AWS Summit 2024

I went to AWS Summit New York today, a free one day conference. It’s the first time I’ve gone. I didn’t live blog but am writing a summary post of my day after the fact.

Overview

AWS spent a ton of money on this event. They rented out all or most of the Javits Center in NYC (this is where NYC Comic Con is held). They gave coffee/soft drinks and even free lunch. They also spent a lot of money on security. For cause. There were protesters right outside the front door.

I tried to experience the major parts of the event.

Expo

The exhibit hall was large on the third floor with lots of vendors related to cloud. There were also some fun activities like a drone and toy car racing. Lots of space for sitting/networking.

There were also some stages in the expo for shorter (15-30 minute talks). They had headphones for people who couldn’t filter out the background noise of the expo. It was nice because you could flit by and see if you were interested. I listened to some pieces of cert/education talks and a full one from Elastic on LLMs and summarizing security incidents

Breakouts

There were lots of one hour breakout sessions on the first floor. I went to two customer success stories (Venmo and Fannie Mae). It was dark in the breakout rooms. Like most places have for keynotes

Learning highlights for Venmo

Key strategies

  • Distribute load to maximize processing throughput
  • Use event based systems for anything not in critical path

Other notes

  • Django app. Used Celery for async work,  reader db instanes for queries that can use
  • Then added DynamoDB, MongoDB, OpenSeach Service, data lake, microservics, Cassandra (for microservices), Kafka
  • Split Mysql into Auroa MySQL comatible secondary ySQL and Analytics MySQ databases

Social feed data migration

  • Transactions visible, high traffic because home screen
  • Every transaction geerations a feed story along wih certain profile operations
  • 3.6TB of data, 5.6 bllion entries
  • Since digit lateny on data retrieval
  • 90% of memory usage
  • switched to DynamoDB due to cost (90% less), performance (equivlanet), managed servie, data encrption at rest, integration with other AWS offerings
  • Migrated via backfill followed by dual writes. Let verify performne under pro load and confirm data consistent. Then started ramping reads on new database. Started with 1% reading from new DynaeoDB. Finally cut off writes to ol MongoDB

Offloadng transaction history

  • For each payment put message on Kafka queue and write to Cassandra via microservice.  Implemented as best effort write Needed to guarantee 100% of data so could move over use caes taht required full fidelity data
  • Switch to write ahead log – write log essage saying intend to peror action and store in DynaeoDB Then proess transaction/pblish essage. FInally, delete inteded action message ow that completed. Background process looks for pendin messages 
  • Asyc payment processin using Kinesis
  • Problem batches huge and inconsistent for credit car sage, delays, outage costly, can’t send 500 error/need to reconcile, not a way to replay transactions internally
  • Added Kineis Data Stream via think wrapper to put mesage on strea and ackowledge success to upstrea. From KInsis, have consumers/lambda procss. Also usig Auora, DocumentDB, ElastiCache, DynamoDB and SQS

Key learnings for Fannie Mae

data science research 

  • compared research vs deveopment – ex: research has poc, live prod data, latest tools/patterns
  • pilars of platform:
  • data access – prod data, data usage contracts
  •  governance  – control by business, not tech, autoamted integration with governace
  •  operationalization – testing, validation, Ci/CD
  • data science controls
  • register research activities in CMDB so can provision/tag resources. Automated provisioning, strealined architect review process
  • Data access.sharing  contracts, perissions, ingress/egress rules, sensitive data protection rules
  • Cde deployment and change managment  CI/CD, scanning
  • Data science platform architecture
  • code/image repo
  • pblic data endpoints
  • code/package library
  • read only access to enterprise data lake
  • research envs –
  •  collaboration – just in time access – read only access to prod enterprise data lake. results an’t be shared; considered dev
  • validation  – testing/shakeot – still read only
  • operaiton – headless execution/- now can write to prod, create reports and share exterally
  • data access JIT (just in time). Fannie Mae has a patent on this
  • request access to data. could be from many data sources
  • JIT access engine checks against coarse grained contracts
  • Then goes to policy manager to check fine graine access controls. Use UI to create rules. creates new role dynamically so can use token to access

