web stats

vUnity Networking Blog

28 May 2015

vUnity is now a Polycom Certified Partner!

By Brian Brandon

polycom

 

Polycom is a world leader in VoIP and video communications. They’re delivering some of the most innovative communications products on the market today, and we’re eager join them.

“That’s nice,” I hear you ask, “but what does it mean for me?”

We’re now fully accredited to supply their cutting-edge VoIP and audiovisual technologies, and to receive full support for any feature requests and enhancements our growing customer base needs. When you buy a VoIP phone from vUnity, you can rest assured you’re dealing with a company with direct access to the manufacturer and all the tools we need to make sure your network gets set up and managed as efficiently as possible.

And all the benefits we’re getting out of this deal, we’re passing directly on to you.

That means…

  • Lower equipment costs for vUnity customers.
  • Prioritized tech support: with a direct line to Polycom’s support infrastructure, we can much more effectively implement, manage, and troubleshoot our customers’ VoIP phones.
  • Our early access to cutting edge software means your software updates will be far smoother.
  • A full refund policy and lifetime warranties on all Polycom equipment purchased though us. Even if you transfer to another provider, we can replace your damaged hardware at no cost.

We also have access to their testing labs, sales resources and promotions, all of which will just go to further drive down the cost of your next VOIP deployment.

This is a crucial step in vUnity’s mission to exceed your expectations and deliver seamless broadband and voice deployments, and we’re sure this partnership will be a long and beneficial one: not just for us, but for you.

23 May 2015

QUIC Multipath

By Brian Brandon

quic

Google is pioneering a protocol it calls QUIC (Quick UDP Internet Connections, pronounced quick), which makes several improvements to modern networking. It’s exciting because it’s built into Chrome and there’s a lot to learn from the years of research by Google’s expert dev team.

QUIC is hugely successful in enhancing network connections, especially to websites. It doesn’t require extensive RTT’s to retrieve website data and implements a whole new world of performance enhancements such as built in compression and built in FEC . Webpages load faster with QUIC, even in environments with packet loss. Some other benefits of QUIC are unique identifiers instead of IP Addresses, meaning the session can resume from a new network interface or IP Address.

And since we’re all about multipathing here at vUnity, the first thing on our minds was, of course… How we can make QUIC multipath! So, is it possible? How hard would it be to add these features to create a new QUIC Multipath? Could this mean a commercially supported alternative to MPTCP?

Probably all future protocols will have multipath support… But, QUIC isn’t a transport-layer protocol like MPTCP and it would be the first time a session-layer protocol did multipath. Since QUIC is an extension of Chrome, it doesn’t have direct access to interfaces depends on the operating systems implementation of UDP. There’s also no definitive answers on how to best implement bonding even when we’re talking about transport layer protocols. MPTCP is the “right” way to do multipath, but even MPTCP has isn’t complete with a packet scheduler to bonding links of drastically different latencies. These challenges, as well as the new challenges in the development of a first-ever session layer bonding application makes the QUIC Multipath a job for an experienced team.

QUIC would need (among other things) the following to multipath:
1. Congestion control algorithms for individual links
2. Coding of a packet scheduler (i.e. how much traffic to put out each interface)
3. A method to configure OS routing tables to route traffic to and from QUIC.
4. Direct access to the operating systems network interfaces (IP/MAC/GW/Etc.)

Multipath QUIC would mean a revolution to the networked world. It would be awesome to work alongside the community to build a QUIC Multipath. We were told Google will be taking a more serious look into developing QUIC Multipath later this year. If anyone has any other information or insight, please don’t hesitate to contact us!

-Brian

23 Sep 2014

What Time Warner’s Outage Could Mean for the Future

By Brian Brandon

As you may know, Time Warner suffered a nationwide internet outage on August 27th.

It lasted just three hours, but with a system as big and centralized as Time Warner’s, what would have been an inconvenience for a smaller provider became a debacle big enough to warrant a story in Time.

The cause of the outage was surprisingly small.  In the words of the company’s own press release: “During an overnight network maintenance activity in which we were managing IP addresses, an erroneous configuration was propagated throughout our national backbone, resulting in a network outage.”  In layman’s terms, that means someone messed up a setting on the central line that local and regional networks connect to in order to carry data.TWC-Outage-Map 2

But that little configuration left eleven and a half million subscribers without internet, shutting down coverage for large swaths of the country, including New York and Los Angeles, where they hold a virtual monopoly.

Our last article focused on the costs a single business is hit with when their internet goes out, so just try to imagine the amount of money that’s lost when an ISP fails on that scale.

This is one of the inherent problems with monopolies and centralization.  If one company’s providing everyone’s internet, and that company goes out, their customers are left stranded.

This is especially noteworthy in light of Time Warner’s ongoing attempts to merge with Comcast.  If this deal goes through, the resulting creation will be the world’s largest cable provider, controlling almost half of America’s wired broadband services.   However, the merger, in the words of Georgetown University professor Betsy Sigman, “could increase the likelihood of a massive super outage.”  And if that super-outage happens, customers will have fewer places to turn to get their internet back.  Many won’t have any, forcing them to watch helplessly as their online businesses bleed out.

