The API pattern was created in the 1970's when 'distributed architectures' didn't even exist and was established mainly for 'centralized architectures' as it bound the communication data/logic to the business logic.
In a modern world, we have moved to distributed architectures where we now have to share the I/O... but that communication logic still remains bound in the application due to an old API pattern. This makes it so that the IO data and functionality related to a request/response cannot be shared with our edge services without duplication/entanglement. This in turn means the data/functionality in our services then cannot be synchronized.
This leads to dropped threads, poor security, bad data, bad user experience, broken application, etc.
This can ALL be fixed and improved and even lead to better speed, scalability and automation through a new API Pattern.
1. The document discusses microservices architecture and how Netflix transitioned from a monolithic architecture to microservices. Key aspects discussed include breaking the monolith into many small, independent services that are loosely coupled.
2. Netflix's microservices architecture is composed of hundreds of microservices running on thousands of servers. Each service focuses on doing a small, well-defined piece of work. Services communicate through well-defined APIs and share no code or databases.
3. The document provides examples of how other companies like Samsung and Vingle have also adopted microservices architectures on AWS, breaking monolithic applications into independent, scalable services. This allows for independent deployments, rapid innovation, and improved resilience.
Zero Data Loss Recovery Applianceによるデ`タベ`ス隠oのア`キテクチャオラクルエンジニア宥佚
?
デ`タ楚の寄、I佞24rg晒に育い、愔瓦離丱奪アップ?ソリュ`ションではデ`タ隠oのニ`ズをすべて困燭擦覆なってきています。これを盾すべくOracle Databaseの隠oに蒙晒してOされたエンジニアド?システム、Zero Data Loss Recovery Applianceが鞠しました。これからのr旗のデ`タ隠oテクノロジ`にvして、ア`キテクチャを嶄伉にB初します。
The API pattern was created in the 1970's when 'distributed architectures' didn't even exist and was established mainly for 'centralized architectures' as it bound the communication data/logic to the business logic.
In a modern world, we have moved to distributed architectures where we now have to share the I/O... but that communication logic still remains bound in the application due to an old API pattern. This makes it so that the IO data and functionality related to a request/response cannot be shared with our edge services without duplication/entanglement. This in turn means the data/functionality in our services then cannot be synchronized.
This leads to dropped threads, poor security, bad data, bad user experience, broken application, etc.
This can ALL be fixed and improved and even lead to better speed, scalability and automation through a new API Pattern.
1. The document discusses microservices architecture and how Netflix transitioned from a monolithic architecture to microservices. Key aspects discussed include breaking the monolith into many small, independent services that are loosely coupled.
2. Netflix's microservices architecture is composed of hundreds of microservices running on thousands of servers. Each service focuses on doing a small, well-defined piece of work. Services communicate through well-defined APIs and share no code or databases.
3. The document provides examples of how other companies like Samsung and Vingle have also adopted microservices architectures on AWS, breaking monolithic applications into independent, scalable services. This allows for independent deployments, rapid innovation, and improved resilience.
Zero Data Loss Recovery Applianceによるデ`タベ`ス隠oのア`キテクチャオラクルエンジニア宥佚
?
デ`タ楚の寄、I佞24rg晒に育い、愔瓦離丱奪アップ?ソリュ`ションではデ`タ隠oのニ`ズをすべて困燭擦覆なってきています。これを盾すべくOracle Databaseの隠oに蒙晒してOされたエンジニアド?システム、Zero Data Loss Recovery Applianceが鞠しました。これからのr旗のデ`タ隠oテクノロジ`にvして、ア`キテクチャを嶄伉にB初します。
designing, implementing and delivering microservices with event storming, spr...uEngine Solutions
?
Implementing Microservices is something like an adventure. Analyzing and decomposing microservices with applying DDD and make them into code, all is not easy. With new simple approach - Event storming, designing and implementing an event-driven MSA became easier ever seen before.
ATAM is an architecture evaluation method that assesses architectural decisions based on quality attribute requirements. It identifies risks, tradeoffs, and sensitivity points. The ATAM process has 3 phases - introduction, analysis and investigation, and follow up. It involves stakeholders generating scenarios, analyzing how architectural approaches address quality attributes, and identifying risks. The results include documented architectural approaches, utility trees, scenarios, and risks/tradeoffs to improve the architecture.
The document discusses GPRS (General Packet Radio Service) networks. It provides an overview of the evolution of mobile network generations from 1G to 4G. It then describes the key components and protocols of GPRS networks, including the GPRS architecture, interfaces, radio interface protocols, protocol stacks, and functions of network elements like the SGSN and GGSN.
