Investigators are piecing together details about one of the most aggressive computer attacks in recent memory — a powerful “denial-of-service” assault that overwhelmed computers at U.S. and South Korean government agencies, companies and institutions, in some cases for days.

Here are some questions and answers about the attack.

Question: What is a “denial-of-service” attack?

Answer: Think about what would happen if you and all your friends called the same restaurant over and over and ordered things you didn’t even really want. You’d jam the phone lines and overwhelm the kitchen to the point that it couldn’t take any more new orders.

That’s what happens to Web sites when criminals hit them with denial-of-service attacks. They’re knocked offline by too many junk requests from computers controlled by the attackers.

The bad guys’ main weapons in such an attack are “botnets,” or networks of “zombie” personal computers they’ve infected with a virus. The virus lets the criminals remotely control innocent people’s machines, which are programmed to contact certain Web sites over and over until that overwhelms the servers that host the sites.

Question: How often do these attacks happen?

Answer: People try denial-of-service attacks all the time — many government and private sites report being hit every day. Often the assaults are unsuccessful, because Web sites have ways of identifying and intercepting malicious traffic. However, sites really want to avoid blocking legitimate Web users, so more often than not, Internet traffic is let through until a problem is spotted.

Question: Some organizations appear to have fended off these recent attacks, while other Web sites went down. How can this be?

Answer: The sites that went down probably were less prepared, because they are less accustomed to being hit or aren’t sensitive enough to warrant extra precautions.

Popular Web sites, like e-commerce and banking sites, have a lot of experience dealing with denial-of-service attacks, and they have sophisticated software designed to identify malicious traffic. Often that’s done by flagging suspicious traffic flowing into the site, and if there’s enough of it, preventing it from ever reaching the site’s servers.

Another approach is to flag suspicious individual machines that seem to be behind an attack, and ban any traffic from them from reaching the site.

That can often be difficult, though, because criminals use “proxy” computers to route their traffic, masking the source of the original requests.

Question: Is there usually evidence of who the culprits were? Or is the nature of the attack such that it leaves few fingerprints?

Answer: It’s usually easier to stop a denial-of-service attack than it is to figure out who’s behind it. Simply identifying where the malicious traffic is coming from won’t get investigators very far, since the infected PCs that get roped into a botnet are owned by innocent people who don’t know their computers are being used for nefarious purposes.