5 Reasons to Attend Cumulus Linux Boot Camp

Sure, you’re smart and can figure out Cumulus Linux on your own. Or maybe you can’t spare 8 hours for training. If you don’t think you have the time or can benefit from attending the Cumulus Linux Boot Camp as you get started with open networking, here are five good reasons to think again.

1. Save Time

Yes, you can actually SAVE time deploying and managing your network by investing time in training. Our Cumulus Linux Boot Camp is an instructor-led, hands-on training course for networking and sysadmins, optimized to get you going quickly with open networking using Cumulus Linux. You could read the documentation and try to figure things out on your own, but why not maximize your time by leveraging a course where the essential information is gathered in a single place, presented in a methodical order?

2. Access Experts

Our class is taught by technical trainers and consultants who are well-versed with open networking. It’s a great venue to get answers to your questions.

3. Get Stick Time on Switches

This course includes hands-on labs using switches in our Cumulus Workbench. You’ll run through practical exercises and be provided with lab answers and sample code to increase Continue reading

The Magnificent 7

Our Cumulus Networks team is very excited that Supermicro has joined our Open Hardware partner program, the latest major IT systems provider to join the industry-wide open networking movement.

Now there are Seven. 

Supermicro is a leading innovator in high-performance, high-efficiency server, blade, storage, and networking technology for Green Computing – worldwide. Cumulus Linux on Supermicro bare-metal switches further extends the reach of the Supermicro solutions, enabling rapid deployment of a highly scalable, cost effective software-defined network infrastructure for data center, cloud, enterprise IT, big data and HPC.

As our seventh Open Hardware partner, Supermicro is now part of a very impressive list of providers on the Cumulus Linux HCL: Agema, Dell, Edge-Core, HP, Penguin, Quanta Cloud Technology (QCT), and Supermicro.

What does it mean for the industry? Open Networking is inevitable and Cumulus Networks is leading the way.

Major changes are underway in the IT industry that improve data center networking, allowing organizations of all sizes to leverage efficient technology that was developed by the world’s largest cloud operators. The resulting data center networks scale more easily, enable much faster innovation, and cost significantly less to build and operate. With data center infrastructure leaders like Supermicro embracing Continue reading

Managing your Out-of-Band Network with Linux

Initially when I was asked to blog about out-of-band management I thought to myself, as most people would, “this is too basic!”  What new thing could I cover?  Generally speaking, out-of-band management, like management in general, is an afterthought.  With that typical attitude, we make the mistake of placing low value on access to our network devices, seeing it as a simple back door when in reality it could provide so much more.

The idea of creating the Cumulus® RMP (Rack Management Platform) came about after talking to several customers whose approach was to purchase low-end switching platforms to meet their out-of-band management needs.  These closed network platforms provide such limited feature sets that it’s easy to dismiss their usefulness.  The team sat down and came up with the idea to “complete the rack.” Why not provide the same open networking capabilities that Linux servers and Cumulus® Linux® switches offer for out-of-band management? Thus Cumulus RMP was created.

Typical Deployment Scenarios

In general there are two basic scenarios when it comes to out-of-band management.  The first provides a simple but versatile L2 flat design leveraging VLANs to manage the switches and servers in the rack.  The Cumulus RMPs Continue reading

Open Networking: Worth the Learning Curve

When’s the last time you had the opportunity to learn about a new area of technology? Not just the latest API but something big. With the emergence of open networking, now may be the time to do just that.

Open networking is becoming more wide spread in data centers, partly because of CapEx savings and customer choice from disaggregation of hardware and software. From an operational standpoint, the efficiencies of automating and managing switches like servers may be even more compelling. Those efficiencies – and resulting OpEx savings – are from using a real Linux OS for the network.

As a Linux operating system running network switches, Cumulus Linux presents a learning opportunity for two specific groups of users: network administrators who are new to Linux, and Linux administrators who are new to networking. Open networking represents a radical improvement in network management, but as with most major changes, there’s a learning curve.

Learning a new language (but still saying the same things).

