Your unfair advantage in cloud certifications
Cloud exams hand you a scenario and four plausible architectures. Choosing well means keeping many similar services straight: which one is serverless, which one is regional, which storage class fits the access pattern, and what each one costs you in trade-offs.
The advantage isn't a shortcut. It's a study method: pulling answers from memory and reviewing them just before they fade, instead of rereading, highlighting and cramming.
13 programs · 10,804 cards · organised by public exam outlines
What your memory has to carry for cloud certifications
Different exams, the same problem: a lot to hold, for a long time, recalled on demand.
Many similar services
Databases, queues, storage classes and networking options whose names and features blur together.
Limits and defaults
Quotas, retention periods, consistency models and the defaults a scenario quietly relies on.
A moving target
Services and exam guides change between versions, so you keep refreshing what you know.
Why the old way fails for cloud certifications
Rereading, highlighting and cramming are the most common ways to study, and they feel productive. The research says they're the weakest.
Rereading
Rereading documentation makes every service sound reasonable. The exam asks you to rule three of them out.
Retrieve it
Tested students recalled 61% of a passage a week later; students who reread it four times recalled 40%.
Roediger & Karpicke (2006), Psychological Science
Highlighting
Highlighting a feature list gives you no practice at telling two services apart.
Test yourself
A review of ten techniques rated highlighting and rereading low utility, and practice testing high.
Dunlosky et al. (2013), Psychological Science in the Public Interest
Cramming
A weekend of cramming before the exam can carry a foundational cert. Associate and professional exams span more services than a weekend can hold.
Space it
Word pairs reviewed across four days were recalled 54% of the time; the stack crammed the day before the test, 34%.
Kornell (2009), Applied Cognitive Psychology
The numbers behind the method
Published finding about the method, not NoReread results. Each one is shown with the conditions it was measured under.
recalled a week later: kept testing vs only restudied
Service-to-feature pairs are pair learning. Continuing to test them is what held a week later.
College students learned 40 Swahili–English word pairs. Pairs that kept being tested after first recall were recalled about 80% of the time a week later; pairs dropped from testing but restudied, 36%.
Karpicke, J. D., & Roediger, H. L. (2008). The critical importance of retrieval for learning. Science, 319(5865), 966–968.
Published finding about the method, not NoReread results.
of learners did better with spacing, while most believed cramming worked better
Cramming feels faster. For nine in ten learners, spacing worked better.
Across three flashcard experiments, spacing beat massing for 90% of participants, yet after the first session 72% believed massing had been more effective.
Kornell, N. (2009). Optimising learning using flashcards: Spacing is more effective than cramming. Applied Cognitive Psychology, 23(9), 1297–1317.
Published finding about the method, not NoReread results.
utility: practice testing and spacing vs rereading and highlighting
Reading docs and highlighting are rated low utility; practice testing and spacing high.
A review of ten learning techniques rated practice testing and distributed practice high utility; rereading, highlighting and summarization were rated low utility.
Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4–58.
Published finding about the method, not NoReread results.
See the research
Three short loops. Two chart published results exactly as reported; the forgetting curve is an illustrative model and is labelled as one.
Spacing beats cramming
From Kornell (2009), Experiment 2: over four days, one stack of 20 word pairs reviewed every day versus four stacks of five each crammed on one day. On day 5, spaced pairs were recalled 54% of the time, massed 21%, and the stack crammed the day before 34%. Across three experiments spacing beat massing for 90% of participants, yet after the first session 72% believed massing worked better.
Kornell, N. (2009). Optimising learning using flashcards: Spacing is more effective than cramming. Applied Cognitive Psychology, 23(9), 1297–1317.
Published finding about the method, not NoReread results.
The forgetting curve, with and without reviews
An illustrative model, not measured data: after one exposure, memory decays quickly toward zero over 30 days. With reviews on days 1, 3, 7 and 14, each review restores it and it decays more slowly each time, staying high at day 30.
Ebbinghaus, H. (1885). Über das Gedächtnis. Duncker & Humblot, Leipzig.
Illustrative model, not measured data.
Testing yourself beats rereading, a week later
Bar charts from Roediger & Karpicke (2006), Experiment 2. Five minutes after studying, students who read a passage four times recalled 83%, read three times and tested once 78%, read once and tested three times 71%. One week later the order reversed: 40%, 56% and 61%.
Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255.
Published finding about the method, not NoReread results.
Cloud & data programs
Ready-made decks, organised by each exam's public outline. Open any program to try its free sample cards.
