fast but not loose
play

Fast but Not Loose: Typesafe Clients in a Distributed Service - PowerPoint PPT Presentation

Jun 30, 2023 •232 likes •1.11k views

Fast but Not Loose: Typesafe Clients in a Distributed Service Architecture, a retrospective #gotocon #gotoaar #gilttech Eric Bowman @ebowman VP Architecture @ Gilt Groupe ebowman@gilt.com Monday, September 30, 13 Monday, September 30, 13

  1. Fast but Not Loose: Typesafe Clients in a Distributed Service Architecture, a retrospective #gotocon #gotoaar #gilttech Eric Bowman @ebowman VP Architecture @ Gilt Groupe ebowman@gilt.com Monday, September 30, 13

  2. Monday, September 30, 13

  3. Monday, September 30, 13

  4. • Scala • Aster Data • Play • Mahout • PostgreSQL • Jersey • MongoDB • SBT • Voldemort • Docker • Kafka • Continuous Delivery Microservices Monday, September 30, 13

  5. http://upload.wikimedia.org/wikipedia/commons/1/16/Ruby_on_Rails-logo.png http://wiki.postgresql.org/wiki/File:PostgreSQL_logo.3colors.svg Monday, September 30, 13

  6. http://logonoid.com/images/thumbs/christian-louboutin-logo.jpg Monday, September 30, 13

  7. http://data.iluxdb.com/data/christian-louboutin-daffodile-160mm-python-masai-1130127cm09_001.jpg?dd80c0 Monday, September 30, 13

  8. http://upload.wikimedia.org/wikipedia/commons/7/79/Operation_Upshot-Knothole_-_Badger_001.jpg Monday, September 30, 13

  9. Monday, September 30, 13

  10. Monday, September 30, 13

  11. Monday, September 30, 13

  12. Monday, September 30, 13

  13. Monday, September 30, 13

  14. Monday, September 30, 13

  15. Monday, September 30, 13

  16. http://blog.verwilst.be/wp-content/uploads/2008/12/java.gif Monday, September 30, 13

  17. swift user_service order_service payment_service product_service cart_service pagegen Monday, September 30, 13

  18. http://megmurph.com/wp-content/uploads/2013/03/success-1.jpg Monday, September 30, 13

  19. Service X Front End Service Y Service Z Monday, September 30, 13

  20. Service X Front End Service Y Service Z Monday, September 30, 13

  21. Content from Service X Service X Content from Z Front End Service Y Content from Service Y Service Z Monday, September 30, 13

  22. Front-End Tier Service Tier Data Tier Monday, September 30, 13

  23. Caching Front-End Tier Light Computation Orchestration Service Tier Data Tier Monday, September 30, 13

  24. Caching Front-End Tier Light Computation Orchestration Caching Service Tier Heavier Computation Separation of Concerns Data Tier Monday, September 30, 13

  25. Caching Front-End Tier Light Computation Orchestration Caching Service Tier Heavier Computation Separation of Concerns Data Tier Data Access (Disk/SSD/RAM) Monday, September 30, 13

  26. Caching Front-End Tier Light Computation Orchestration Caching Service Tier Heavier Computation Separation of Concerns Data Tier Data Access (Disk/SSD/RAM) • Runtime Temperature • Development Temperature Monday, September 30, 13

  27. Service Front End Javascript Data Source Application JDBC ➾ HashMap ➾ JSON ➾ HashMap ➾ JSON JSON http://gilt.com Monday, September 30, 13

  28. Service Front End Javascript Data Source Application JDBC ➾ HashMap ➾ JSON ➾ HashMap ➾ JSON JSON http://gilt.com Chaos Grows Quickly Monday, September 30, 13

  29. Front End Service JSON ➾ HashMap ➾ JSON Data JDBC ➾ HashMap ➾ JSON Source Service Javascript Application JSON ➾ HashMap ➾ Data JSON JDBC ➾ HashMap ➾ JSON Source http://gilt.com Service JSON ➾ HashMap ➾ JSON Data JDBC ➾ HashMap ➾ JSON Source Monday, September 30, 13

  30. Front End Service JSON ➾ HashMap ➾ Service JSON Data JDBC ➾ HashMap ➾ JSON Source JSON ➾ HashMap ➾ JSON Javascript Service Application JSON ➾ HashMap ➾ JSON ➾ HashMap ➾ JSON JSON Data JDBC ➾ HashMap ➾ JSON Source http://gilt.com Service JSON ➾ HashMap ➾ JSON Data JDBC ➾ HashMap ➾ JSON Source Monday, September 30, 13

  31. http://www.recruitmenttakeout.com/wp-content/uploads/2013/04/brian.png Monday, September 30, 13

  32. • Fast • Service Decomposition • Implicit core model was good Monday, September 30, 13

