The cloud era has been anything but simple for businesses. In fact, a lot of the challenges they face may be attributed to how hard it is for network operations teams to consistently provide reliable connectivity among their employees, assets, applications, and services.
So the opportunity is to simplify end-to-end reliable WAN for businesses, large and small. That's why we were at ONUG in NYC (ok, were there for the bagels also). We saw an amazing presentation by Steve Garson of SD-WAN Experts, titled "Measuring Internet Core Variability."
We've been blogging about the Internet, how it's generally great, but perhaps problematic when it's used as part of an SD-WAN solution for mission-critical business applications.
We've told you about the "good parts" version of the Internet — separating Access from Core. And we've implicated the Core as the primary cause of unpredictability.
We are always amazed when others produce corroborative evidence. At ONUG, others = Steve Garson of SD-WAN Experts.
Let's take a look at the first part of Steve's presentation.
Steve chose response time (i.e. send + wait) as the ideal methodology, because it excludes one-time events like DNS and connect, and it's a better measure of real Internet response than ping.
In his first series of tests, Steve calculated Core performance by subtracting access performance from a long-haul performance. Of course, performance varies over time. I'm no mathematician, but I seem to remember that in cases of two parts of a series with variability, the variance of the whole (variance = standard deviation, squared) is equal to the sum of the variance of each part, plus two times the covariance (the correlation of variance in part one with part two).
I think it's a safe assumption that, in general, performance variance of the last-mile is wholly independent of core variance, meaning covariance = 0. Steve makes this assumption, which means the long-haul variance = core variance + access variance. Whew. We are done with the math.
Steve collected data from servers originating in San Jose, London, Tokyo, Sydney, and Virginia. He looked at the performance to end users in Bangalore, Washington D.C., Tokyo, London, Melbourne, and San Francisco. Here is a table of the raw data:
The results are pretty clear. For long hauls, the vast majority of response variance occurred in the Internet Core, not the last-mile. In fact, 99.5% of response variance happened in the Internet Core. That means that your business traffic, over long hauls, is experiencing the vast majority of dropped packets, jitter, etc., as a result of the Internet Core. Even with the lovely software-defined benefits of SD-WAN at your corporate edge.
99.5% of long haul Variance happens in the Internet Core. Why?
It's largely an issue of economics. Internet Access networks receive 300% to 1000% the investment of the Internet Core, most of it coming from customers. The Core is based on least-cost peering and routing.
Steve went on to provide additional tests of Internet Core performance, and we will cover them in upcoming blogs.
But for now, we know one thing for sure. If you want to simplify reliable end-to-end WAN so your businesses can hum, you can't stop at SD-WAN. You need to replace the Internet Core with something far more reliable. You could use something overpriced and rigid like MPLS. But you really want something that's SD-WAN friendly, flexible, and affordable. You may not know it yet, but your want a software-defined core (SD-CORE).
Jennifer English at TechTarget has done a nice job reviewing Steve's presentation. Her conclusion — you may not be able to rely on SD-WAN + Internet, but you no longer have to pay for expensive solutions like MPLS. SD-CORE might be just the thing for simplifying the reliability of your SD-WAN.
Who doesn't love The Princess Bride? I got my first copy at the Dartmouth bookstore outside of summer camp (for those who've been reading this blog, you'll recognize the tie-in to the very first Mode post). No, the movie isn't nearly as good.
It was a tight book. A real page-turner. In fact, this masterpiece by William Goldman was an abridged version of the hideously plotted, unedited, and unreadable original version, penned by S. Morgenstern. Or so Goldman claimed. In fact, it was a clever, but bogus, literary device. Now I digress.
So what do R.O.U.S. and "as you wish" have to do with this next part in our blog series of "Is the Internet good enough for mission-critical business?" Quite a bit, actually. Or at least one bit.
In our last blog, we showed evidence that the Internet was too unpredictable to support mission-critical business applications (or other applications like streaming gaming, and a whole host of next-generation devices, applications, protocols, and services — from IoT to blockchain to 5G).
We must ask: do we dismiss the entire Internet on this basis? Or is there a "good parts version" of the Internet that we can carefully separate, and use with confidence as part of an end-to-end cloud connectivity solution? (I know you were waiting for that tie-in, and yes, you're welcome).
It's common practice to divide the Internet into two parts — the on and off ramps we call Access. The center, we call Core. Access has different names like last mile (or first mile), and Core is sometimes called backbone or middle mile. So can we attribute the unpredictability of end-to-end Internet entirely to either Access or Core? Is it really that easy?
