In a previous blog post, What’s in a Network, we took a look at the various components of a network and how they relate to each other. We’re going to take things a step further in this post and take a deeper dive into the world of Internet Service Providers (ISP’s). Before we jump in, let’s recall that an ISP is simply a company that sells you access to the web and that the Internet is simply a decentralized network of networks that sends packets of information around.
IP Addresses and Protocols
When you send an email, two Internet Protocol (IP) addresses are needed to link the device you sent the email from to the device that is receiving it. Each IP address is a string of numbers that is uniquely tied to a computer. They’re basically just street addresses for digital postal workers, allowing information to be delivered to the correct location. An IP address is a 32 bit number that identifies senders and receivers and relays information such as the country, region, network, sub network, and relay information about the device that is being used. Protocols are the set of rules and standards that direct the flow of information.
So when you surf the internet and visit a website, behind the scenes your computer pings the server that hosts the website you’re visiting and then returns its information to your computer. This information retrieval process is called a post or a get request, and it forms the basis of the protocol process.
Peering vs. Transit:
ISP’s can have a private network or use an internet exchange to access the Internet. Both have their pros and cons, but the biggest factor is that an internet exchange is a public system that costs less while a private network is a more expensive private system. Since the internet is a network of networks, communication between them is needed, especially between service providers. Peering occurs when two or more ISPs agree to share their networks so they can both route data between them. On the other hand, internet transit is a contractual agreement between two ISPs that allows one ISP’s network traffic to cross or "transit" another ISP’s network for a fee. Transit arrangements are usually used to connect a smaller Internet service provider (ISP) to the larger Internet.
Generally, the lower tier an ISP is the less it has to rely on Peering or Transit to reach other networks.
Tier 1: A network that can reach every other network on the Internet without purchasing IP Transit or paying settlements.
Tier 2: A network that peers with some networks, but still purchases IP transit or pays settlements to reach at least some portion of the Internet.
Tier 3: A network that solely purchases transit from other networks to participate in the Internet.
Types of ISPs
While a majority of ISPs are classified as access ISPs, which simply provide internet access to its users, several other types of ISPs exists that offer a range of services; these include:
Mailbox ISP: A mailbox ISP hosts email domains and provides servers for its users to send, receive, and store emails.
Hosting ISP: A hosting ISP provides email, web-hosting, and online storage services to its users.
Transit ISP: A Transit ISP sells access to the internet to smaller ISPs that do not have as large of a network.
Virtual ISP: A virtual ISP is a provider that may not own its own infrastructure. Instead it simply acts as a middleman and purchases services from another ISP to allow its users access to the internet.
Free ISP: A free ISP provides its services for free. Free ISPs usually either run advertising or are set up as a municipality or nonprofit organization.
Wireless ISP: Provide internet access via a wireless network.
Generally, ISP’s offer either a Standard connection or High Availability connection. High Availability connections are crucial for organizations that rely on stability and continuous network uptime. Often when choosing a high availability connection, organizations sign a Service Level Agreement (SLA), from the service provider to determine a scope of work as well as outline performance metrics.
One of the metrics that is essential for an organization is availability, which is measured by the percent of uptime per year. The percentages are expressed in the amount of nines (see graph below). It is essential for an organization to look at the opportunity cost when deciding on different levels of availability and how it will affect the business.