  33. • Org Scaling • APIs • Implicit core model was ... implicit Monday, September 30, 13

  34. A data model and APIs for services Monday, September 30, 13

  35. A data model and APIs for services (aka, RPC) Monday, September 30, 13

  36. core service core service core service core service Legacy http://git-scm.com/images/logos/downloads/Git-Logo-2Color.png Monday, September 30, 13

  37. core service commons.jar core core clients service core data core model service async client framework core service Greenfield Legacy http://git-scm.com/images/logos/downloads/Git-Logo-2Color.png Monday, September 30, 13

  38. core service commons.jar • Users core • Sales core clients service • Products core data • Skus core model service • Assets async client • Targeting framework core • Auth service Greenfield Legacy http://git-scm.com/images/logos/downloads/Git-Logo-2Color.png Monday, September 30, 13

  39. service client core Monday, September 30, 13

  40. • RESTful service • Scala clients • All APIs futures-based client • Case class schema core Monday, September 30, 13

  41. service core Compile/Runtime Dependency client client service core Monday, September 30, 13

  42. Consumer Service Client Core Core “Embassy Soil” Monday, September 30, 13

  43. Consumer Service Client Core Core • Easy functional testing • Service response capture • Test linking • Upgradable • Emergent Regression • Automated upgrades • Compile farmers Monday, September 30, 13

  44. Consumer Service Client Core Core Monday, September 30, 13

  45. Consumer Service Client Core Core • Environment ⊕ Config • Live updates • Indirection • Circuit Breaker Monday, September 30, 13

  46. Monday, September 30, 13

  47. “All of this is completely wrong.”* *Not an actual quote Monday, September 30, 13

  48. 1.The network is reliable 2.Latency is zero 3.Bandwidth is infinite 4.The network is secure 5.Topology doesn't change 6.There is one administrator 7.Transport cost is zero 8.The network is homogeneous https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  49. 1.The network is reliable https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  50. d e v l o 1.The network is reliable s e R https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  51. 2.Latency is zero https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  52. d e v l o 2.Latency is zero s e R https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  53. 3.Bandwidth is infinite https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  54. d e v l o 3.Bandwidth is infinite s e R https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  55. 4.The network is secure https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  56. d e v l o 4.The network is secure s e R https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  57. 5.Topology doesn't change https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  58. d e v l o 5.Topology doesn't change s e R https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  59. 6.There is one administrator https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  60. d e v l o 6.There is one administrator s e R https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  61. 7.Transport cost is zero https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  62. d e v l o 7.Transport cost is zero s e R https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  63. 8.The network is homogeneous https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  64. d e v l o 8.The network is homogeneous s e R https://blogs.oracle.com/jag/resource/Fallacies.html Monday, September 30, 13

  65. “Still Wrong.”* *Possibly an actual quote Monday, September 30, 13

  66. Convenience Over Correctness http://steve.vinoski.net/blog/2008/07/01/convenience-over-correctness/ Monday, September 30, 13

  67. Public Data Model Client Layer Client Data Model Client Backend Client/Server Serialization Model Service Frontend Service Model Service Backend Database Model Database Monday, September 30, 13

  68. Public Data Model Client Layer • So many models Client Data Model Client Backend • Corners are cut Client/Server Serialization • Typesafe helps Model Service Frontend • Conflation? Service Model • Just the data Service Backend Database Model Database Monday, September 30, 13

  69. Public Data Model Client Layer • So many models Client Data Model Client Backend • Corners are cut Client/Server Serialization • Typesafe helps Model Service Frontend • Conflation? Service Model • Just the data Service Backend Database Model Database “Works in Practice for some use cases” Monday, September 30, 13

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

A Coq - definition al implementation of the Lax Logical Framework LLF P , for fast and loose

A Coq - definition al implementation of the Lax Logical Framework LLF P , for fast and loose

A Coq - definition al implementation of the Lax Logical Framework LLF P , for fast and loose reasoning F. Alessi, A. Ciaffaglione, P. Di Gianantonio, F. Honsell, M. Lenisa name.surname@uniud.it Department of Mathematics, Computer Science,

650 views • 30 slides
Respondents Playing Fast and Loose?: Antecedents and Consequences of Respondent Speed of

Respondents Playing Fast and Loose?: Antecedents and Consequences of Respondent Speed of

Respondents Playing Fast and Loose?: Antecedents and Consequences of Respondent Speed of Completion Randall K. Thomas & Frances M. Barlas GfK Custom Research Sub-optimal Response in Surveys Survey satisficing occurs when respondents

496 views • 49 slides
Parametricity Types Are Documentation Tony Morris The Journey Fast and loose reasoning is

Parametricity Types Are Documentation Tony Morris The Journey Fast and loose reasoning is