Building a generative AI use case

  • Used Anthropic’s  Claude 3 Sonnet via Amazon Bedrock and Aazon Neptune (graph db)
  • A lot of analysis of unstructured documents, average of 5 hours per doc and 8K dos per year
  • Deep Insight for LLM driven knowledge extraction. Uses ontology (schema( an LLM t generate knowledge graphs. Human in the loop to validate Then knowledge utiilization step to use natural language via a chatbot
  • taxonomy – linear top down hierarchy. Ontoogy – interconnected network representation
  • Disambigution important to avoid duplication
  • graph database  
  • reduces risk of hallucinations because more context
  • two types – 
  • Property Graph (Apache Tinkerpop) . Query with Gremlin or Cypher
  • RDF Graph (from W3C). query with SPARQL
  • extraction uses Bedrock, fargate, lambda, neptune, s3
  • utilization uses – bedrock, fargate, neptune and a chatbot
  •  also uses LangChain – Neptune Open Cypher QA chain (converts natural langague queries into Cyper so can do query(  and Amazon OpenSearch
  • challenges
  • pick onthology framework – Chose Turtle (Terse RF Triple Language for reeasability/ease of reading
  • find best way to chunk. Chose at sections so handle complex tables btter
  • Picking graph type. Chose property graph due to better OSS framework support
  • Amazon Kendra (enterprise search( did not integrate with Amazon Neptune. Used LangChain’s NeptuneOpenCypher QA Chain instea

Chalk Talks

Chalk talks were also on the first floor. They were also an hour but had less prepared content. The one I went to had 20 minutes of talking/demos. Most of the time was Q&A or discussion. They had a whiteboard with a camera to show what was on it so the speakers could write/draw real time. This meant one projected screen was the computer and one was the physical whiteboard.

Learning highlights

  • gen customers what to know what model to use, how to move quickly and how keep data secure/private
  • Bedrock provides foundational models via single API, customize model, RAG (Retrieval Augmented Generation), agents for multi step tasks, security/privacy/safety
  • Models include – amazon’s models, anthorpic,, cohere, meta, etc. ANd lots of variants/versions of each.
  • Two use cases: observability of generative AI itself, using gen AI to help with observability
  • gather metrics – ex: number tokens used for input/output
  • collected metadata/requests/responses so understand how customers use
  • governance/controls/guardrails
  • Cloudwatch – analyze inovcation logs, protect sensitve date, real time metrics and alarms (Ex: more latency on different version of claude), single pane of glass/dashboard
  • recorded demo #1 (while video was recorded, he narrated live. also paused periodically to say more
  • can send model invocation logs to either s3 (if using other loggiing system) or cloudwatch

Builder Sessions

Also on the first floor, these were small group labs. I went to one on Amazon Q. They had 4 areas on the room with 10 chairs each. An instructor from AWS was allocated to each group. After a short intro, the instructor helped anyone stuck and answered questions. This was great.

The lab had an access code good for three hours so you continue a little longer if you wanted. In theory, there was separate wifi for the lab but it didn’t work. The main conference wifi was fine though.

Learning highlights

  • Amazon Q Developer has a free and paid version.
  • The paid version promises not to learn from your data, It’s licensed per person but only billed if the developer uses in a month.
  • IDE integration for VS Code and IntelliJ.
  • Chat bar. Often gives sources/links. From 2023 for public internet. RAG for Amazon so more recent
  • Can explain code, refactor code, fix code and migrate to later version of Java. Can also write a plan for writing code and write code (with some errors)
  • Code Whisperer was folded into Q
  • It was slow, but I was on a conference network

