Lesson One: Big B and Little b

One of the initial challenges that you will encounter when attempting to research bandwidth rates is the lack or ambiguity of standards when these metrics are discussed. Let's investigate a couple problem areas here for reference:

Bits or Bytes: This is where we talk about big and little B. A bit is simply a one or a zero and is usually fairly meaningless by itself. A byte is eight bits and is capable of representing 256 unique binary addresses. A bit is written as lowercase b and a byte as uppercase B.

You've probably noticed that data speeds for networking gear is rated in megabits or gigabits, whereas some storage data (and capacity) rates are discussed in megabytes and gigabytes.

In this case, we have a simple conversion:

1 gigabit per second of bandwidth equals 1,000,000,000 bits which is equal to 125 decimal megabytes per second. We achieve this result by simply dividing by 8.

Now why did I use the term "decimal" megabytes? Because networking bandwidth is generally described using power of ten factors (decimal) whereas server bandwidth and usually data storage capacity is measured in powers of 2 (binary). Confusing? Of course it is, but it allows for a higher degree of creativity in the marketing of products. Actually, this is one of the areas where a standard would be nice. I've seen many bright engineers over the years get confused on this topic, sometimes comically and sometimes tragically.

For simplicity and all server bandwidth results, I will use big B (bytes) and power of 2 conversions. And if you ever require clarification, please drop me a note.