If we assume a crytpo-system with a large key (and therefore a large key space) a brute force attack will likely take a good deal of time -anywhere from several hours to several years depending on a number of variables. If you use a session key for each message you encrypt, then the brute force attack provides the attacker with only the key for that one message. So, if you are encrypting 10 messages a day, each with a different session key, but it takes me a month to break each session key then I am fighting a loosing battle.
The other answers are not correct because: “The use of good key generators” is not correct because a brute force key attack will eventually run through all possible combinations of key. Therefore, any key will eventually be broken in this manner given enough time.
“Nothing can defend you against a brute force crypto key attack” is incorrect, and not the best answer listed. While it is technically true that any key will eventually be broken by a brute force attack, the question remains “how long will it take?”. In other words, if you encrypt something today but I can’t read it for 10,000 years, will you still care? If the key is changed every session does it matter if it can be broken after the session has ended? Of the answers listed here, session keys are “often considered a good protection against the brute force cryptography attack” as the question asks.
“Algorithms that are immune to brute force key attacks” is incorrect because there currently are no such algorithms.
References: Official ISC2 Guide page: 259 All in One Third Edition page: 623