Main dev activities

  • planning – docs, examples, deisgn
    • creating = generate cpde,amage omfra
    • test amd secure – test cases, scan for security vulnerabiliteies
    • operate – identify and mitigate code issues, monitor performance and efficiencey
    • maintenance and modernization – modernize and update old code languages and dependencies

Amazon Q Developer tries to help with all phases

  • plan – explain code with conversational coding (chatbot)
  • create – inline code complete, conversational coding
  • test/secure – unit test generation, OWASP top 10 security scanning
  • operate – debug/optimize code with conversational coding
  • maintenance and modernatization  update code with agent from legacy

Keynote

The keynote was in a big room that wouldn’t fit everyone. They also used all the breakout rooms as overflow and streamed to the stages in the expo. I like that as it was easy to eat and listen. Or talk to the vendors and listen to parts. Or not.

How I recommend studying for the AWS Associate Architect Exam

Studying for the AWS Associate Architect Exam? Keep reading for what to expect as I share my tips for passing the exam on the first try *and* making the best use of your time. Also see:

Choosing the Right Associate Exam

There are three Associate level exams. The Architect and Developer exams have a lot of overlap. The Sys Ops one is very different. Which means those reading this page are probably deciding between the Architect and Developer exams!

If you look online, different people will say each of the associate exams is the “hardest”. The problem with “hard” is similar to the problem with “easy”.

This was my advice when I choose the Developer exam to take first. I completely agree with it now that I’ve taken both!

In a nutshell, I’d say developers will probably find the developer one more interesting. Those with a networking or architecture background will probably find the architect one more interesting. Another tip is to look at the ACG (A Cloud Guru) outline for the architect and developer exams to see which you find more interesting. For example, Dynamo was one of my favorite topics and VPCs one of my least favorite so Developer was clearly the right choice!

Checking the version numbers

As of September 2020, the current version is the SAA-C02. I do not recommend using materials for the SAA-C01 unless you are using them in very specific ways (ex: to learn a specific topic)

Official Study Guide/Outline

Amazon’s official Solution Architect Associate page has an outline and sample questions.

Jeanne’s study notes

While I’m posting my notes, they are just to give you a feel for the types of things you should know. They are not meant to learn from.

A Cloud Guru (ACG)

John Bonso Practice Exams

John Bonso’s practice exams on Udemy are $13 for 6 exams. (Udemy runs frequent sales to get it to that price) Each question has a detailed explanation to help you learn.

Amazon Practice questions

  • If you’ve taken another AWS exam (like the Practitioner), you get to take a 20 question timed practice exam for free. Go into the cert dashboard using your login to get a coupon code for this.

Studying for AWS Associate Architect in the Time of COVID-19

Last year, I took the AWS Cloud Practitioner and AWS Associate Developer. This year, I was supposed to take the AWS Associate Architect. [Edit: got a 760. Passing is 720]

Also see

How I recommend studying for the AWS Associate Architect

Why I took the exam in September

I wanted to finish working on our second OCP 11 book first. I was ok with taking the AWS cert at the same time we were working on our practice tests book.I find the practice book easier to work on because questions are independent (well mostly) so I don’t need as large a chunk of time to make progress.

I also wanted to wait for the robotics competition season to be over as I was going to busy with that 4/8 weekends in March/April. I turned out to only be busy one of those weekends. Then COVID-19 hit NYC. For the first 11 weeks, I was barely a functioning person. I know some people were able to use those weeks of isolation to study. I am not one of them. it was a miracle I could get up and do my job (well most of it) every day.

In early summer, I still wasn’t considering taking the exam because testing centers in NYC were not open yet. I am not willing to take the exam online. In mid-August, I noticed testing centers were open. I still didn’t feel up to it, but I studied anyway.

I went into the Practitioner and AWS Developer not knowing I’d pass and figuring I’d take it again if I failed. That won’t work here because Oracle announced the 1Z0-819 exam for Java and I’m taking that on 9/26. So this was my only attempt.

