Introduction (pain point analysis)

Small game development team, have you ever experienced such a scene? Due to a successful social sharing or platform recommendation, the number of users of your casual competition or board game instantly skyrocketed several times or even dozens of times!

This is originally a happy event, but may be due to insufficient server resources, resulting in players can not log in, the game lags or even the service is completely down, not only seriously affect the user experience, but also may make the initial users who have worked hard to obtain a rapid loss. What's even trickier is that small game teams usually have limited resources, and it's difficult for them to invest a lot of infrastructure costs in advance for possible peaks like large game companies do!

If you are struggling with how to cope with the unexpected traffic of small games and ensure smooth launch and operation at a controlled cost, then this elastic scaling solution from Tencent Cloud is tailor-made for you.

Solution Architecture Diagram and Overview

The following diagram illustrates the core architecture of this solution, which is designed to provide automated, resilient and cost-effective infrastructure support for small games:

Tencent Cloud's Elastic Scaling Solution for Small Games: Responding to Instantaneous Traffic Waves, Helping Casual Games Launch Easily and Operate Smoothly - LikaCloud

Its workflow is as follows: player requests are accessed nearby through Tencent Cloud's global acceleration network, and after being cleaned by a security protection layer, they are distributed by a load balancer (CLB) to a network consisting of theElastic Stretch Group (AS)managed clusters of cloud servers (CVMs). The elastic scaling service, based on pre-defined policies (e.g., CPU utilization, network traffic, or custom metrics), theAutomatically and seamlesslyIncrease or decrease the number of CVM instances. The data generated by the game server is processed by the high-performance cloud databases MySQL and Redis, while the logging service CLS is responsible for monitoring and analyzing and triggering expansion and contraction decisions, forming an intelligent closed loop

Core Products and Components

The core of this solution is to build an automated, resilient architecture utilizing a range of hosted products so that you don't have to care about the underlying resource scheduling:

component nameplay a roleKey configuration/selection recommendationsWhy choose it
Elastic Stretch (AS)The brain of the program. Automatically adjusts the number of CVM instances based on monitoring metrics to cope with traffic fluctuations.The recommended configuration is based on theCPU utilization(e.g., 80%) andNetwork Inflow BandwidthThe multidimensional expansion and contraction strategy of theWithout the need for manual intervention, it can automatically expand capacity at peak times and shrink capacity at low times.Significant cost optimization.
Cloud servers (CVM)The computational unit that runs the logic of the mini-game server.optioncomputationally optimizedinstance and configure theUncapped bandwidth packages, in case a sudden increase in traffic causes the bandwidth to hit full.Provides stable, high-performance computing power, seamlessly integrates with AS, and supports rapid scaling up and scaling down.
Load Balancing (CLB)Traffic Portal. Distribute player requests evenly to healthy CVM instances in the backend.configureHTTP/HTTPS ListenerOpenhealth checkupFunction.Eliminates single points of failure, ensures high service availability, and automatically directs traffic to the new instance after expansion.
Cloud Database MySQLStore core relational data such as player accounts, game progress, etc.optionHigh Performance Architecture, planning specifications in advance based on the expected number of players. Supportelastic capacity expansion.Provides highly available architecture and data reliability with strong performance (240,000 QPS), eliminating the burden of building your own database for operation and maintenance.
Cloud Database RedisCache hot data (e.g. player sessions, leaderboards) to reduce database pressure.optionmemory-basedspecifications, can be enabledHybrid Storagemodel to optimize costs.ultra-high performanceThe system guarantees the read/write speed of real-time in-game data and enhances the player experience.
Logging service (CLS)Monitoring and Diagnostic Center. Collect business logs and utilize AI capabilities to quickly locate faults.configureCustomized monitoring metrics(e.g., number of online players) and use it as a trigger for elastic scaling.AI log analysisIt can quickly locate the root cause of faults, improve the efficiency of operation and maintenance, and achieve finer elasticity scaling.
Dayu DDoS High Defense PackageSecurity Shield. Defense against DDoS attacks and CC attacks that may be triggered by bursting fires.Protection is automatically turned on upon service access, eliminating the need to adjust network architecture.For the gaming businessT-class protection bandwidth, precisely cleansing the attack traffic and guaranteeing the stable operation of the business.

Summary of program benefits

  • ⚡ Seconds of elasticity, unlimited scaling.The elastic scaling group can complete the expansion of hundreds or even more computing resources in minutes, so that it can comfortably cope with any traffic peaks, and the business is not aware of it.
  • ? Cost-optimized, pay-as-you-go.You only pay for the computing resources you actually use. Automatic scaling down of resources during low peak periods, as opposed to keeping a large number of idle servers all year round.Costs can be reduced by more than 30%.
  • ? ️ Safe and solid, create with peace of mind.No need to worry about the malicious attacks that follow after a firestorm. the DDoS high defense package provides all-around protection for the business, allowing the team to focus on the game development itself.
  • ? Intelligent operation and maintenance, saving effort.From resource scheduling to troubleshooting, a large number of automated hosting services greatly reduce the complexity and workload of operation and maintenance, so that small teams can also have the operation and maintenance capabilities of large factories.

Application Scenarios and Applicable Customers

This program is ideal for the following scenarios:

  • Light casual game.Such as chess, elimination, IO competition, casual competition, etc.Ultra Casual, H5 & Mini-GamesThese games are highly susceptible to transient fluctuations in traffic due to social fission.
  • New game launch and event promotion.When a game is first released or a major operational event is held, it faces abursting at the seamsof deterministic traffic pressures that require the early deployment of elastic resource pools.
  • Cost-sensitive teams.hoping toMinimum initial investmentSmall and medium-sized development teams that quickly validate game play and dynamically adjust resource costs to maximize benefits based on business growth.

Related links