Please enable JavaScript.
Coggle requires JavaScript to display documents.
clustering(distance 계산), cloud computing, NoSQL 2, clustering 방법, NoSQL 1,…
clustering(distance 계산)
cosine
백터
consine distance = 각도(0~180)
jaccard
jaccard similarity
jaccard distance
집합
euclidian distance
점
edit distance
문자열
LCS
hamming distance
비트벡터
cloud computing
cluster
구성요소
rack
network switch
nodes/blades
storage device
data center
hot aisle/cold aisle
문제점
difficult to dimension
waste resources
lose customers
expensive
difficult to scale
power plant
쓴만큼 돈내자!
서비스 종류
Saas
Web
Paas
Iaas
Public cloud
Private cloud
hybrid cloud
virtualization
VM
VMM
migration
time sharing
isolation
NoSQL 2
data distribution
sharding
horizontal scalability
replication
master-slave
master
slave
peer-to-peer
inconsistent read
inconsistent write
consistency
conflicts
read-write
write-write
해결책
pessimistic approach
safety
optimistic approach
liveness
종류
strong consistency
RDBMS
ACID
eventual consistency
LDBMS
BASE
Basically Available
Soft state
Eventual consistent
inconsistency window
read-your-write consistency
session consistency
version vector
CAP theorem(brewer's theorem)
Consistency
Availability
Partition tolerance
clustering 방법
Hierarchical
agglomerative(bottom up)
divisive(top down)
개념
centroid
clustroid
nearness 판단
min distance
max distance
diameter
group average distance
centroid distance
density-based approach
Point assignment
k-means
BFR
discard set(DS)
compressed set(CS)
retained set(RS)
summarizing sets of points
N
SUM
SUMSQ
points set 정보
count
centroid
standard deviation
mahalanobis distance
CURE
단계
initialization
merge
NoSQL 1
RDBMS
ACID
Atomicity
Consistency
Isolation
Durability
concurrency control
NoSQL
storage model
data model
aggregate models
key-value
<key, value>
TTL based
column-family
<row key, <column key, value>>
document
<key, document>
graph-base models
nodes
relationships
properties
<key, value>
Apache Spark
구성요소
SparkContext
RDD
partition
operation
transformations
actions
persistence
Driver Program
Cluster Manager
deploy mode
client
cluster
worker node
executor
task
stack
Spark SQL
Spark Streaming
Spark MLlib & GraphX