네트워킹 설계

Understanding Traffic Requirements

Performance efficiency design for cloud-based information systems starts with identifying end user locations and selecting the Region where the information system will be deployed.

This is because the geographical distance between the end user and the data center significantly impacts user request latency.

You must consider the following factors when determining the data center Region.

Legal Regulations

You must consider legal regulations regarding the geographic location where end users’ personal information and sensitive information are stored.

For example, public institutions may need to use a CSAP-certified public cloud.

Latency

Minimize the time it takes for data to travel from the source to the end user.

It is advisable to identify the services that users frequently use and to deploy the system in a data center that is geographically close to those users.

When providing a global service, you should review whether to distribute the system to each country or minimize latency through a Global CDN.

Evaluate network requirements to determine appropriate workload services and configuration.

• Identify the data transmission requirements and network request frequency required by the information system.

• Understand bandwidth requirements. Analyze the workload to calculate network traffic based on the number of service requests per unit time and request data size, and secure bandwidth capable of accommodating the traffic. Generally, traffic and bandwidth are expressed in bps (bits per second).

• Establish strategies to minimize network latency. When delivering web content over the public internet, you can reduce latency by using a Global CDN. If a private network connection is required, you can configure a private connection using VPN, Direct Connect, Transit Gateway, etc., and you can also select the bandwidth of VPN or Direct Connect to meet the Application’s required transmission speed.

• Understand throughput. Review whether to configure a single VM or multiple VMs, considering the throughput required for the workload. You can distribute load to multiple VMs using a Load Balancer, and by applying GSLB, you can route request load to multiple Regions (IP addresses).

Establishing Latency Reduction Strategies

The user undergoes a series of processing steps across various components to request a service from the web application and receive a response.

We previously examined performance improvement measures regarding computing and data storage.

Based on the diagram below, the user and the web application are connected via the internet, and the latency in this segment varies depending on the geographic distance, or more accurately, the number of network hops.

Generally, the time required for data transmission increases as the geographical distance increases.

We use a Global CDN service to reduce this latency.

Global CDN leverages numerous edge servers distributed across a global network to deliver static content stored on web servers or object storage to users faster and more securely.

Additionally, even during traffic spikes, it distributes the load on the origin server to prevent overload, and end users can download content from the nearest edge server to enjoy fast and reliable web services.