Diese Präsentation wurde erfolgreich gemeldet.
Wir verwenden Ihre LinkedIn Profilangaben und Informationen zu Ihren Aktivitäten, um Anzeigen zu personalisieren und Ihnen relevantere Inhalte anzuzeigen. Sie können Ihre Anzeigeneinstellungen jederzeit ändern.

Evolution of message brokers: from helpers to cores

378 Aufrufe

Veröffentlicht am

Codemotion Berlin 2017

Veröffentlicht in: Ingenieurwesen
  • Als Erste(r) kommentieren

Evolution of message brokers: from helpers to cores

  1. 1. Evolution of message brokers. From helpers to cores. Artem Demchenkov
  2. 2. 2 Artem Demchenkov ● CTO at Billie ● ardemchenkov@gmail.com ● https://www.linkedin.com/in/artem- demchenkov-76b69934/
  3. 3. Billie We are an innovative financial platform, specialized on factoring. Some numbers: ● Start: 01.12.2016 ● Closed beta: 01.05.2017 ● Public launch: 01.06.2017 ● Team: 18 ● Engineering crew: 8
  4. 4. is an intermediary program module that translates a message from the formal messaging protocol of the sender to the formal messaging protocol of the receiver. Boring, but necessary theoretical information 4 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Message-oriented middleware (MOM) is software or hardware infrastructure supporting sending and receiving messages between distributed systems. Message broker
  5. 5. A Universe of Message Brokers 5 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 And more...
  6. 6. PAST 6 PRESENT FUTURE ?
  7. 7. Message broker is a helper on a side of an infrastructure 7 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Website RabbitMQ Consumer Past
  8. 8. 8 PAST PRESENT FUTURE ?
  9. 9. Message broker is a core element of an infrastructure. 9 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Message broker Internal tool Microservice Microservice Present
  10. 10. Real-life example. Google Cloud Platform. Present 10 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017
  11. 11. Use case 1. Synchronization of data between parts of a system 11 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Website Google Pub/Sub Salesforce Partner portal
  12. 12. Distribution Values 12 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Allows getting rid of a central data storage. Each application can use an own one. Real-time task forwarding between different systems, responsible people or departments. Improves data consistency. Instantness Consistency
  13. 13. Use case 2. Safe and secure triggering of microservices 13 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Backend Google Pub/Sub Microservice
  14. 14. Values 14 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Safety You won’t loose a message. Each message can be secured.Security
  15. 15. Use case 3. Persistent data storage 15 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Backend Apache Kafka AWS Cloud Storage Google Cloud Storage Microsoft Azure Blob Storage
  16. 16. Values 16 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Persistence You have a full set of historical data. You have a right to choose how to store data, where and in which format. Cloud storage services provide a permission management mechanism to secure data. Flexibility Security
  17. 17. Use case 4. Log aggregation. Commit log 17 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Backend Apache Kafka Instance 1 Storage 1 Instance 2 Storage 2
  18. 18. Values 18 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Replicability Message broker log allows replicating any state of a system at any moment in time You have an full set of historical data. Message broker log can be used by many different clients. Persistence Multifunctionality
  19. 19. Use case 5. Log aggregation. Fulltext search 19 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Backend Apache Kafka Google Cloud Storage Elastic
  20. 20. Use case 5. Log aggregation. Fulltext search 20 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Backend Apache Kafka Apache Solr Cloudera Hue
  21. 21. Values 21 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Searchability Any information in the past could be found. It’s relatively easy to detect wrong messages and define when and how it has been consumed. Discoverability
  22. 22. Use case 6. Delayed messaging 22 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Website RabbitMQ External data provider
  23. 23. Values 23 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Effectiveness Allows consuming a message at the most effective time. Saves traffic, memory usage and disk space.Thriftiness
  24. 24. Use case 7. Multi-cloud application 24 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Google Cloud Storage Google Pub/Sub AWS Cloud Storage
  25. 25. Values 25 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Distribution Allows synchronising data across applications running on other clouds. Allows backing up data across applications running on other clouds. Provides more flexibility in terms of choosing right services for an architecture. Security Variety
  26. 26. 26 PAST PRESENT FUTURE ?
  27. 27. Future 27 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Let’s look again at the current situation.
  28. 28. 28 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Google Pub/Sub RabbitMQ Apache Kafka My prediction is the following: soon projects will use more than one message broker, communicating to each other. Future
  29. 29. 29 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Future Let’s see in 3-4 years, whether I was right
  30. 30. 30 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 Future Conclusion
  31. 31. 31 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 “Because it’s there” George Mallory - English mountaineer, answering a question why did he want to climb Everest. Made 3 attempts and died during the 3rd one in 1924. It is still a mystery whether he had reached the top of it or not.
  32. 32. 32 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 “If you don’t know that something exists, you, probably, don’t need it” Tenzing Norgay - Nepali Sherpa mountaineer. One of the first two people reached the top of Everest in 1953.
  33. 33. 33 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 “No matter what you’ve seen in that presentation, think twice whether you product needs it” Artem Demchenkov - CTO at Billie. I hope, successfully presented these slides on Codemotion Berlin 2017.
  34. 34. Thank you ● ardemchenkov@gmail.com ● https://www.linkedin.com/in/artem- demchenkov-76b69934/
  35. 35. We have just started, but there will be interesting: Check out our “Billie Engineering Corner” on Medium 35 Artem Demchenkov. Evolution of message brokers: from helpers to cores. Codemotion Berlin 2017 https://medium.com/billie-finanzratgeber/tagged/engineering

×