And are Time Warner themselves prepared for a “super-outage?”  Just five days after this one, Time Warner agreed to pay a $1.1 million penalty to the Federal Communications Commission for neglecting to file “a substantial number” of other outage reports.  These weren’t little ones, either.  Federal law states that a company only has to report an outage to the FCC if it lasts more than thirty minutes, and could (among a few other circumstances) either cause massive interruptions to customers’ phones and internet, or interfere with emergency services.  Apparently, even with these standards, there were still too many for Time Warner to report them all.  The writing was on the wall, but it seems they often chose to ignore it.

That brings me to another problem with mega-corporations.  As proved by recent tragedies like the banking crisis and the gulf oil spill, they’re also very difficult to hold accountable.  With no competition and no one except the government, which is very susceptible to lobbyist influence, to oversee them, they often suffer no severe consequences for causing problems that affect millions of people.

However, the fact that a mere three hours of outage can cause such crippling effects tells us not just about our internet service providers, but about the internet itself, and how integral it is to the functioning of successful businesses today.  New York governor Anthony Cuomo called the situation “a stark reminder that out economy is increasingly dependent on a reliable broadband network.”

And here at vUnity, we agree.

21 Sep 2014

What Do Internet Outages Cost Your Company?

By Brian Brandon

outage_2

Here at vUnity, we talk a lot about how few outages we have. But why does that matter? How do internet outages affect your business, anyway?

Well, a quarter of business owners report that their business shuts down completely when their internet goes out. And even if you stay up, things can go south in many other ways. And these can run up all kinds of costs, both countable and not.

Let’s look at some of them.

Calculating the Costs

Most outages occur in the middle of the night for scheduled maintenance. These are annoying, especially if you weren’t expecting them, but they’re not the worst ones. Chances are your customers are in the same country as you, so unless your business is massive, it’s not likely that many of them will be buying at two in the morning.

So, for the purposes of this article, we’ll focus on the ones that make up the exception, not the rule. The outages you truly have to worry about are the ones that occur during the day, or the ones that go on for longer than you could’ve reasonably expected.

How much do these cost you? A lot of it depends on the size and configuration of your network, how quickly your provider handles them, and whether or not your site stays up and functioning during the outage. But you can still arrive at a decent approximation.

Netblazer has an Internet Outage Cost Calculator that’ll do it for you if you plug in some variables. But For those of you who prefer to calculate it manually, or just want to know how it’s done, the easiest way, according to DVI Consulting, is to assign a dollar value to each business area impacted by the outage. And UC Berkley’s Dave Patterson published this detailed chart (warning: comic sans) that lays out how to do this.

First, you’ll need to know the duration of the outage and when it occurred. Next, take your site’s income per quarter and divide it by the number of hours in a quarter. (It’s 2,191, for reference.)

First, if the outage knocked your site out, multiply that last number by the number of hours it was down. That’s how much you’ve lost in pure sales.

Next, if the outage occurred during work hours and employees were forced to stop, figure out how many were unable to do their jobs and whether they remained on the payroll during the outage. If so, add in all their hourly salaries. Did it cost you anything to collect all that information? And what about the repairs? Add those in too.

Steep, isn’t it?

When Amazon went down last year, it cost them over $66k per minute. Chances are, your business won’t hemorrhage anywhere near that much cash, but it’s all proportional. Even though you might lose a smaller dollar amount, it could cost you more relative to what you make. Outages lasting days can even bankrupt a small business.

And even worse, this is a conservative estimate, as there’s often no way to account for certain costs, especially ones that may or may not have occurred. Did you lose any offline sales, as well? For example, if you have a brick-and-mortar store and the internet outage made you unable to process credit cards, did you lose those sales?

If you have a package deal and your phones went out as well, could you have missed some customer service inquiries, or a call from a vital client?

And there are some losses that can’t be quantified at all. For example, will too-frequent website outages cause customers to assume your products will be just as unreliable? (This is especially a problem if you sell anything electronic.) Could they cause a loss of employee trust in your IT department? Could you have missed that one-in-a-million customer while your website was down?

Loss Prevention

Don’t panic, though. There are ways to minimize the damage.

First, resolve to take disaster preparedness seriously. Look into establishing a comprehensive disaster recovery plan. 56% of North American businesses don’t have one, so you’ll be putting yourself ahead of more than half the competition. If you’re big enough to have an IT department, they can handle it. If not, here’s a good place to start.

Next, offload online services like email and web hosting to a certified datacenter that will remain online during a local outage. All data centers with SSAE16-certifiication feature advanced fire suppression systems, uninterruptible power supplies, and ready-to-go backup generators in case of emergency, all in strictly temperature-controlled environment that guarantee your site will stay up through the worst of times.

And finally, consider investing in an ISP that guarantees uptime and responds to outages quickly… For example, one with a 99.99% uptime guarantee, built-in failsafe technology—including entire backup circuits—and response times of less than an hour, like vUnity.

 

07 Jun 2014

The Best FCC Smackdown.. ever.

By admin

We are a net-neutrality ISP, with your business Net Neutrality will not go quietly into the good night.

Head to http://www.fcc.gov/comments to voice a productive comment on the future of communications.

See http://consumerist.com/2014/05/15/how-to-tell-the-fcc-exactly-what-you-think-about-the-proposed-net-neutrality-rule/ for tips on creating a productive comment.