The document discusses telecommunication standards from 1G to 4G including GSM, CDMA, WiMAX and LTE networks. It specifically focuses on the GSM network architecture, characteristics and interfaces. The GSM network uses a cellular structure with Base Transceiver Stations communicating with Mobile Stations through the air interface. It also describes the components of a Mobile Station including the Mobile Equipment and Subscriber Identity Module.
3. 3
Infrastructure Evolution
? IT ???????? ??
Electronic accounting
machine
1930 ~ 1950
? ??? ??? ?? ?? ??? ??? ??
? ?? ???? ?????
General-Purpose
Mainframe and
minicomputer
1950 ~ ??
? 1958: IBM ?????
? 1965: DEC ???
Personal computer 1981 ~ ??
? 1981: IBM PC
? 80~90?? PC ? SW ??
Client Server 1983 ~ ??
? ??? ?????? ????? ?? ?? ??
? N-tier
? ??? ??? ??: ????, ??????, Web
Enterprise Internet 1992 ~ ??
? ?? ????
? ??? ?? ? ?????? ??????
Cloud and mobile
computing
2000 ~ ??
? ??? ? ??????? ????? ?????
?? ??
? ?? ??? ???? ??? ??
4. 4
Software Development Methodology
? Waterfall ? Agile ? DevOps
Waterfall
Agile
DevOps
Analysis Development Build Test Deployment
DEV QA OPS
Test Deployment
DEV QA OPS
A D B A D B A D B
Sprint Sprint Sprint
A D B T D
Sprint
DEV/QA/OPS
A D B T D
Sprint
DEV/QA/OPS
A D B T D
Sprint
DEV/QA/OPS
5. 5
Prolog - History
Innovation
1985 1995 2005 2015 2025
OO application
(offline, monolith)
Online systems
(offline, monolith)
Distributed systems
(distributed monolith)
Microservices Technology X
Challenge:
Reaching customers
(diskettes, CDs)
Challenge:
Scaling
Performance
Challenge:
Zero downtime
Upgrade
Challenge:
Comprehension
Monolith
Single unit
SOQ
Coarse-grained
Microservices
Fine-grained
11. 11
Microservices - Definition
? Microservices is ´
A software architecture style in which complex applications are composed of small, Independent process
communicating with each other using language-agnostic APIs.
These services are small, highly decupled and focus on doing a small task, facilitating a modular approach
to system-build - Wikipedia
??? ??? ????? ?? ???? ??? ?? ??? ??????? ???? ????, ??? ????
???? ???? ????, ??? ??? ? ??/????? ?? ?? ????? ??? ? ??. ????
???? ?? ??? ???, ? ???? ?? ?? ????? ??, ?? ??? ???? ??? ??? ???
?. - Martin Fowler at ThoughWorks
Data Store
Public API
Software Modules: Application code, libraries, ´
Microservices = Fine-grained service-oriented architecture + ^small ̄ public API
Microservice
12. 12
Microservices C Pros, & Cons.
Data
Business Logic
Web
Monolithic Design Microservices Design
Http Listener Page Content
ReportingOrder ManagementUsage Analytics
AuthenticationOrderSearching
Monolithic Style Microservices
Single Unit with all components interwoven Applications are suite of service
Overall Scaling of the structure required Each service independently scalable
Overall deployment Independently deployable
Increases complexity for developers More developer friendly
One errant component crashes the whole system Fault diagnosis and isolation is easier
Testing is streamlined Difficult to test
14. 14
Microservices C Organization: Conway¨s law
Organizations which design system (in the board sense used here) are
constrained to produce design which are copies of the communications
structures of these organizations
- Melvin Conway, Datamation, 1968
15. 15
Microservices C Most Important thing!
? Only rely on each other¨s public API
? ?? ?? ??? ?????? ?? ???? ???? ??? ????
? ??? ??? ?????? ?? ?? ????
? ?? ??? ?????? ?? ?? ???? ??? ????? ????? ??
Microservice A Microservice B
30. 30
V-Model & Test Pyramid
? Each phase has corresponding test or validation counterpart
Requirements
Analysis
System
Design
Program
Design
Implementation
Unit Test
Integration
Test
Acceptance
Test
Nested V-Model
Nested multiple V-Model
31. 31
Microservice Deploy
? Use Continuous Integration/Delivery/Deployment
? Decreased lead times and higher quality with smaller batches
Monitoring