How data centers save holidays
During the year, Russians regularly go on holidays - New Year's holidays, May and other shorter weekends. And this is the traditional time for serial marathons, spontaneous purchases and sales on Steam. In the pre-holiday period, retail and logistics companies are experiencing increased pressure: people order gifts in online stores, pay for their delivery, buy tickets for trips, and communicate. Calendar peaks in demand are a good stress test for online cinemas, game portals, video hosting and streaming music services - all of them work on the holidays during the holidays.
We tell how organized the uninterrupted availability of content is organized using the example of the online cinema Okko, which relies in its work on the power of the Linxdatacenter data center.
Earlier, seasonal equipment purchased additional equipment for local placement, moreover, “with a margin”. However, when Vremya Ch came, it often turned out that companies either couldn’t or didn’t manage to cope with the correct configuration of servers and storage. It simply did not work out to solve these problems during the development of emergency situations. Over time, the understanding came: the demand peaks for content and online services are perfectly worked out with the help of third-party resources, which can be purchased using the pay-as-you-go model - payment for the actually consumed volume.
Today, almost all companies that anticipate a surge in demand for their resources during the holidays (the so-called burst), pre-order the expansion of the bandwidth of communication channels. Those companies that host applications and databases on the data center resources increase the computing power in the clouds for holiday peaks, ordering the necessary virtual machines, the amount of storage, etc. in data centers.
How not to miss the calculations
To prepare for peak loads, the coordinated work of the provider and client is important. The main points in this work include the accurate forecast of the load jump in terms and volumes, careful planning and the quality of interaction with colleagues within the data center, as well as with a team of IT specialists on the side of the content provider.
A number of solutions help organize the rapid allocation of resources necessary so that the new series of your favorite series on the screen of your tablet does not freeze.
- Firstly, these are workload balancers: these are software solutions that carefully monitor the load level of servers, storage and networks, allowing you to optimize the performance of each system for the task. Balancers evaluate the level of availability of both hardware and virtual machines, not allowing sagging system performance on the one hand, and preventing the infrastructure from overheating and slowing down, on the other. Thus, a certain level of stock of resources is supported, which can be quickly transferred to solving urgent tasks (a sharp jump in calls to the portal with video content, an increase in orders of a certain product, etc.).
- Secondly, CDN. This technology allows users to receive content from the portal through access to it from the geographical location closest to the user without buffering delays. In addition, the CDN eliminates the detrimental effect on traffic transmission processes caused by channel congestion, communication disconnections, packet loss at the channel junctions, etc.
Consider the example of preparation for the holidays of the online cinema Okko, using our sites in Moscow and St. Petersburg.
According to Alexei Golubev, Okko’s technical director, in addition to calendar holidays (high season), the company has periods when the main movie novelties from majors come out:
“Every year during the holiday season, Okko grows in terms of traffic volume by about two times compared to the previous year. So, if in the past New Year season the maximum peak load was 80 Gb / s, then in 2018/19 we expected 160 - the traditional double. However, they received more than 200 Gbit / s! ”Okko always prepares for the peak load slowly, throughout the year, as part of a project codenamed “New Year”. Previously, Okko used its own infrastructure, the company has its own distribution cluster of content delivery, on its own hardware and with its own software. Over the course of the year, Okko technicians gradually bought new servers and increased the bandwidth of their cluster, suggesting annual growth by half. Additionally, new uplinks and operators were connected - in addition to large players like Rostelecom, Megafon and MTS, they also connected traffic exchange points and the smallest operators. This approach allowed us to deliver the service to the maximum number of customers by the shortest route.
Last year, after analyzing the cost of equipment, labor costs for expansion and comparing it with the cost of using third-party CDNs, Okko realized that it was time to try a hybrid distribution model. Behind double growth during the New Year holidays there is a decline in traffic, and February is the lowest season. And it turns out that their equipment is idle at this time. By the summer, the decline is leveled, and by the fall season a new rise begins. Therefore, in preparation for the new 2019 year, Okko went the other way: finalized their software in order to be able to distribute the load not only on themselves, but also on external CDNs (Content Delivery Network). Two such CDNs were connected, into which excess traffic merged. Okko’s internal IT infrastructure bandwidth was ready to sustain the same double growth, but in case
“The decision not to increase their CDN saved Okko about 20% of the distribution budget in CAPEX. Plus, the company saved several man-months by shifting the work on setting up the equipment to the partner’s shoulders. ” - comments Alexey Golubev.
Distributing cluster (internal CDN) at Okko is implemented at two Linxdatacenter sites in Moscow and St. Petersburg. Full mirroring of both content and its caching (distributing nodes) is provided. Accordingly, the Moscow data center processes Moscow and several regions of Russia, while the St. Petersburg one processes the Northwest and the rest of the country. Balancing occurs not only on a regional basis, but also depending on the workload of nodes in a particular data center, the presence of a movie in the cache and a number of other factors are also taken into account.
The enlarged service architecture looks like this in the diagram:
Physically, service and product development support consists of about ten racks in St. Petersburg and several racks in Moscow. There are a couple of dozen servers for virtualization and almost two hundred “iron” servers for everything else - distribution, support of the service and its own office infrastructure. The interaction of the content provider with the data center during peak loads does not differ in any way from the current work. All communication is limited to the application for support, and in case of emergency - by call.
Today, as never before, we are close to the present, for a 100% uninterrupted scenario of online content consumption, since all the technologies necessary for this are already available. The development of online streaming is very fast. Legal models of content consumption are growing in popularity: Russian users are gradually getting used to the fact that they need to pay for content. Moreover, not only for cinema, but also for music, books, educational materials on the network. And in this regard, the delivery of the most diverse content and with the least network latency is the most important criterion in the operation of online services. And our task, as a service provider, is to close the resource requirements on time and with a margin.