Prof. Hammoud - Research Questions

1. What happens when the problem you want to solve becomes too big for any one computer?

The problem can be divided into smaller tasks and distributed across multiple computers. These computers communicate over a network and work together to solve the larger problem.

IBM - What is Distributed Computing?

2. Suppose 1,000 computers work together. Do you now have one computer that is 1,000 times more powerful? Why or why not?

Not necessarily. Computers need to communicate and coordinate their work, which takes time. Some problems are also harder to divide than others, so performance does not increase perfectly with every additional computer.

3. Can 1,000 computers agree if some fail or lie?

Yes, as long as fewer than one-third of them are faulty. This is the Byzantine Generals Problem: honest computers must agree even when some crash or send false messages. To handle f bad computers, you need at least 3f + 1 in total, so 1,000 computers can handle up to 333 liars. The tradeoff is that agreeing takes many rounds of messages, which makes it slower.

Source - Byzantine Generals Problem

4. When you use ChatGPT, Google, Instagram, or an online game, where is the computation actually happening?

Much of the computation happens on remote servers in data centers, rather than directly on your phone or computer. Your device sends requests over the Internet, and these servers process the data and send the results back.

Cloudflare - What is the Cloud?

5. If you could make millions of computers behave like one dependable machine, what could humanity build that we cannot build today?

We could build much more detailed climate simulations. According to the source, older climate models see the atmosphere at about 100 km resolution, while exascale supercomputers can simulate it down to about 1 km. This lets researchers predict regional weather extremes far more accurately. Getting there takes an enormous number of computers working together as one reliable machine, because a failure partway through would ruin a long simulation.

Source - Exascale Computing

Home Page