It turns out that we can, for the most part. Internet Access outages are exceedingly rare. Physical network distances at the last-mile are relatively short, and global, long-distance studies show that Access contributions to latency and jitter are relatively insignificant vs. end-to-end values. Solutions like SD-WAN offer additional Access resiliency via redundancy. Additional studies of last-mile U.S. operators reinforce this truth: Internet Access is predictable, can be trusted, and is getting better all the time.
This makes sense when you consider how well-capitalized and well-marketed Access solutions are. It's a highly competitive marketplace, and poor performance is a deal-breaker.
The Internet Core is an entirely different story. It's not nearly as visible, not heavily marketing, and is composed of complex peering arrangements that shift continually. Its "shortest path" priorities are often economy, not performance. Third-party tests show that the vast majority of latency variation (jitter) happens in the Internet Core, as do daily globally-dispersed Core outages that have not gotten better over time.
This leaves us with a remarkable conclusion, one our SD-WAN partners have known for some time: the "good parts version" of the Internet is Internet Access. It is the Internet Core than requires a more predictable, higher-performance alternative.
SD-WAN has led the way by bringing software-based control the edge of the enterprise WAN. What if we adopted this approach, and brought software control to the Core? That would be remarkable. But how?
This isn't a Princess Bride-style "battle of wits." We know the answer, and we'll happily tell you in our next blog installment.
We aren't alone in our search for an Internet Core alternative that's SD-WAN-centric. Steve Garson @ SD-WAN Experts just published an article detailing various ways to take the goodness of SD-WAN and extend it end-to-end across the WAN core. You can read Steve's article here.
There was no Bogart. No tearful goodbyes. But the Mode team left Paris with something better — confidence.
Not just our confidence, but yours. You, the CIO, the VP Network Operations. For some time, you've feared the transition to the cloud.
You've woken up at night in a sweat. And that fear has made you cling to the past. To rigid, expensive, and decidedly cloud-unfriendly connectivity solutions like MPLS.
We left Paris, and we gave you the confidence to let go. The confidence that the cloud could deliver both where MPLS succeeds (reliability and QoS), and where it falls short (flexibility and economy).
We didn't start the ball rolling. That was SD-WAN. It gave you confidence in the edge of your network, with visibility, and the flexibility to chose among different networks. It made site-to-site connectivity a snap. And it did it without breaking the bank.
Still, you knew that at its core, it was still using the Internet. And that made you nervous. And for good reason. The Internet Core is anything but reliable. At least SD-WAN let you pick and choose, and that's exactly what you did, using MPLS like the preciously expensive resource that it is.
SD-CORE sounds a lot like SD-WAN, and that's not an accident. Where SD-WAN brings software-defined control to the edge of the enterprise WAN, SD-CORE takes it end-to-end, across the world's highest-performing private global network as a service.
The need for an SD-CORE has never been greater, and in Paris we learned we are not alone. We saw SD-CORE like projects from Cisco (NGENA) and VMWare. All driven by the need for a better middle mile and the increasing reliability and performance of Internet access. But only Mode offers a network operating at the theoretical limit of packet-switched efficiency. Translation: an untouchable combination of performance and economy.
Mode has replaced that pesky Internet Core with the world's highest-performing SD-CORE, and it works with any SD-WAN with the turn of a key. Still, it was Paris, and you know what those arrondissements can do to star-crossed technology companies. In fact, rumor has it that Mode and Versa Networks were seen holding hands at the Tuileries (read the press release here).
We also saw lots of traditional vendors piling into the SD-WAN space, each trying to differentiate via a better end-to-end experience at a reasonable price. Enter Mode, and our SD-WAN + SD-CORE partner program.
Paris is now a memory. But now you're more confident in the cloud than ever. You know that the combination of SD-WAN and SD-CORE gives you MPLS-level QoS and reliability AND cloud flexibility, and it does it at a business internet price point.
Bonne nuit et beaux reves!
Talk about a busy week. You know the way fast-growth startups can be. Some weeks, you're just pushing that boulder. And other weeks, it comes at you nonstop, and you just don't have enough hours in the day.
This week is one of those.
First, we are incredibly pleased to welcome Versa Networks to the Mode fold (read the press release). Big handshake. Versa Networks is a leading provider of SD-WAN solutions, and the first to join our newly minted SD-WAN + SD-CORE Partner Program. What's that, you ask?
