{"id":29840,"date":"2019-10-31T21:32:16","date_gmt":"2019-10-31T18:32:16","guid":{"rendered":"https:\/\/prohoster.info\/blog\/istoriya-uspeha-nginx-ili-vozmozhno-vsyo-probuj\/"},"modified":"2019-10-31T21:32:16","modified_gmt":"2019-10-31T18:32:16","slug":"istoriya-uspeha-nginx-ili-vozmozhno-vsyo-probuj","status":"publish","type":"post","link":"https:\/\/prohoster.info\/en\/blog\/istoriya-uspeha-nginx-ili-vozmozhno-vsyo-probuj","title":{"rendered":"The Success Story of Nginx, or 'Anything is Possible, Just Try!'","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\"The
\n
\nIgor Sysoev<\/b>, web server developer nginx<\/a><\/noindex><\/strong>, a member of a large family HighLoad++<\/a><\/noindex>, was not just a part of the inception of our conference. I see Igor as my professional mentor, a master who taught me how to work with and understand high-load systems, which has defined my professional path for a decade.<\/p>\n

Naturally, I couldn't overlook the resounding success<\/a><\/noindex> of the NGINX team... And I interviewed not Igor (he remains the introverted programmer), but the investors from the Runa Capital<\/strong>, who recognized nginx ten years ago, built a business infrastructure around it, and are now conducting a deal unprecedented in size for the Russian market.<\/p>\n

The aim of the article below is to confirm yet again \u2014 anything is possible! Give it a try!<\/a><\/noindex><\/p>\n

Oleg Bunin, Head of the HighLoad++ Program Committee: Congratulations on the successful deal! As far as I can tell, you managed to preserve and support Igor's desire to continue working as a programmer while building the entire business infrastructure around him \u2014 that's a dream for any developer, isn't it?<\/strong><\/p>\n

My interlocutor, managing partner of Runa Capital, Dmitry Chikhachev:<\/strong> That's right. This is largely thanks to Igor himself and his co-founders Maxim and Andrey (Maxim Konovalov and Andrey Alekseev), because they were originally ready for this infrastructure to be built around them. Not all startup founders adequately assess their own strengths and capabilities.<\/strong> Many want to lead or manage the entire process.<\/p>\n

\u2014 So, did the NGINX team essentially distance themselves from the business part, or what?<\/strong><\/p>\n

Dmitry:<\/strong> No, they did not distance themselves from the business part, why would they? Maxim managed the operational side as the Chief Operating Officer. Andrei took care of BizDev, while Igor continued with development \u2014 the aspect he enjoys.<\/p>\n

Everyone focused on their strengths and what they liked.<\/p><\/blockquote>\n

\nBut they all understood that building a multi-million dollar business in the U.S. required a person of a different caliber, with a different background. Therefore, there was an agreement with the investors from the very first round of negotiations that such a person would be found. This person was Gus Robertson, who met all these criteria.<\/p>\n

\u2014 So, was it initially planned to enter the American market?<\/strong><\/p>\n

Dmitry:<\/strong> NGINX is a B2B business. It's not particularly well-known to users, as it operates at the infrastructure level, one could say middleware. The primary market for B2B is the USA, which accounts for 40% of the global market.<\/p>\n

Success in the American market predetermines the success of any startup.<\/p><\/blockquote>\n

\nTherefore, a logical plan would be to head to the USA, immediately hire someone to lead the American company, and develop the business while attracting American investors. If you want to sell infrastructure software in the USA, it's important to have American investors backing you.<\/p>\n

\u2014 Who approached whom: you to nginx, or nginx to you?<\/strong><\/p>\n

Dmitry:<\/strong> We had many different points of contact. We probably took more initiative, because even then nginx was noticeable. Although it wasn\u2019t a company yet and its market share was relatively small (6%), there was already significant investor interest. The deal was competitive, so we were certainly proactive.<\/p>\n

\u2014 What was the state of the product? There was no company, but were there drafts of a commercial enterprise version?<\/strong><\/p>\n

Dmitry:<\/strong> There was an open-source web server nginx. It had users representing 6% of the global market. In reality, that's millions, even tens of millions of websites. However, there was still no company and no business model. And since there was no company, there was no team: there was Igor Sysoev \u2014 the developer of nginx and a small community around it.<\/p>\n

It's a very interesting story. Igor started writing nginx quite a while ago \u2014 in 2002, and released it in 2004. Real interest in it only emerged in 2008, and in 2011 he attracted funding. Few people ponder why there was such a long delay. There is actually a logical technical explanation for this.<\/p>\n

In 2002, Igor was working at Rambler, and there was one problem he was solving as a system administrator \u2014 the so-called C10k problem, which is ensuring a server can handle more than ten thousand simultaneous requests at peak load. This issue was just emerging, as heavy loads on the internet were starting to become common. Only a few individual sites faced it \u2014 such as Rambler, Yandex, Mail.ru. For most websites, it wasn't relevant. When there are only 100-200 requests a day, there\u2019s no need for nginx; Apache would do just fine.<\/p>\n

As the internet became increasingly popular, the number of websites facing the C10k problem grew. More and more websites required a faster web server to handle requests\u2014something like nginx.<\/p>\n

However, the real surge in load occurred between 2008 and 2010 with the advent of smartphones.<\/p><\/blockquote>\n

