Relocation is always stressful, no matter what. Moving from a more uncomfortable two-bedroom apartment to a more comfortable one, relocating from one city to another, or even gathering yourself to move out of your parents' house at 40. It’s not so simple with infrastructure transfer either. It’s one thing to have a small website with a couple of thousand unique visitors a day, ready to spend a few hours and a couple of cups of coffee on data transfer. It’s quite another when you have a complex infrastructure with a ton of dependencies and workarounds in place for specific spots in a particular cloud.
And if we add 1C to this, the process starts to take on new colors.
My name is Sergey Kondratyev, I am in charge of our striped cloud, BeeCLOUD, and in this post I will tell you about the transition to our cloud by the company 'AeroGeo'.
Why move at all
First of all, let me tell you about the specifics of the 'AeroGeo' business. It’s a Krasnoyarsk airline that has been transporting passengers and cargo for 13 years, with over 40 aircraft in its fleet, including helicopters. They operate exclusively in Russia, covering the entire territory, meaning you can spot the company’s planes from Altai to Kamchatka. A distinctive hallmark is that 'AeroGeo' ensures the full operation of the Seasonal Drifting Station of the Russian Geographical Society.
Bell 429, photo from Company
In general, there are enough clients, more than 350 internal employees, and aviation work of any complexity. Therefore, a properly functioning infrastructure is extremely critical for the company. And you know how capricious 1Cs can be without me having to explain.
So, a year ago, the client had a clear need to update their infrastructure. Naturally, they began looking towards operational cloud solutions, and it turned out that, first of all, the company’s management had some doubts about cloud solutions (whether everything would actually be available 24/7 or not), and secondly, they definitely did not want to work through a public channel. To their credit, upon deciding to move, they put us through a serious examination: the IT director personally flew in to look around and understand how things work with us. He walked around, observed, made conclusions, and approved the pilot project.
The infrastructure that needed to be migrated was designed for the work of 30 specialists peaking from three different offices (read — from three different networks, the head office, the "Emelyanovo" airport, and the "AeroGeo" airport). We thought it over and decided to consolidate everything into a single network, which we then reserved via the IPSec protocol, connecting a dedicated "Krasnoyarsk-Moscow" tunnel at 100 Mbps. The hardware key is located in our Data Center on a USB hub and connected to the client pool.
The migration took just one evening because the representative from "AeroGeo" simply brought us the main database on a physical medium directly to the Data Center where the platform is deployed. In fact, we were concerned about the key bindings; there were worries that the keys would drop during the migration, but it went smoothly, as the keys were linked to similar hosts.
The pilot project lasted about a month, during which we actively gathered feedback from 1C specialists. Over this month, they did not notice any performance degradation or inconveniences.
Why choose us
There are many clouds now; almost every major player in the market already has its own cloud with a bunch of perks. It’s understandable, if you want to compete, create an excellent cloud and add a little extra.
We currently have three Data Centers (Moscow), the cloud runs on OpenStack (if you're interested, I can write about this in detail in a separate post), we've gained experience in migrating very different 1C solutions to the cloud. BeeCLOUD has hosts at both 3 GHz and 3.5 GHz (we specifically chose one with a dedicated HP Synergy cluster at 3.5 GHz for "AeroGeo"), depending on the client's needs.
Since 1C is such that in its setup and tweaking, the principle of "Each to his own" actively continues to apply, we created an excellent cluster where clients can transfer their highly customized, demanding 1C solutions without spilling anything along the way. Everything will work. TIER 3, SLA 99.97, FZ-152, standard scenario.
But these are just numbers and technologies. In our product, the main focus is on people. We have successfully gathered an excellent team of great engineers, some based in Moscow and others working remotely from various regions. This gives us a crucial opportunity to assist our clients on-site. It is one thing when you (even as a VIP client) call support and spend some time on the line explaining what has broken this time, after which the support team goes to check things remotely. It’s quite another when network engineers and experts can solve all potential problems on-site, with their own hands.
Of course, the cloud is also great because it relieves the client of all headaches and ceremoniously transfers them to the provider. At "AeroGeo", everything was tied to this 1C system. Now they know that we maintain the relevance and functionality of the system. Whenever something new comes from the vendor, and a patch needs to be implemented, we simply inform the client, coordinate a convenient time for them in their time zone to perform the work, and get to it. For example, when fresh patches from Intel and HP were rolled out to the hosts, our team did this during the lowest load times according to Krasnoyarsk time.
Moreover, we managed to do everything within a single window. Different services sometimes experience hiccups when you, as a provider, offer a service, but you have a bunch of contractors. If something goes wrong with the contractors, you waste time communicating with them as well. The client doesn't care; since they are paying you, all the problems should be solved by you.
Therefore, in the case of BeeCLOUD, we decided to eliminate this approach and do everything ourselves. Our own backbone channel, our own support, our own hardware. This means it’s faster for the client in case something goes wrong; if there’s an issue, it’s definitely our problem, and we will resolve it. Plus, it really saves a lot of time on internal processes when everything is in-house — you have just one service desk, without a bunch of clones and synchronizations or constant ping-pong with contractors.
And about the money
Where would we be without it? I can't reveal too many figures within the confines of this post, but they will still convey the scale. When "AeroGeo" calculated the cost of modernizing the existing infrastructure, they estimated it would exceed 2,000,000 rubles. And these are preliminary figures, typically accompanied by a note stating 'From'. This is for the update alone, without any maintenance or support.
For the migrated infrastructure in BeeCLOUD, which includes the hardware and 24/7 support, the client pays 45,000 rubles per month. This means that two million rubles will be enough for nearly 4 years of hassle-free operation.
We strive to be as transparent as possible; if a client wants to visit us and see what everything runs on, they're welcome. By the way, about the cloud itself: .
If you have any questions about this case or about our cloud in general — feel free to write, I'd be happy to respond.
Source: habr.com