Well, it's like peanut butter and chocolate. Or Gracie and George (Meghan and Harry for you millennials). Two great things that are just better together. SD-WAN brings cloud service flexibility to the edge of the enterprise WAN. That means easy setup and management, and edge visibility. Still, lots of folks continue to use rigid, expensive legacy connectivity solutions like MPLS because of its rock-solid reliability (hint: that's not good cloud etiquette). But we understand. There was no good alternative. Until now.
Mode has stretched the software-defined goodness of SD-WAN, and brought it to the core of the network. We use our breakthrough autonomous routing control to make our cloud private network as reliable as MPLS, but as affordable as business internet. We call this flexible network-as-a-service SD-CORE, and it's built to literally turnkey enhance any SD-WAN. All that SD-WAN edge goodness of flexibility, reliability, and visibility now goes end-to-end, over the entire corporate WAN — for the cost of business internet. We think that's pretty sweet. Peanut butter and chocolatey goodness.
Versa is our first SD-WAN partner. Together we are making enterprises big and small confident in their transition to the cloud. No more clinging to legacy connectivity solutions. It's ok, you can let go, little by little if that makes you feel better.
In other news, Mode is at the SD-WAN Summit 2018 in Paris! And ... we are a Diamond Sponsor!!! Funny, because of the shape of those tetrahedral carbon bonds sort of looks like an autonomously routed network, but that's just me. Our CEO Paul Dawes is giving the keynote all about the need for the SD-CORE I just described. Nithin Michael, Ph.D., and Mode co-Founder will talk about how he brought the world's first autonomous network to life.
When we aren't presenting, you can find us celebrating in the Versa booth, where we will show you how Versa + Mode = SD-WAN + SD-CORE = Better Together = Cloud with Confidence. That's math that I know you'll appreciate. See you there!
In our last blog, we asked a simple question: is the internet good enough? In fact, this is typically how our sales conversations start:
Mode Sales Guy: Hi VP IT, why do you use MPLS today when the internet is so great and cheap?
This question elicits two different responses:
VP IT: Oh, I don’t use MPLS at all. We don’t really use UC or VoIP, or have any sensitive applications that run in our data center or cloud. Basically we just pump everything over the internet.
Mode Sales Guy: Great. Have a nice day!
VP IT: Look, I hate the cost of MPLS. It’s also a real pain to work with — slow to setup, hard to change, cloud unfriendly. But I don’t have a choice. Any problem — video conference glitches, voice call dropouts, access or performance issues — it’s all my fault. Saving money over reliability isn’t worth it. So we use MPLS for mission-critical, and the internet for everything else.
Mode Sales Guy: What if I told you that you could save money and gain flexibility, without affecting reliability?
It's pretty common for IT departments to complain heartily about MPLS, but not believe that they can rely on the internet to replace it.
Remember that this question is the first in a series of three:
- Is the internet good enough for mission-critical business applications?
- If not, is there a specific part of the internet that’s the culprit?
- If so, can you isolate/replace the bad parts, and build a cloud connectivity solution that’s ready for mission-critical business?
Let’s assume for a moment that IT is just being conservative, and look for outside, broader answers to the first question beyond just businesses that use MPLS.
If you think about it, the entire CDN market came into being a while ago because the internet wasn’t good enough for delivering video. The persistence of CDN solutions like Amazon and Akamai suggests it still isn’t.
On the gaming side of things, companies like Riot Games spent millions of dollars to build their own backbone because the Internet isn’t good enough for their gamers. Imagine that — a gaming company becoming a network operator. That’s desperate. And they’re not alone. Nvidia built the GeForce NOW edge network because the Internet isn’t good enough for interactive streaming.
Finally, it’s pretty clear that the $40B+ MPLS market is evidence that the Internet isn’t good enough for mission-critical business applications. Here you’d have a ton of IT professionals nodding in unison about the need for global, consistent reliability with an SLA for mission-critical cloud access, unified communications, VoIP, etc.
It’s pretty clear that there are growing number of applications which require more reliability than the Internet can deliver. In this post-HTTP world, that trend is accelerating.
So the next logical questions are: why is this true? And, can we do anything about it?
If you think Mode might have something to do with the solution, let’s just say you’d be getting pretty warm right about now.
See you next week!
How's that for clickbait? OK, in fairness, the internet is pretty darned good. I like to think of it as the biggest infrastructure project in the history of humanity (truth). It's transformed our lives and our world in profound ways. It seems pretty short-sighted to rail against it.