Leaving IOS, NX-OS, or JunOS for Cumulus Linux means rediscovering how to perform tasks that network administrators have been doing the same way for years. This often means tweaking existing knowledge to understand where familiar commands Continue reading

Feeling the Love from our IT Brethren

I had the honor of accepting two awards yesterday on behalf of the Cumulus Networks team. We’re pleased because the honors were awarded based on a broad poll of real people in real IT shops, including some of the world’s biggest enterprises and service providers. You can’t buy these; we didn’t even know that the poll was being taken. Those voters, who we think of as our IT brethren, chose us for “Bare Metal Switch OS Market Leader,” for one of only two “Special Achievement” awards (together with our partner, VMware), and three other accolades.

“Well, how did I get here?”

When we founded Cumulus Networks, we were driven by a vision of how high-capacity interconnect would change modern applications.

We set out to make great networking technology available to the masses, addressing three critical needs: affordability, an efficient operating model, and availability via a variety of channels. These tenants have become the definition of open networking.

To fulfill these goals, we built the company with IT professionals from all disciplines: development, operations, support and logistics. Continue reading

Bringing Enterprise Class Automation to Open Networking

Today Puppet Labs announced that Cumulus Networks has joined its Puppet Supported Program. We’re very excited about this and, if you’re implementing a software-defined data center, you should be excited too.

Is it finally possible to manage the data center instead of just managing stacks?

Because Cumulus Linux is Linux, our customers are able to use the same tools they know and love for managing Linux servers to manage their networks. The joint integration work we’ve done means it’s easier than ever for anyone that wants to automate their data center to extend their change management procedures across both servers and switches, unifying data center and network infrastructure under a single dashboard.

Beyond the streamlining of management consoles, this integration brings a host of business benefits to any organization. For example:

• Businesses can more quickly deploy applications with integrated, end-to-end application deployment from provisioning the VM to the full-stack (see the diagram below)
• Unified management and automation significantly reduces human error, meaning more uptime for services and applications
• Support for multiple Puppet masters across dev, test and production assures full visibility of change management
• Integration gives networking teams the ability to contribute to the Puppet code that manages infrastructure configuration

Continue reading

OCP Summit 2015: Open Networking as the Norm

Open Networking Thrives at the 2015 OCP Summit

A few weeks ago I passed my 2 year work anniversary at Cumulus Networks. In this short amount of time we’ve helped change the networking industry as a whole. This statement may seem a bit bold, but I feel it’s justified and we’ve all earned it. I’ll attempt to back this up with a few anecdotes collected along the journey — the men and women of Cumulus Networks, along with our great partners and customers, deserve the credit here.

When I was first hired, our on-boarding process was a bit … unconventional to say the least. The first week included JR (our CEO and cofounder) taking a headshot photo for the website (at that time it was just a Brady Bunch collage of pictures, not much else) and being issued a switch for “jailbreaking.” Now jailbreaking is a bit of an extreme term, but in these dark days before open networking, the software and hardware for networks were tightly coupled together. Items such as debugging headers, a soldering iron, and even a bit of swearing were all part of the installation process on a fresh hardware platform. Fast forward to today, Continue reading

Arista Jumps into Open Networking — or Do They?

On March 31 Arista announced it will offer its EOS operating system as a subscription license separate from its switch hardware.  While the rest of the industry is realizing disaggregation of hardware and software for the sake of customer choice and innovation, Arista’s “disaggregation” is only a new pricing model. Arista hardware and its EOS operating system remain locked. Customer choice is limited to pricing models.

That is not open networking. And it’s not what customers need. This is an open pricing model at the best.

Currently Arista sells its hardware and software as a single bundle, but with this new pricing model, Arista claims it gives customers a better way to balance their CapEx and OpEx budgets. Arista also says this will let customers scale their cloud deployments as they want, paying for their network resources only as they consume them and — given the subscription service — helping them avoid exorbitant upfront investments.

Disaggregation?