Parametricity Types Are Documentation Tony Morris The Journey Fast and loose reasoning is morally correct Danielsson, Hughes, Jansson & Gibbons [DHJG06] tell us: Functional programmers often reason about programs as if they were written

1.41k views • 72 slides
Good Data Gone Bad: playing fast and loose with data Colin Rowat DiNardo and Tobias (2001, JEP)

Good Data Gone Bad: playing fast and loose with data Colin Rowat DiNardo and Tobias (2001, JEP)

Good Data Gone Bad: playing fast and loose with data Colin Rowat DiNardo and Tobias (2001, JEP) How do we explain this bimodal distribution? the bimodal distribution survives slicing of the data: e.g. recruits from 1851 1855 or from 1856

412 views • 12 slides
Loose in the Street Yard Material Pile Collection Service Levels Review October 30, 2018 1

Loose in the Street Yard Material Pile Collection Service Levels Review October 30, 2018 1

Loose in the Street Yard Material Pile Collection Service Levels Review October 30, 2018 1 Focus on the Loose in the Street Yard Material Collection o Until July 2016, Davis Waste Removal (DWR) collected loose in the street (LITS) yard

115 views • 8 slides
Announcements Exam 2: 03/11, PA2______,HW3______. Today: Loose ends Associative Arrays Hash

Announcements Exam 2: 03/11, PA2______,HW3______. Today: Loose ends Associative Arrays Hash

Announcements Exam 2: 03/11, PA2______,HW3______. Today: Loose ends Associative Arrays Hash Tables Intro to Hash Functions Loose ends Classic balanced BST structures: Red-Black trees max ht 2log 2 n. Constant # of rotations

298 views • 8 slides
In any organization there is always a lot of loose, unused power lying about which can be

In any organization there is always a lot of loose, unused power lying about which can be

In any organization there is always a lot of loose, unused power lying about which can be picked up without alienating anyone, said Demara, who also posed as a prison guard, a professor, a monk, a deputy sheriff. Make your own rules

727 views • 23 slides
What is Loose Parts, Nature Play? Children are invited to engage in unstructured

What is Loose Parts, Nature Play? Children are invited to engage in unstructured

Loose Parts, Nature Play August 2013 Ilisa Goldman, RLA San Diego Children and Nature Collaborative What is Loose Parts, Nature Play? Children are invited to engage in unstructured outdoor play with a collection of nature

83 views • 5 slides
Community Update MST T Fast st Facts cts MST T Fast st Facts cts MST T Fast st Facts

Community Update MST T Fast st Facts cts MST T Fast st Facts cts MST T Fast st Facts

Monter nterey-Sali Salinas nas Transit ansit Community Update MST T Fast st Facts cts MST T Fast st Facts cts MST T Fast st Facts cts MST T Fast ast Facts acts 231,706 miles between preventable accidents 89% on time

198 views • 19 slides
Schools Out for Summer! But. Not for You! It is time for you to cut loose and plan

Schools Out for Summer! But. Not for You! It is time for you to cut loose and plan

Schools Out for Summer! But. Not for You! It is time for you to cut loose and plan aheadand get ready to close this year and prepare for next school year! Introduction/Goals Welcome and thank you for joining us today. Introduction of

1.24k views • 109 slides
Loose Ends: Side Channels, Government Surveillance & Targeted

Loose Ends: Side Channels, Government Surveillance & Targeted

CSE 484 / CSE M 584: Computer Security and Privacy Loose Ends: Side Channels, Government Surveillance & Targeted Attacks Spring 2015 Franziska

622 views • 27 slides
Spring in Action Agenda What is Spring? Loose coupling with Dependency Injection

Spring in Action Agenda What is Spring? Loose coupling with Dependency Injection

Spring in Action Agenda What is Spring? Loose coupling with Dependency Injection Declarative programming with AOP Eliminating Boilerplate with templates E-mail: craig@habuma.com Blog: http://www.springinaction.com Twitter:

584 views • 22 slides
Tying up loose strands: Defining equations of the strand symmetric model Colby Long and Seth

Tying up loose strands: Defining equations of the strand symmetric model Colby Long and Seth

Tying up loose strands: Defining equations of the strand symmetric model Colby Long and Seth Sullivant North Carolina State University June 8, 2015 Colby Long (NCSU) Tying up loose strands: Defining equations of the strand symmetric model

685 views • 21 slides
1 04/12/12 Too much food and too many birds The birds loose their fear Goose dung

1 04/12/12 Too much food and too many birds The birds loose their fear Goose dung

04/12/12 Problem, what problem? Geese at Watermead Country Park (South) Campaign 2012 What are they all doing there? 1 04/12/12 Too much food and too many birds The birds loose their fear Goose dung creates a of people and get

