This is a good course. Lectures are very good, albeit dense. A lot more math in this course than what I expected going in. The course readiness thing mentions calculus but you never use it (at least, not in lectures and not in our exams). Algebra and very basic linear algebra are good enough

Course is in 3 parts: memory-hierarchy-aware algorithms (e.g. external sorting), SMP multithreaded algorithms, and message passing distributed algorithms. Each section has its own mathematical model that's used for analyzing performance Would recommend taking this course alone. Getting an A is already hard, but even if you manage to get an A there's still more to understand

This class was the halfway point for me, best one I've seen so far in OMS. Grateful for the professor and TAs

DISCLOSURE: I dropped the course about 5-6 weeks in, with no plans of retaking it.

I attempted the course as #5 in OMSCS, but I didn't find it particularly useful, so I decided to drop accordingly.

The purpose of this review is not necessarily to discourage others from taking the course (or to disparage the course and/or staff, either, for that matter; on the contrary, both the students cohort and staff were very engaging and solid folks on the whole, for the record), but more so a "word to the wise" in terms of some due diligence I wish I had done before enrolling, in order to facilitate making a more informed decision if you're considering taking the course yourself. Part of what pushed me into enrolling in the first place was "blind faith," in terms of the generally overwhelmingly praising/positive reviews of the course here and elsewhere, against my own better judgment, in hindsight...

For starters, bear in mind that high performance computing is a specific topic/niche within computer science (basically dealing with cluster computing, which is more so research-focused, such as doing large simulations). I would strongly advise to use Google, YouTube, etc. to search this term/phrase to get a better general sense of the topic. This recommendation is not meant patronizingly/sarcastically, for the record; on the contrary, I was none the wiser going in (as per the aforementioned "blind faith"), and accordingly failed to do this basic step of due diligence myself in the first place (i.e., a "rookie mistake," despite myself being no stranger to academia at this point in terms of cumulative experience/exposure, no less)...

HPC broadly covers parallelism, but in a pretty oddly specific manner, with heavy emphasis on algorithmic analysis of some rather specifically hand-selected/showcased algorithms (and similarly for the projects, too). My own intention for taking the course was to "improve my familiarity/competency with parallelism in general," but more specifically in the context of applications programming (i.e., running your application on a multi-core processor more efficiently, using a language-provided facility such as .NET Task Parallel Library, Java Virtual Threads, Go Goroutines, etc.). For this particular purpose, I personally did not find the material particularly compelling/useful (at least based off of the first 1/3 or so I had gotten through by the point of dropping, which was far enough into it for my own liking to constitute "cutting my losses" by that point accordingly).

Additionally, bear in mind that the course heavily focuses on research papers; in fact there are separate, dedicated TA-led weekly office hours just for going over the papers (i.e., separately from the "main" office hours with the lead instructor). This may be a plus or a minus, depending on your own particularities and such.

In summary, my recommendation to better prospect the course in order to see if it's the right fit for you would be to do the following:

• Look up "high performance computing" to understand the basics of what it entails (i.e., cluster computing)
• Watch the first few lectures* to get a general sense of the topical coverage and presentation style
• Skim some of the papers* to see if there are any particular topics there that resonate with your own interests
• Peruse the non-required-ish textbook*
  • While not "strictly required," there are a few chapters assigned as readings, and the book generally covers the same "subject matter at large" as the course, so it's still useful for "vetting" in that regard

* Requires GT credentials to access.

Lastly, on a logistics front, the course projects focus on C/C++, using specific libraries such as OpenMP, CUDA, and OpenMPI. So, whether or not that is of particular interest to you may also dictate your decision-making accordingly.

Hope this review helps!