Is Arista opening their pricing model because of growing industry support for open networking? Their announcement nods towards a “disaggregated offering,” but EOS still requires Arista hardware, and Arista switches still require EOS. Customers can buy the hardware/software components separately, but they still can’t Continue reading

Cumulus Linux Switch Monitoring with Datadog

As a Linux platform, one of the cool things is that we often don’t have visibility into how customers use their switches running Cumulus Linux. They buy HCL-compatible hardware from our partners, and with some training and enablement, are off to the races.

The idea for running Datadog in Cumulus Linux came about for the simple reason that we were in adjacent booths at PuppetConf last year, and we all figured it would be cool to try it out. Further, since Datadog already provides visibility across systems, apps and services, they were interested in seeing how networking can be added into the mix. As you will see, it turns out to be pretty simple.

Installing Datadog in Cumulus Linux

The Datadog agent, as with most things Debian, installs easily on Cumulus Linux. For x86 switches, this is as simple as installing a Debian package and performing simple changes in the Datadog agent files, such as the application/API key, which is the tag associated with the switch. You can easily automate this installation using common automation tools like Puppet and Ansible.

Configuration

Since the Datadog agent is designed for servers, metrics can be collected using Datadog’s SNMP plugin, a custom sFlow Continue reading

The Facts Behind the Myth

I hate getting into lengthy discussions regarding open networking (or bare-metal) pricing as there are benefits other than price. However, with so many people trying to understand the industry transition, I feel compelled to jump in when I see confusing information.

Forrester analyst Andre Kindness recently published a report called The Myth of White-Box Network Switches which is causing a pretty interesting debate.  The discourse forms, as Andre puts it, “I think there is misunderstanding/reading my research. I’m not saying one solution is cheaper. It highlights the cost”; however, the title of the report necessarily creates bias. Luckily, we had a chance to speak with Andre to better understand his perspective and intentions as well as relay our observations.

For the hardcore, we’ve gone through some of the market basics in prior blog posts; most notably Democratizing Capacity and Death of the Multipler Effect.  Some of the absolute numbers in those analyses have changed; however, these hold true and directly relate to both the points that Andre was trying to make as well as the gaps in his analysis.

Bill of Materials Cost

The report makes two observations that we completely agree with. One is that, most of Continue reading

Cumulus Networks, sFlow and data center automation

The Cumulus Linux platform makes it possible to run the same open source agent on switches, servers, and hypervisors – providing unified end-to-end visibility across the data center. The open networking model that Cumulus is pioneering offers exciting opportunities. Cumulus Linux allows popular open source server orchestration tools to also manage the network, and the combination of real-time, data center wide analytics with orchestration make it possible to create self-optimizing data centers.

Install and configure Host sFlow agent

The following command installs the Host sFlow agent on a Cumulus Linux switch:

sudo apt-get install hsflowd

Note: Network managers may find this command odd since it is usually not possible to install third party software on switch hardware. However, what is even more radical is that Cumulus Linux allows users to download source Continue reading

Cumulus Networks Support — Community Style

Sometimes, it’s not what you know, it’s who you know. Today we launch a new method of support to the Cumulus Networks family, a community question and answer site. This is a place where you can ask either a simple question for which you couldn’t find an answer, or maybe something you’ve always wanted to know. It’s also a place to engage in conversation with other users that may be experiencing things you have seen in the past or may encounter in the future.

To be sure, Cumulus Networks employees will be on hand to assist and direct you to the documentation and knowledge base as appropriate. If a question is too complex, we will assess your needs based upon your support entitlements and work with you to open a ticket so our excellent support team can assist you in figuring out your issue. The community portal is a supplement to the support team, not a replacement.

Our growth is directly related to you, our community, and in that vein, we want to offer more ways that you can be involved — this is just the beginning. Look for more to come in upcoming blog posts.

The post Cumulus Networks Continue reading

Open Hardware that Just Runs

When you buy a server, you don’t worry whether or not Windows will run on the server.  You know it will. That’s because the server industry has a comprehensive solution to a hard problem: rapid, standard integration between the OS and underlying open hardware.  They’ve made it ubiquitous and totally transparent to you.

This is not the case for embedded systems, where you have to check whether an OS works on a particular hardware platform, and oftentimes you find out that it’s not supported yet. Bare metal switches are a good example of this.

It’s time to change that.  We need the same transparent model on switches that we have on servers.  

To make open networking ubiquitous — and to give customers choice among a wide variety of designs, port configurations and manufacturers — the integration between the networking OS and bare metal networking hardware must be standardized, fast, and easy to validate. We need open hardware.

The Starting Point

Today, a bare-metal networking hardware vendor supplies their hardware spec to the NOS (network OS) provider. The NOS team reads the spec, interprets it and writes drivers/scripts to manage the device components (sensors, LEDs, fans and so forth). Then Continue reading

Striking Gold with Copper in the Data Center

10G Ethernet (10GbE) is a very popular interconnect technology in today’s data center. It’s widely used for servers/storage devices connecting to top-of-rack (leaf) switches, as well as connecting those leaves to the aggregation (spine) switches. According to a Dell’Oro Ethernet Switch Market report published in 2014, 80% of server connections will be 10GbE-based by 2018.

In general, two types of physical media are used in 10GbE interconnects: fiber and copper. For intra-rack (server/storage to leaf switch) connections, most deployments use copper cabling as it is the most cost effective for short distances within the rack.

There are two copper cable types: twinax and twisted pair. Twinax is used in the 10GBASE-CR standard in the DAC (direct attached cable) format, which is a fixed length cable with SFP+ plugs integrated into both ends. Twisted pair, on the other hand, is something that should be very familiar to every IT person. Remember CAT cables and RJ-45? The 10GbE interconnect standard that uses twisted pair is 10GBASE-T, which is officially defined in the IEEE 802.3an standard.

10GBASE-CR with DAC is great and used in many deployments. However 10GBASE-T over twisted pair offers some unique benefits:

Distance & Interoperability

10GBASE-T over twisted pair Continue reading

SCALE13x – My talk: Switch as a Server

This past weekend, I had the opportunity to speak at SCALE13x in Los Angeles, on the Switch as a Server — treating your network switches in the same way you treat your servers.  It’s a topic I feel very strongly about!

As strong as my feelings are about open networking, I also love non-automotive forms of transportation!  So I decided to bike to the airport.  SFO has a lot of bicycle facilities so it was no problem to find parking.

Got to LA on the plane and then Rocket Turtle enjoyed the view by the airport!

… and met some of our great customers!

Scale is a unique conference in that they encourage canine attendance  — doggies!

 Friday night I helped out with a birds of a feather (BOF) event, giving advice to job hunters.  Did I mention we’re hiring?

On Saturday evening I won the Weakest Geek — a Weakest Link-style geek-themed trivia contest, run this Continue reading

Introducing Redistribute Neighbor

The existing landscape

Are we feeding an L2 addiction?

One of the fundamental challenges in any network is placement and management of the boundary between switched (L2) and routed (L3) fabrics. Very large L2 environments tend to be brittle, difficult to troubleshoot and difficult to scale. With the availability of modern commodity switching ASICs that can switch or route at similar speeds/latency, smaller L3 domains become easier to justify.

There is a recent strong trend towards reducing the scale of L2 in the data center and instead using routed fabrics, especially in very large scale environments.

However, L2 environments are typically well understood by network/server operations staff and application developers, which has slowed adoption of pure L3-based fabrics. L3 designs also have some other usability challenges that need to be mitigated.

This is why the L2 over L3 (AKA “overlay” SDN) techniques are drawing interest; they allow admins to keep provisioning how they’re used to. But maybe we’re just feeding an addiction?

Mark Burgess recently wrote a blog post exploring in depth how we got here and offering some longer term strategic visions. It’s a great read, I highly encourage taking a look.

But taking a step back, let’s explore Continue reading

Network Infrastructure at Faithlife

The network infrastructure powering Faithlife has seen a massive transformation in the last eighteen months. We’re really excited about all the cool new changes, and the measurable impact they’ve had on our employees, customers, and the products / features we’re able to offer. Given that, we thought that sharing our solution was a fun way to live our values and showcase what we think is pretty cool.

Philosophy

At Faithlife we value smart, versatile learners and automation over expensive vendor solutions. Smart, versatile learners don’t lose value when technology changes or the company changes direction, as vendor solutions often do. If we can use commodity hardware and open source software to replace expensive vendor solutions, we do.

Commodity hardware is generally re-configurable and reusable, and lets us treat our hardware like Lego bricks. Open source software allows us to see behind the curtain, and more easily work with other existing tools. We’re empowered to fix our own issues by utilizing the talent we already employ, not just sit on our hands waiting for a vendor support engineer to help us out (though we do like to keep that option available when possible). Additionally, combining commodity hardware with automation tools like Continue reading

Guest Blog: REST API for Cumulus Linux ACLs

RESTful control of Cumulus Linux ACLs included a proof of concept script that demonstrated how to remotely control iptables entries in Cumulus Linux.  Cumulus Linux in turn converts the standard Linux iptables rules into the hardware ACLs implemented by merchant silicon switch ASICs to deliver line rate filtering.

Previous blog posts demonstrated how remote control of Cumulus Linux ACLs can be used for DDoS mitigationand Large “Elephant” flow marking.

A more advanced version of the script is now available on GitHub

The new script adds the following features:

1. It now runs as a daemon.
2. Exceptions generated by cl-acltool are caught and handled
3. Rules are compiled asynchronously, reducing response time of REST calls
4. Updates are batched, supporting hundreds of operations per second

The script doesn’t provide any security, which may be acceptable if access to the REST API is limited to the management port, but is generally unacceptable for production deployments.

Fortunately, Cumulus Linux is a open Linux distribution that allows additional software components to be installed. Rather than being forced to add authentication and encryption to the script, it is possible to install additional software and leverage the capabilities of a mature web server such as Apache. The Continue reading

Hyper-convergence and Open Networking: A Match Made in Cloudy Heaven

A lot of the early hype around cloud computing focused on grand visions related to there being only 5 or 6 extremely large cloud providers across the globe. While public clouds continue to grow at a breakneck pace, private clouds are also starting to see immense traction, especially in key verticals like financials, SaaS providers, and telecom service providers.

Over time and through extensive trial and error, the marketplace is realizing that there are two key requirements for successfully implementing cloud computing:

• Simplicity: This primarily refers to breaking down silos that have plagued IT departments of all sizes, allowing for a unified framework across compute, storage and networking.
• Infrastructure automation: Ranging from automated provisioning to full lifecycle of infrastructure, implemented in a software defined manner. Often referred to as Infrastructure as Code, or Idempotent IT.

Simplicity and infrastructure automation have been extensively covered by leading IT analysts and, along with application-level paradigms like Hadoop, have often been referenced as the way to achieve the extraordinary scale and success of Web scale IT shops like Google, Facebook and Amazon.

But until now, having the entire set of components and knowing how to assemble and automate them effectively still required open Continue reading

The cloud journey continues: HP and Cumulus Networks

Is open networking mainstream?

If the HP announcement of new data center networking solutions with Cumulus Linux is any indication, then we will venture to answer.

Yes.

On Feb. 19, HP announced a partnership with Cumulus Networks for new HP open network switches with Cumulus Linux, the operating system for open networking. HP has designed an efficient supply chain model including a joint venture with Accton, delivery worldwide via HP logistics centers, and HP local sales and support.

Open networking was born when web-scale providers developed proprietary cloud networking that reduced CapEx costs while improving OpEx and enabling automation.  Similarly, hosting companies, service providers and high-tech firms re-architected their data centers realizing similar benefits.  Next, financials, government, and education entities found affordable capacity and new innovation opportunities with their cloud journey.  Today, a broad range of enterprises are adopting open networking for a wide set of use cases – and for automation and rapid service delivery using Linux tools already standardized for servers.

In that context, the new HP open network switch offerings with Cumulus Linux address a range of use cases but the sweet spot is hyper-scale data center operators. Of course HP also offers the servers, storage and Continue reading