AWS Cloud Practitioner
4 decks · 961 cards
AWS Solutions Architect Associate
4 decks · 944 cards
AWS Solutions Architect Professional
4 decks · 944 cards
AWS AI Practitioner
5 decks · 520 cards
AWS Machine Learning Engineer Associate
4 decks · 968 cards
Azure Fundamentals (AZ-900)
5 decks · 537 cards
Azure Solutions Architect Expert (AZ-305)
4 decks · 708 cards
Google Cloud Professional Cloud Architect
7 decks · 1,085 cards
Certified Kubernetes Administrator (CKA)
5 decks · 989 cards
Certified Kubernetes Application Developer (CKAD)
5 decks · 928 cards
Terraform Associate
6 decks · 805 cards
Databricks Data Engineer Associate
7 decks · 911 cards
PL-300 — Power BI Data Analyst
4 decks · 504 cards
| Exam | Questions | Time |
|---|---|---|
| AWS Cloud Practitioner65 items (50 scored, 15 unscored) · 90 minutes | 65 items (50 scored, 15 unscored) | 90 minutes |
| AWS Solutions Architect Associate65 items (50 scored, 15 unscored) · 130 minutes | 65 items (50 scored, 15 unscored) | 130 minutes |
| AWS Solutions Architect – Professional75 items (65 scored) · 180 minutes | 75 items (65 scored) | 180 minutes |
| AWS AI Practitioner65 questions (50 scored, 15 unscored) · 90 minutes | 65 questions (50 scored, 15 unscored) | 90 minutes |
| AWS Machine Learning Engineer Associate65 items on the standard exam (50 scored, 15 unscored); the MLA-C02 beta has 85 items · 170 minutes for the MLA-C02 beta (MLA-C01 was 130 minutes) | 65 items on the standard exam (50 scored, 15 unscored); the MLA-C02 beta has 85 items | 170 minutes for the MLA-C02 beta (MLA-C01 was 130 minutes) |
| Azure Fundamentals (AZ-900)Approximately 40–60 questions · Approximately 45 minutes of exam time | Approximately 40–60 questions | Approximately 45 minutes of exam time |
| Azure Solutions Architect Expert (AZ-305)40–60 items · Approximately 100 minutes | 40–60 items | Approximately 100 minutes |
| Google Professional Cloud Architect50–60 items · 2 hours | 50–60 items | 2 hours |
| CKA15–20 hands-on tasks · 2 hours | 15–20 hands-on tasks | 2 hours |
| CKADReal-world tasks performed in live Kubernetes clusters · 2 hours | Real-world tasks performed in live Kubernetes clusters | 2 hours |
| HashiCorp Terraform Associate~57 questions · 1 hour | ~57 questions | 1 hour |
| Databricks Data Engineer Associate45 scored questions (plus unscored pilot items) · 90 minutes | 45 scored questions (plus unscored pilot items) | 90 minutes |
| PL-300 Power BI Data Analyst40-60 items · 100 minutes | 40-60 items | 100 minutes |
How NoReread puts it to work
1. Cards from public exam outlines
Each program is organised by its exam's publicly available outline, so the decks follow the structure you will be tested on.
2. Recall first, then flip
Every card asks before it tells. That moment of pulling the answer from memory is the retrieval practice the research measures.
3. Reviews scheduled before you forget
An SM-2 schedule brings each card back just as it starts to fade: sooner if you missed it, later if you knew it.
Cards contrast one service or feature at a time, taken from each exam's public guide. You recall the answer before flipping, and the schedule keeps services you confuse in front of you until you stop confusing them.
What is the On-Demand pricing model for EC2?
Pay for compute by the second or hour with no long-term commitment or up-front payment.
Learn first, then retain.
Flashcards are a memory tool. Learn the material first, then use NoReread to retain it and refresh what you already know. It doesn't replace studying the subject.
Cards are created from publicly available sources and may contain errors. Check anything you rely on against your official materials.
Straight answers
Is NoReread enough to prepare for cloud certifications?
No, and we won't pretend otherwise. Flashcards are a memory tool. Learn the material first, then use NoReread to retain it and refresh what you already know. It doesn't replace studying the subject.
What does “unfair advantage” mean here?
The advantage is the study method: retrieval practice and spaced review, instead of rereading, highlighting and cramming. It is not a promise of a pass or a score. The numbers on this page are published findings about the method from the studies cited, not NoReread results.
Do I still need hands-on labs?
Yes. Labs build the skill; flashcards keep the service facts and trade-offs in memory. The performance-based Kubernetes exams in particular need hands-on practice.
What happens when an exam version changes?
Each program follows the public exam guide for the version named on its page, and cards may lag behind a new version. Check anything you rely on against the current exam guide.
Can I rely on the cards being accurate?
Cards are created from publicly available sources and may contain errors. Check anything you rely on against your official materials. We're not experts in every field we cover, and NoReread does not guarantee exam results.
Is NoReread affiliated with the organizations behind these exams?
No. NoReread is not affiliated with or endorsed by the organizations whose exams and outlines it references, and exam names are trademarks of their respective owners.
How can I try it?
Every program page shows 6 free sample cards per deck. The All-Access annual plan starts with 7 days free, then $119.99/year, or you can buy a single program for life from $5.99.
Exam names are trademarks of their respective owners. Content disclaimer · All the research
Study cloud certifications the way memory works
7 days free, then $119.99/yr for every program · cancel anytime · or one program for life from $5.99