Note: No Sql

Note: NoSql

1 Relational DB

1) value:
-to get consistent data
-syncnization
-integration

2)weakness
– impedence mismatch

3)ACID transaction:
-atomic, consistent, isolated, durable

2聚集導向(aggregate- orientated)
-No SQL does NOT support ACID transaction
– No Sql DB: Google BigTable
-THINK how to construct data

3 資料模型

1 relationship
2 graphic DB: very suitable to record any complex relational data
3 無綱要DB: informal storage for any kind of complex data
4 materialized views 具現檢視表
5 modeling for data: key/ value

4分散式模型

1 single server
2 partition
-horizontal , very helpful for efficiency
3 master-slave replication
4 peer 2 peer replication
5 combine with segmentation and replication

5 Consistency

1 update
2 read
3 loose

6 CAP theory
– consistancy
-availibility
-partition tolerance

7 Version tag

1 enterprise and system transaction
2 version tag on multi-nodes

8 Map reduce
-basic:

MAP
input =single aggregation
output= key-value pair

REDUCE
key- value pair REDUCE to single aggregation

-partition and combine

partion

9 Key-value DB

value= bucket (桶)
storage : BLOB
key- value DB:

Riak
Redis
Berkeley DB
HamsterDB
Amazone’s Dynamo
Project voldemort

key- value storage funcion
– consistancy
– transaction
– find out
– scalable

Suitable for online conversation storage, avatar info, shopping cart,

廣告
舊文章
發表留言

發表迴響

在下方填入你的資料或按右方圖示以社群網站登入:

WordPress.com Logo

您的留言將使用 WordPress.com 帳號。 登出 / 變更 )

Twitter picture

您的留言將使用 Twitter 帳號。 登出 / 變更 )

Facebook照片

您的留言將使用 Facebook 帳號。 登出 / 變更 )

Google+ photo

您的留言將使用 Google+ 帳號。 登出 / 變更 )

連結到 %s

%d 位部落客按了讚: