In distributed systems, quick coding and testing is essential for understanding the complex concepts like Paxos. I developed a small framework to quickly write and test, various replication mechanisms. I use this to teach #distributedsystems. Happy coding! github.com/unmeshjoshi/re…
Unmesh Joshi
323 posts
- I just published "Choosing a Stream Processing framework. Spark Streaming or Kafka Streams or Alpakka Kafka"? link.medium.com/wKFEGkiQFT @lightbend
- A detailed blog post on “Service Discovery With CRDTs” following my @reactivesummit talk with @mushtaqA
- Interesting thread. Knowing what goes on at the code level matters. Abstraction is a disciplined form of ignorance and what we ignore bites back in production.One of the things that surprised me following last week's Jepsen report on Radix DLT (jepsen.io/analyses/radix…) was seeing both blockchain/DLT people *and* the database community go "Hang on, 16 transactions per second can't be right"--and expecting wildly different figures.
- Recently I recorded a podcast with @martinfowler @rebeccaparsons and @zhamakd where we talked about the patterns approach to distributed systems and its relevance in today's technology landscape.
- NEW POST: Sometimes client requests can be responded to only after certain criteria are met based on the responses from several other cluster nodes. Request Waiting List tracks and handles such client requests.
- Recording of my talk at Distributed Systems Conference, 'Patterns of Implementing Consensus' youtu.be/yCMj_dfeO0k @dist_sys @thoughtworks
- We will be discussing 'Kafka's proposed RAFT usage with some sample code to see possible implementation, in the next meetup. @dist_sys @apachekafka meetup.com/Distributed-Sy…
- I will be showing step by step implementation of Gossip Protocols at this meetup meetup.com/Distributed-Sy… #Meetup via @Meetup
- I just published “Aggregate Oriented MicroServices” medium.com/p/aggregate-or…
- "Code is like the mathematics of our profession. It's where we have to remove the ambiguity" I am reminded of this quote from @martinfowler again and again. Writing working code is the real test.
- At @dist_sys meetup, looked at possible code changes in @apachekafka in KIP-500 mode, by building a mini kafka like implementation. Amazing to see how metadata changes will be available as an event stream to all the brokers.
- Thanks to a great write up by @shipilev. (shipilev.net/jvm/diy-gc/). It was fun implementing GC code on JDK12. I have noted some learnings here medium.com/@unmeshvjoshi/…