What was different at the exam center

  • There was a sign on the door saying that the door was locked and do not enter. The sign said the proctor would come out at the exam start time (9, 11:30, 1, and 2) to let you in so they could clean. So I stayed outside. A few minutes to 1:00, someone walked right in. I went in and there were 4 people in the waiting room. I might have been waiting outside forever.
  • Only some exams are requiring a picture. (I don’t remember if AWS did last time either) Mine did not, but people taking a picture did have to remove the mask for a minute
  • I didn’t have to sign anything.
  • They took everyone’s temperature before letting you into the exam room. (But not before letting you into the waiting room)
  • The first computer he was going to assign me to didn’t work. He moved me to a different computer that was right across from someone not wearing a mask (maybe 4 feet away). I said something and the proctor made her put the mask back on.
  • I was given paper and pencil. (I don’t remember if I was given that for the AWS Associate. I might have been)

Why this was the hardest exam (for me) to study for

While things have been better since Memorial Day, I don’t know if I’d all them good. One of the problems I have with working from home full time during pandemic is remembering stuff. I remember things spatially. And apparently, I can only remember so much that happens in the same place in a day. So during the week, work “takes” all of that. So I’ve found it incredibly difficult to retain anything and had to learn the same things over and over and over…

Last year, I was given a choice of which Associate exam to take. I picked Developer. It matches my background and interests best which makes it easier to study for. Alas, that means this one is harder to study for. I work for a large company. I’m not going to be personally designing a 100 petabyte data transfer or setting up the VPN. So it’s hard to convince myself this is important for me to know.

The practice materials (ex: A Cloud Guru, John Bonso) test you on a lot of details and number. Which suggests you need to know them. At least on my exam, this was not the case. The only numbers that were important had to do with data migration. Luckily, I got enough questions on the topic that I could “reverse engineer” the answers by cross referencing the questions.

I also got frustrated during (practically) every mock exam when I couldn’t remember stuff I knew I “should” be able to. This gave my brain the opportunity to freak about about all the other stuff I’m worried about (going back to a coronavirus winter and the like)

How long was the exam

You get 130 minutes. I used 70 minutes on my first pass. (I had to stop a few times to relax though due to frustration about not remembering stuff I felt I should. I just closed my eyes for a few minutes in the exam center and then continued. I then spent 5 minutes switching answers where I got the answer based on a question later in the exam. Finally, I spent 5 minutes starting to review my answers. I stopped because I was worried about changing a right answer to the wrong answer.

This was a good amount longer than the Associate Developer took me. Probably because there was more reading in some of the questions.

The actual exam

I got 5 questions on one topic (Direct Link and Storage Gateway). I also got two questions on SQS that were almost the same. Luckily this was a topic I felt comfortable answering so I know I got both. But AWS has a big test bank. That doesn’t feel representative. But I didn’t get a ton of questions on VPCs (one of my least favorite topics) so who am I to complain.

As I went through the exam, I noted which questions I was unsure of. The answer was 35. Which was more than half of them. Enough of those I was able to narrow down to two options though so statically I should have come close to passing.

Test taking techniques

Applying test taking techniques definitely helped my score. In particular:

  • Write down facts that I’m unsure of. That lets me go back to that question if another question in the exam answers them.
  • Process of elimination. There were *many* questions where you could rule out two of the four answers without even reading the question. For example, there is no such thing as a LIFO queue. (That would be a stack).
  • Look for keywords in the question. For example, if the question manages cost or performance, you know what to look for.

The surprise at the end

I pressed “End test” and nothing happened. The countdown kept going. Well “something” happened. I could no longer click previous/next or anything else. I showed the proctor. He closed the window and said he’ll look at it. He closed the browser and told me to wait in the lobby, he did something with the computer and said that I passed. I don’t see that from Amazon online yet (it takes a few days) so I’ll take his word for it!

So phew. I don’t have to take it again. However, this meant I was separated from my notes for longer than usual at the end so don’t have as good a grasp of what I wanted to remember from the exam. Luckily I passed so it isn’t that important. Or I should say, I allegedly passes. The proctor saw the “pass” result, not me.

Did I think I passed?

The AWS exams have been interesting in terms of whether I thought I passed when I clicked the “end test” button:

  • Practitioner – I had no idea -> Passed
  • Associate Developer – Confident that I passed -> I did pass, but didn’t do as well as on the Practitioner.
  • Associate Architect – Didn’t think I passed -> passed. (I replied more on test taking skills than knowledge)

Something I learned during the exam

Around the middle of the exam, I learned that the “side effects” of working form home full time during pandemic are going to follow me back to work for a while. It wasn’t as intense as when i was at home practicing, but I definitely felt the frustration of not being able to recall info that I studied umpteen times. Which means when I go back to working at the office, I’m likely to feel the same on information I should be learning now. Sigh.

What I used to study

  • Wiley’s AWS Associate Architect Kit – Note that I used the version for the previous exam (SAA-C01) even thought I was taking the SAA-C02. I like learning from books. And now that so much of life is online, I’d like to spend less time on the computer, not more. So I figured I could read from the old book while sitting outside and make what I needed to learn on the computer easier. Wiley is publishing a SAA-C02 version of the book soon. The books definitely helped me understand the material even if I wasn’t retaining it as much as I’d have liked. It would have been easier using the right version of the book though. (Note: I write a book in this series for Java)
  • The A Cloud Guru course – I like that they added subtitles and transcripts to the videos. (And that they’ve always let you speed up the videos). This makes learning from video somewhat better for me. My employer paid for the description. I paid for myself last year and crammed everything I wanted to watch into a month. I did notice the “practice test” had a bunch of questions from each module. Which made it feel like an inaccurate reflection of how I was doing. Also, they added SAA-C02 videos after the course existed. Which means some sections have the original video which says you don’t need to know things or it is the last video in the section when it is not.
  • John Bonso’s practice exams on Udemy – Remember to check for Udemy discounts. They usual (if not always) have them at half price. Which made this 6 exams for $13.
  • Amazon’s official practice exam. This is only 20 questions. But you get a free code to take it if you’ve passed a prior exam.
  • Amazon’s free practice questions

What I didn’t do

  • Follow on with the labs hands on. I started doing that for the Associate Developer and stopped. It’s a good way to learn the content. It’s a terrible investment of time (for me) to retain stuff for the exam. So this time I didn’t even try.
  • Play with the A Cloud Guru sandbox. It looks really cool. And I’ll definitely use it when I want to learn something. But studying for an exam requires a different skill set for me.
  • Read the A Cloud Guru recommended FAQs (S3 and ELB), whitepapers, Re-Invent videos, etc. I did that for the Associate Developer. It was a terrible use of time. They are interesting, but reading a long FAQ feels a little like “read the dictionary; you’ll learn to spell.” That’s why I paid money for a book/etc. To not read everything. Similarly for Java, one could read the entire Java Language Specification or they could read my book. Besides in this case, I couldn’t even retain the stuff in the study guides. Adding more to that pile was never going to help.

What I did the last minute

Since trying to learn the material and get it into my long (or even medium term) memory wasn’t working the day before the exam, I:

  • Didn’t do anything technical/problem solving at work to store up energy. (One of the parts of my job is to be our team Scrum Master. The day before the exam was sprint planning day so a good part of the day was SM stuff anyway. Then I did some paperwork in the afternoon and left early)
  • Went for a walk with someone in the neighborhood. (I find that my brain works better when i haven’t been alone for days on end and don’t have that to look forward to)
  • Read my study notes
  • Did full mock test in Wiley practice test book
  • Read my study notes
  • Do last Wiley online practice test
  • Watch all the A Cloud Guru summary views and do the practice questions for each section.
  • Read my study notes
  • Final Bonso exam
  • Read my study notes

And the day of my exam:

  • Read my study notes
  • Try to convince myself I am ready
  • Repeat A Cloud Guru challenge exams
  • Spend two hours at the local park weeding and throwing out trash. (I actually rescheduled my exam from 9am to 1pm to do this). Another thing to help convince my brain things are “normal”
  • Read my study notes
  • Walked to exam center. I had to wait. So… read my study notes

Practice scores

Like last AWS exam, I’m sharing my scores because it’s hard to determine if you are ready.