\nIt's easy to imagine how quickly the number of requests to servers skyrocketed. First, internet usage time increased because clicking on links became possible everywhere, not just while sitting at a computer. Secondly, user behavior changed as well\u2014navigating via touch screens led to more chaotic link transitions. One can also include social media.<\/p>\n

This led to exponential growth in peak loads on the internet.<\/strong>Overall load increased fairly evenly, but the peaks became increasingly pronounced. It turned out that the C10k problem had become widespread. At that moment, nginx took off.<\/p>\n

\"The
\n
\n\u2014 Tell us how events unfolded after meeting Igor and his team? When did the development of infrastructure and business ideas begin? <\/strong><\/p>\n

Dmitry:<\/strong> At first, the deal was being formed. I already mentioned that it was a competitive deal, and eventually, a syndicate of investors emerged. We became part of this syndicate along with BV Capital (now e.ventures) and Michael Dell. We first closed the deal and then started considering the search for an American CEO.<\/p>\n

And how did you close the deal? It turns out you didn\u2019t even know what the business model would be or when it would become profitable? You just invested in a team and a great product?<\/strong><\/p>\n

Dmitry:<\/strong> Yes, it was purely a seed deal. We weren\u2019t thinking about the business model at that moment.<\/p>\n

Our investment thesis was based on the fact that NGINX is a unique product with a significantly growing audience.<\/p><\/blockquote>\n

\nIt solved a serious problem for this audience. I have a favorite test, a litmus test for any investment\u2014does the product solve a widespread and painful problem? NGINX passed this crash test with flying colors: the problem was widespread, loads were increasing, and sites were going down. And it was painful because it marked the era when a website became, as they say, mission critical.<\/p>\n

In the 90s, people thought this way: the website is down\u2014I'll call the sysadmin, and in an hour it will be up again\u2014everything's fine. By the late 2000s, a 5-minute downtime had become a real loss of money, reputation, etc., for many companies. The fact that the problem was painful is one side of the story.<\/p>\n

The second aspect we, as investors, look at is the quality of the team<\/strong>. Here we were impressed by Igor and his co-founders. It was a complementary experience and a unique product that was developed by one person.<\/p>\n

\u2014 Clearly, the team with a certain level of complementary competencies also played a role.<\/strong><\/p>\n

Dmitry:<\/strong> I think it\u2019s right that Igor developed the product alone, but when it came time to create the business, he didn\u2019t jump in alone but with partners. Looking at my 10 years of investment experience, I can say that having two co-founders definitely reduces risks. The optimal number of co-founders is two or three.<\/b> One is too few, while four is already too many.<\/p>\n

\u2014 So what happened next? When the deal was already done, but the business idea hadn\u2019t been developed yet.<\/strong><\/p>\n

Dmitry:<\/strong> The deal is made, the company is registered, documents are signed, money is transferred\u2014everything\u2019s set. At the same time, we worked on the business aspects and hired a team of developers who started working on the product. Andrei Alexeyev, as BizDev, built the initial relationships with potential clients to gather feedback. Everyone thought together about the business model, and everyone searched for a top manager to develop the American business and essentially head the company. <\/p>\n

\u2014 And how did you find him? Where? I can't even imagine how to do that.<\/strong><\/p>\n

Dmitry:<\/strong> All the investors and the board of directors were involved in this. Ultimately, the choice fell on Gus Robertson. Gus worked at Red Hat, whose top manager was our investor. We reached out to Red Hat since it\u2019s open source and said we were looking for someone capable of leading the business and growing it into a billion-dollar company. They recommended Gus.<\/p>\n

The deal with NGINX was closed in 2011, and by 2012 we had already met Gus, and we liked him right away. He had a background in open source from Red Hat\u2014at that time it was the only company with a multibillion-dollar valuation in open source. Plus, Gus was focused on business development and sales\u2014that's exactly what we needed!<\/p>\n

In addition to background and experience, we appreciated his personal qualities\u2014he is a perceptive and insightful person with a quick mind, and, importantly, we felt he had a good cultural fit with the team. Indeed, that turned out to be the case. When they met, it turned out that everyone was on the same wavelength, all interacting excellently.<\/p>\n

We made G\u0430\u0441 an offer, and at the end of 2012, he started working with us. G\u0430\u0441 also offered to invest his own money in NGINX. This impressed all the investors. Due to G\u0430\u0441's high involvement, he integrated into the founding team and was regarded by everyone as one of the co-founders of the company. Subsequently, he became one of the four. There is a famous photograph where all four are in NGINX t-shirts.<\/p>\n

\"The
\nThe photograph is taken from notes<\/a><\/noindex> Dmitry Chikachev about the history of collaboration between NGINX and Runa Capital.<\/i><\/p>\n

\u2014 Did you manage to identify the business model right away, or did it change later?<\/strong><\/p>\n

Dmitry:<\/strong> We managed to identify the model right away, but before that, we debated for a while about how and what to do. However, the main discussion revolved around whether to continue supporting the open-source project, to keep nginx free, or to gradually push everyone to pay.<\/p>\n

We decided that it would be right to leverage the strength of the community supporting nginx, not to disappoint them, and to continue supporting the open-source project.<\/p><\/blockquote>\n

\nTherefore, we decided to keep nginx open source, but additionally create a special product called NGINX Plus. This is a commercial product based on nginx that we license to enterprise clients. Currently, the main business of NGINX is selling NGINX Plus licenses.<\/p>\n

The main differences between the open and paid versions are:<\/p>\n