ll have experts on hand to discuss the SLX Insight Architecture (discussed in this video) for network visibility, as well as our DevOps-style automation solution, the Brocade Workflow Composer, powered by the event-driven StackStorm technology.
Registration for NANOG 68 is still available! You can find the agenda here and the current attendee list as well, so you can see if partners or customers of your organization will be present. If you decide to attend you will also want to attend Rohit Bothra’s presentation entitled “BGP: The Highway of the Internet” at 9:30AM.
If you’ve wanted to get more familiar with the SLX 9850, including the interface and management modules, switch fabric cards, even the cooling system, this is your chance to do so. Here’s a shot of the chassis we’ll be showing, prepped for travel in its shipping crate.
Dressed for Success
Stay tuned to follow the SLX 9850’s travels—it will be headed to Supercomputing 16 next.
Jay Etchings has a key mission as director of research operations at Arizona State University – make sure his customers don’t know about the network.
“From a researcher’s point of view—a computational biologist or a genomicist—the network is not important unless it ceases to function,” say Mr. Etchings in this video detailing ASU’s use of this technology.
Figure 1: Brocade Solution for ASU’s Research Network and Science DMZ
This solution uses MLXe routers, the Brocade SDN Controller and of course the Brocade Flow Optimizer. Providing high availability and resiliency, and supporting IPv4, IPv6 and OpenFlow 1.3, the network serves as a scalable, high-performance solution with a Layer 2 firewall bypass.
In particular, the high-speed interfaces (100GE) into the DMZ are ideal for the cutting-edge research that ASU supports. Etchings notes that this performance allows ASU to pursue and support the most impactful global research: “Unlike competing institutions that are focused on papers and theoretical components, we’re actually doing these things, and Brocade is instrumental in helping us do them.”
“Cybersecurity is one of the leading disciplines these days, so I can identify new problems and new ideas almost every day,” notes Dr. Gail J Ahn, Ph.D., CISSP and professor of computer science and engineering. Sometimes these problems may be security related, and other times they may just be poor performance.
But with the new network, Etchings said his team can “take some development that Dr. Ahn put into production, observe the latency, and if necessary remove that flow.” In the past, he notes, the department would have had to put that flow into production without examining it, and it would have been more difficult to back out.
The new process results in less maintenance and less downtime—exactly the way to keep a happy team of researchers, who are always too busy to think about the network.
For more detail, see the written success story on this network.
Here are a couple of options for how you can get orchestration and visibility across container networks and physical networks:Read more...
Have you ever thought about what types of cars are passing through an intersection, whether there are any accidents or if cars are ignoring traffic lights? Operators use cameras and sensors to get visbility into traffic intersections. Do you have the same visibility into your network? ...
Looking back on the changes of the last two decades, we can see that networks were at the center of the disruption that swept through every industry twenty years ago with the introduction of the Internet...Read more...
We had a great afternoon with the NFD team as we walked them through our vision, design principles and demos of the Brocade Workflow Composer... here’s what happened:
It’s August and that means it’s time for VMworld 2016 in Las Vegas! This is THE destination to find all the sexy things you’ve been peeking at online and just never get the chance to sample in real life. And there’s also Vegas entertainment if that’s your thing instead…Read more...
By leveraging SDDC, CA Department of Water Resources is extending their data center to the cloud to enhance the overall customer experience. Learn more here:Read more...
Network automation can’t be addressed with “protocols” alone! The missing piece in network automation puzzle is a software framework that addresses the inefficiencies in today’s model with the following characteristics:Read more...
What must networks deliver to deploy application containers with enterprise-class reliability? How can network fabrics provide the best infrastructure for containers in production?Read more...
The network is undergoing historic transformation with overlays, NFV-based services and SDN control. To leverage the full value of these technologies, a new automation strategy must be front and center with the network operations team...
Why are public and private organizations still readily being hacked or breach? Simple. Security isn’t. Many organizations focus on securing data being stored, but not data in-flight within or across campuses or data centers. For hackers, data in-flight is great. They don’t have to go looking for the data, the data comes to them.Read more...
As optical networking becomes more and more prevalent and as the speeds have increased the importance of creating and maintaining a clean optical infrastructure has become paramount.Read more...
In 2010, Brocade first announced and launched the solar-power project for the company data center (as well as other major green initiatives). I am one who is quite often in awe of these technology marvels and especially regarding the story of what actually goes on behind the curtains. Now that a few years have elapsed and we are coming up on the half way point of 2014, I thought it would nice for a revisit.Read more...
As this market gains more adoption, operation agility and performance must be a part of the future of the virtualized desktops experience.Read more...