SourceScore
ACG – 10K Foot Overview100%, 100%
ACG – IAM & S3100%, 94%
ACG – IAM & S3 (2nd quiz)79%, 89%
ACG – EC275%, 79%
ACG – Databases62%, 69%
ACG – Route 5378%, 78%
ACG – VPCs61%, 79%
ACG – HA Architecture89%, 79%
ACG – Applications88%, 100%
ACG – Serverless44%, 56%
ACG – Challenge Quiz 160%, 81%
ACG – Challenge Quiz 263%, 77%
ACG – Practice test78%
Wiley Assessment Test46%
Wiley Chapter 1100%
Wiley Chapter 285%
Wiley Chapter 380%
Wiley Chapter 470%
Wiley Chapter 562%
Wiley Chapter 675%
Wiley Chapter 755%
Wiley Chapter 880%
Wiley Chapter 945%
Wiley Chapter 1060%
Wiley Chapter 1125%
Wiley Chapter 1270%
Wiley Chapter 1355%
Bonso Test 153%
Bonso Test 269%
Bonso Test 366%
Bonso Test 469%
Bonso Test 564%
Bonso Test 678%
Wiley Practice Chapter 166% (combined)
Wiley Practice Chapter 268% (combined)
Wiley Practice Chapter 373% (combined)
Wiley Practice Chapter 480%
Wiley Practice Chapter 570%
Wiley Practice Mock Exam86%
Wiley online practice exam 160%
Wiley online practice exam 250%
Oracle free sample questions50%
Oracle official practice exam75%

And when I studied

DateSource
7/30 & 8/4First four chapters of Wiley (was traveling) including assessment test
8/5Wiley chapter 5
8/6Wiley chapter 6
8/7Wiley chapter 7
8/8Wiley chapter 8
8/9ACG Intro + Wiley chapter 9
8/10ACG – 10K foot overview + Wiley chapter 10
8/11ACG – Start IAM & S3 + Wiley chapter 11
8/12ACG – Finish IAM & S3
8/13-8/14ACG – EC2
8/15ACG – Databases + Wiley chapter 12
8/16ACG – Advanced IAM + Wiley chapter 13 + Bonso test 1
8/17ACG – Route 53 + Wiley practice chapter 1 (questions 1-50)
8/18ACG – VPCs + Wiley practice chapter 1 (questions 51-100)
8/19Skip day
8/20Wiley practice chapter 1 (questions 100-150)
8/21Wiley practice chapter 1 (questions 150-200)
8/22Wiley practice chapter 1 (questions 200-250)
8/23Skip day
8/24Start ACG – HA Architecture + Wiley practice chapter 1 (remainder)
8/25Finish ACG – HA Architecture + Wiley practice chapter 2 (questions 1-50) + official Oracle sample questions
8/26ACG – Applications + Bonso test 2
8/27ACG – Security + Wiley practice chapter 2 (questions 51-100)
8/28ACG – Serverless + Wiley practice chapter 2 (questions 101-150) + Bonso test 3
8/29ACG – Good luck videos + 2 quizzes
8/30ACG – Practice test+ Wiley practice chapter 2 (remainder) + AWS official practice exam
8/31Wiley practice chapter 3 (questions 1-120 )+ Bonso test 4
9/1Wiley practice chapter 3 (remainder) + Wiley online practice test 1
9/2Wiley practice chapter 4 + Bonso test 5
9/3Wiley practice chapter 5 + Wiley online practice test 2
9/4Last day (see above “last minute” section for what did)
9/5Test day (see above “last minute” section for what did)