SW-PRODUCT/개발

2017. 11. 9. 13:30


완전 간단하게 정리했던 서버 시스템들.




서버시스템_2015.png


서버시스템_2015.mm




서버시스템

  • Front-end :
  • Queue
  • In-memory cache
    • EHCache => Caffeine
  • cache server
    • Redis
  • Circuit Breaker
    • Hystrix
  • disk / file system
  • rdbms
    • 닭치고 read에는 MySQL이 짱이라는소문도...
    • MHA+고사양장비로 고질적 문제들 상당부분 해결됨
    • DokuDB, Lossless Replication. HikariCP(Connection Pool)
  • graphdb / s2graph
  • disk -> DAS/NAS -> hdfs
  • No-SQL
  • Kafka
  • Impala / Hive / Pig / Mahout
  • HBase / Cassandra
  • DSMS / Spark
  • parallel programming
  • multi threading
  • AOP
  • nio -> NIO / AKKA
  • nginx/nodejs (?)
  • AWS
  • docker
    • Why run a container cluster?
      • http://googlecloudplatform.blogspot.kr/2015/01/what-makes-a-container-cluster.html
  • Mesos, Docker, Marathon
    • http://www.recorditblog.com/post/how-to-create-a-web-scale-infrastructure-based-on-docker-coreos-vulcand-and-mesos-and-why-object-storage-becomes-the-de-facto-data-repository/

  • Mesosphere
    • https://mesosphere.com/
    • Datacenter Operating System
  • consul
    • service discovery / failure detection
    • DNS, HTTP 지원 => 기존 DNS에 프로그래밍 특성 제공
    • programmable DNS 정도? 클러스터는 FQDN으로 관리
    • https://www.consul.io/
  • application level

    • 멀티 쓰래딩
    • 멀티 프로세싱
    • In memory cache
    • cache server
  • request level
    • thread pool/worker 방식
    • event queue 방식
    • Message Queue
    • Kafka
  • storage level
    • one disk
    • RAID 1+0
    • NAS/DAS/SAN
    • HDFS
      • no-modify!!
        • 새로운 revision으로 파일 생성!
  • Mgmt of Data
    • file
    • NoSQL
      • Key - Value store
      • Column-family
      • Document
    • RDBMS
    • graph db
      • http://www.quora.com/What-are-the-pros-and-cons-of-using-a-graph-database
  • Big Data
    • Hadoop / MapReduce
    • HBase, Pig, Hive, Mahout
    • Impala
    • Spark Streaming
      • http://www.slideshare.net/yongho/rdd-paper-review p15
  • Distributed
    • Zookeeper
    • Consul