But, it's a living network. It's not static (although most routing techniques are, go figure). So it can get better. And questioning its capabilities is part of this process. Fair?
Beyond the clickbait, the question I'm really asking is this: is the internet good enough for mission-critical applications? You would think it is, because more and more businesses are running hybrid clouds. They're keeping lots of their mission critical data in the cloud. And over 90% of employees rely on the Internet to access these apps and this data. They're putting a lot of faith in the internet. How's that going?
This question opens up a boxful of follow-ons:
- Is the internet good enough for mission-critical business applications?
- If not, is there a specific part of the internet that's the culprit?
- If so, can you isolate/replace the bad parts, and build a cloud connectivity solution that's ready for mission-critical business?
And that, ladies and gentlemen, is the subject of our next series of intriguing and informative blog posts.
We were very excited to read that Cisco was integrating their Viptela SD-WAN into one million of its routers. Cisco is clearly committed to SD-WAN, everywhere. At Mode, we are committed to ENHANCING SD-WAN, everywhere. We are thrilled that we are so aligned!
SD-WAN does amazing things, of course, which is why Cisco wants it ubiquitous. It simplifies enterprise WAN setup and management, and offers edge flexibility and edge transparency. It's all about visibility and flexibility, actually.
Which is why it's so disheartening to hear that so many businesses are dissatisfied with the performance of their cloud applications. Inconsistent SaaS performance. Dropped connections to IaaS. Laggy VoIP and unified communications.
As good as SD-WAN may be, it's only as good as the networks available to it.
And while the ISP last-mile offers the kind of performance that can support mission-critical business applications, the internet's core is inconsistent. A best-efforts internet core means intermittent dropped packets, and lots of latency variance. That translates into poor SaaS, site-to-site, UC, VoIP, and remote access experiences. The alternative to date has been MPLS. Problem solved — but at a cost. A very big cost. Money and time, actually. While SD-WAN helps IT folks keep those costs down by saving only the most-important application traffic for this costly hardware-defined network, enterprise needs something better. Soon.
Enter a new breed of cloud networks, with big promises (we're one of them of course). Some of them offer MPLS-like performance, but at an MPLS cost. Their feature — flexibility. But only with their own SD-WAN. Others still use the internet core network, but offer global POP access and some optimizations. Again, only with their SD-WAN. None offer the performance of MPLS, the flexibility of cloud, and the price of business internet. With ANY SD-WAN.
Except Mode. We are a reliable and transparent core network, and work with any SD-WAN or UCaaS solution. How do we do it? Simple (not really). Mode is built around an NSF-sponsored autonomous routing discovery that triples network resource utilization using just software. And not just any software: the Mode HALO algorithm is the only mathematically optimal routing solution in the world. Really. The result is an unmatched combination of cloud network performance and price. Mode is the world's highest-performing Cloud Private Network for enterprise. Today we enhance any SD-WAN, SaaS/IaaS access, UCaaS, and cloud CPE. Tomorrow we #tripletheinternet
So thanks, Cisco. You keep setting 'em up, and we'll be there, ready to swing.
A few days ago, news broke that Amazon was going to sell its own network devices. This led to a precipitous drop in the value of a number of market leaders, and the quick denial of any such move by Amazon itself — in the form of a direct phone call to Cisco.
Whether or not any or all of this is "fake news" — for me the takeaway is that people can feel that networks as we know them, and the technologies used to run them, are ready for disruptive change. We are all just waiting for the other shoe to drop.
Making networks better, faster, cheaper at moving data begins with rethinking the way packet data is routed. Today, it is done using a fixed heuristic approach, e.g. Shortest Path First, where "shortest" is whatever heuristic you wish to emphasize — capacity, latency, cost, etc. So what's the downside of this approach? Network utilization in the 30-40%. range. OMG. The internet could be three times bigger without spending a penny on new infrastructure if only routing were better. But how?
Mode HALO already does it. It defines packet data routing as a control system. It brings dynamic, autonomous optimization to any network, and triples network utilization, sustaining 90%+ use of resources.
Mode isn't making the internet 3X bigger (today). Instead we use Mode HALO to offer the highest-performing Cloud Private Network in the world, Mode SD-CORE. Our breakthrough in network utilization translates into MPLS-like performance and reliability, as a transparent and flexible cloud service that's available at a business-internet price point.
