156-560 · Question #110
Automatically adding or removing cloud instances based on load is called:
The correct answer is A. Horizontal Scaling. See the full explanation below for the reasoning.
Question
Automatically adding or removing cloud instances based on load is called:
Options
- AHorizontal Scaling
- BSuper Scaling
- CHyper Scaling
- DVertical Scaling
How the community answered
(50 responses)- A76% (38)
- B14% (7)
- C4% (2)
- D6% (3)
Community Discussion
14A is correct, Horizontal Scaling is the term for automatically adding or removing cloud instances in response to load changes. The key word in that question stem is "instances," plural, meaning you are spinning up or tearing down whole compute nodes rather than resizing a single one. Vertical Scaling, option D, is the trap here because it sounds like it could mean "scaling up," but vertical means you are changing the CPU, RAM, or disk on an existing instance, not adding new ones. B and C are distractors that Check Point and most cloud vendors do not use as formal terms, so if you see either of those on a live question, eliminate them first and move on.
Horizontal Scaling is your answer, and the logic is straightforward once you lock in the definition. The stem is describing elasticity, specifically the automatic addition or removal of instances, meaning whole compute units, in response to load changes. That process of adding or removing instances is horizontal movement, more nodes in the pool or fewer, not making a single node bigger or smaller. Vertical Scaling, choice D, is the trap here because it sounds like the "advanced" option, but vertical means resizing a single instance up or down in CPU or RAM, and it does not describe the automation of spinning instances in or out. Super Scaling and Hyper Scaling, B and C, are not real architectural terms in this context and exist purely to fill the option set, so eliminate them immediately and do not waste clock cycles on them.
Worth adding for exam precision: AWS documentation, and Azure's too, consistently labels horizontal movement as "scale out" when adding instances and "scale in" when removing them, so if the stem or an answer choice uses that phrasing instead of "horizontal scaling," it maps to the same concept and you should not be thrown off by the wording swap.
Think about which direction the scaling goes, then ask yourself: does adding instances scale out or up?
That framing clicks for most people, but worth noting that "scale up" trips some folks because on-prem tradition used it for vertical too, so naming the vertical case explicitly once can save a round of confusion.
Honestly I almost went with D thinking vertical was the dynamic one since it felt "elastic" to me, but vertical scaling is just resizing the existing instance, no automation implied. Horizontal scaling is specifically the pattern of spinning up or tearing down instances in response to demand, which is exactly what cloud auto-scaling groups do.
This one actually showed up almost word for word on mine and I almost second-guessed myself because "Super Scaling" sounds like something a vendor would make up to sound impressive, but I kept coming back to the definition, adding or removing whole instances based on load is the textbook description of horizontal scaling, so I went with A and moved on.
Good call trusting the definition over the branding, and that same logic kills the distractor about "Super Scaling" on any other vendor question too, because the moment you see whole instances spinning up or down, that is horizontal no matter what they call it.
The word "automatically" is doing a lot of work here and it's what trips people up, because both horizontal and vertical scaling can technically be automated, but the key is "adding or removing instances," which means spinning up or terminating whole machines, and that is horizontal scaling by definition. D is the trap for anyone who blanks on the terminology and thinks "scaling" just means "making things bigger."
Correct, but do you understand why this differs from vertical scaling?
Vertical scaling throws more horsepower at one machine until you hit a ceiling, while horizontal scaling trades coordination overhead for theoretically unlimited growth, so the "better" choice really comes down to whether your workload can be distributed cleanly.
Honestly I almost picked D thinking "scaling" in cloud always meant bumping up resources on the same machine, but vertical scaling is just upsizing a single instance, not adding or removing instances dynamically. The word "automatically" plus "adding or removing" instances locked it in as A, horizontal scaling, since that is exactly what auto-scaling groups do.
Horizontal scaling is right, but do you know what triggers the automation, load thresholds or schedules?
Load thresholds and schedules are both valid triggers depending on the platform, but most cloud auto-scaling you will actually see on the job reacts to metrics like CPU or request queue depth, not a clock.