358 views • 3 slides
DIMENSIONALITY REDUCTION AND VISUALIZATION Loose ends from HW2 Hyperparameters, bin size =

DIMENSIONALITY REDUCTION AND VISUALIZATION Loose ends from HW2 Hyperparameters, bin size =

DIMENSIONALITY REDUCTION AND VISUALIZATION Loose ends from HW2 Hyperparameters, bin size = 1000, 500, ? Tune on test set error rate Variance of a recognizer Accuracy 100%? 98? 90? 80? Whats the mean and variance of the

976 views • 81 slides
Learning outcomes First session Overview of IT architecture principles; modularization, loose

Learning outcomes First session Overview of IT architecture principles; modularization, loose

Platforms and Architecture Bendik Bygstad IN5210 INFORMATION SYSTEMS SEPT 18, 2017 Learning outcomes First session Overview of IT architecture principles; modularization, loose coupling etc. Understand the technical and economic

731 views • 20 slides
for XCA, XCPD, XDR and XCDR Profiles Presented by Charles Parisot GE Healthcare, IHE ITI

for XCA, XCPD, XDR and XCDR Profiles Presented by Charles Parisot GE Healthcare, IHE ITI

New Asynchronous AS4 Web Services Option for XCA, XCPD, XDR and XCDR Profiles Presented by Charles Parisot GE Healthcare, IHE ITI Technical Committee Member, IHE Intl Board Member Interest of Asynchronous AS4 Web Services To support scaling of

486 views • 14 slides
CS 5150 Software Engineering 24. Presentations William Y. Arms Presentations The following

CS 5150 Software Engineering 24. Presentations William Y. Arms Presentations The following

Cornell University Computing and Information Science CS 5150 Software Engineering 24. Presentations William Y. Arms Presentations The following material has guidance for software presentations in general with hints about preparing for the

253 views • 13 slides
ModelDB : a system for managing ML models Manasi Vartak , PhD Candidate MIT Database Group

ModelDB : a system for managing ML models Manasi Vartak , PhD Candidate MIT Database Group

ModelDB : a system for managing ML models Manasi Vartak , PhD Candidate MIT Database Group mvartak@csail.mit.edu | @DataCereal Why Model Management? IMDB Prediction Task Given data about movies (e.g. year made, studio, genres, actors)

361 views • 24 slides
IoTivity Core Framework: Features & Opportunities Kishen Maloor Intel Open Source Technology

IoTivity Core Framework: Features & Opportunities Kishen Maloor Intel Open Source Technology

IoTivity Core Framework: Features & Opportunities Kishen Maloor Intel Open Source Technology Center iCDG - Intel Communication and Devices Group Outline What is IoTivity and why is it useful? IoTivity stack architecture IoTivity

305 views • 26 slides
CARD: A Congestion-Aware Request Dispatching Scheme for Replicated Metadata Server Cluster

CARD: A Congestion-Aware Request Dispatching Scheme for Replicated Metadata Server Cluster

CARD: A Congestion-Aware Request Dispatching Scheme for Replicated Metadata Server Cluster Shangming Cai, Dongsheng Wang, Zhanye Wang and Haixia Wang Tsinghua University 1 Background: Massive-scale ML in product environments Datasets

260 views • 25 slides
Metaphors Pages 43 - 44 Transform the World Metaphor Critical Faculty/ Barrier Problems

Metaphors Pages 43 - 44 Transform the World Metaphor Critical Faculty/ Barrier Problems

Metaphors Pages 43 - 44 Transform the World Metaphor Critical Faculty/ Barrier Problems Presenting Solution Problem Transform the World Metaphor Design Q2: What is another example of that? Q1: What is this an example of? Critical

213 views • 9 slides
Troubleshooting Service Dr. Susanne Naegele-Jackson DFN / University of Erlangen-Nuremberg

Troubleshooting Service Dr. Susanne Naegele-Jackson DFN / University of Erlangen-Nuremberg

Enabling Grids for E-sciencE perfSONAR-Lite TSS Troubleshooting Service Dr. Susanne Naegele-Jackson DFN / University of Erlangen-Nuremberg EGEE09, EGEE -III Activity SA2 www.eu-egee.org EGEE-III INFSO-RI-222667 EGEE and gLite are

421 views • 27 slides
Socket clients and servers CSCI 136: Fundamentals of Computer Science II Keith Vertanen

Socket clients and servers CSCI 136: Fundamentals of Computer Science II Keith Vertanen

Socket clients and servers CSCI 136: Fundamentals of Computer Science II Keith Vertanen Overview Networking basics IP Addresses (review) Port numbers (review) Reliability, connecting, latency, firewalls Single threaded

336 views • 15 slides

More recommend