Mode enhances any SD-WAN, Unified Communications, Cloud Access, Remote Access, and a host of next-generation applications that demand ultimate network performance without the ultimate price tag.
It's hard for a startup to create a new solution category. It's generally a bad idea. Everyone thinks of the world in a contemporaneous paradigm, and they want to stick you into the right preexisting box. There is little startups can do to change that.
So when we talk about what we've built, for the sake of clarity, we use terms like "Cloud Private Network" or the idea of delivering a flexible, SaaS-friendly private network as a service. It's all very descriptive and easy to digest. Still, there is another term we favor.
SD-CORE. It's no accident that it sounds like SD-WAN. SD-CORE is the yin to SD-WAN's yang. Mode SD-CORE is the world's highest-performing SD-CORE, and it's a perfect complement to any SD-WAN. Here's why:
- SD-WAN, in its purest form, is an edge technology. It does amazing things, but nearly all of its control and transparency is at the edge (CPE). SD-CORE extends the domain of an SD-WAN beyond the edge, through the heart of the global network, and out to the other side where SD-WAN takes over again.
- Transparency. SD-WAN gives enterprises full transparency for the edge of their WAN. SD-CORE extends that complete transparency (and control) end-to-end.
- Reliability. SD-WAN provides last-mile reliability through optional redundancy of ISP connections, and intelligent routing of application traffic. SD-CORE provides an SLA-backed QoS alternative to the internet core transporting data among SD-WAN endpoint CPEs. This is especially important given that over 99% of latency variance happens outside of the last-mile. Furthermore, Mode SD-CORE autonomously routes data over the highest-performing path, including the internet should the private core itself experience any kind of failure. Bits must always flow.
- Security. SD-WAN sends data over encrypted tunnels via the best-efforts internet. Mode SD-CORE preserves customer-control of end-to-end encrypted data, shuttling those encrypted bits over a true QoS controlled private cloud service featuring MPLS-like reliability.
- Flexibility. SD-WAN gives enterprises choice over which application data flows over which network: internet or MPLS. SD-CORE extends this choice, adding Mode SD-CORE to the list. This enables the safe, phased transition of traffic from MPLS or internet to Mode SD-CORE. Mode SD-CORE allows the 60-second spin up of a virtually unlimited number of private global cloud networks, enabling micro-segmentation by application.
- Economics. SD-WAN is heralded for its affordability for applications like site-to-site connectivity. Mode SD-CORE brings business-Internet economics to a global QoS network with dynamic bandwidth allocation. You only pay for what you use.
The combination of SD-WAN and SD-CORE produces a transparent, flexible, secure end-to-end global QoS solution for enterprise, at a business-internet price point.
Two weeks ago, we launched Mode. Launches are full of hope and anticipation, but the truth is you're probably going to have a long wait before the world notices you even exist. So you can imagine my surprise when I woke up to find Mode in a top 10 list from Network World. In an article that used the word "hot" to describe us. Twice. That wasn't expected, at least not so soon.
Of course, it's a short article, and this space is pretty complex. So here are some SD-WAN musings on a sunny Friday in the Bay Area:
- Mode isn't an SD-WAN. The title of the article categorizes us as an SD-WAN startup. Here's the rub: we don't build your SD-WAN, but we do make your SD-WAN better. How is that possible? By enhancing it with an affordable, reliable, QoS, self-service connectivity option. The reliability and QoS of MPLS. The flexibility and transparency of a cloud service. The cost of business internet. And, since we don't offer an SD-WAN, you're free to choose any SD-WAN you want based on any features you'd like.
- We don't use the internet as a backbone. We love the internet. It's great. It's just not great all the time, and no magic tricks, magic beans, or magical routing can make it perform like MPLS, reliability-wise. So we passed on creating a global connectivity service based around an internet core in lieu of a cloud private network built on a state-of-the-art Ericsson private underlay. We can make it incredibly efficient because it's routed with our breakthrough Mode HALO routing algorithm. That efficiency translates into true MPLS reliability with QoS and real QoS control in an affordable cloud service. Today, that just wouldn't be close to possible using the internet. But maybe someday (wink).
- If it's an app that requires solid connectivity, Mode can make it better. SaaS, IaaS, UCaaS, remote access, Express Route, Direct Connect, MS Peering. When we say cloud flexible, we aren't kidding. We like giving our customers plenty of good choices.
So thanks for the swipe right, Jeff Vance. You are definitely in the Mode.