Transcript HBase: An Introduction

Christopher Shain
Software Development Lead
Tresata






Hadoop for Financial Services
First completely Hadoop-powered analytics
application
Widely recognized as “Big Data Startup to
Watch”
Winner of the 2011 NCTA Award for Emerging
Tech Company of the Year
Based in Charlotte NC
We are hiring! [email protected]


Software Development Lead at Tresata
Background in Financial Services IT
 End-User Applications
 Data Warehousing
 ETL


Email: [email protected]
Twitter: @chrisshain

What is HBase?
 From: http://hbase.apache.org:
▪ “HBase is the Hadoop database.”
 I think this is a confusing statement

‘Database’, to many, means:
 Transactions
 Joins
 Indexes

HBase has none of these
 More on this later

HBase is a data storage platform designed to
work hand-in-hand with Hadoop







Distributed
Failure-tolerant
Semi-structured
Low latency
Strictly consistent
HDFS-Aware
“NoSQL”

Need for a low-latency, distributed datastore
with unlimited horizontal scale

Hadoop (MapReduce) doesn’t provide lowlatency

Traditional RDBMS don’t scale out
horizontally

November 2006: Google BigTable whitepaper
published:
http://research.google.com/archive/bigtable.html
February 2007: Initial HBase Prototype
October 2007: First ‘usable’ HBase
January 2008: HBase becomes Apache
subproject of Hadoop
 March 2009: HBase 0.20.0
 May 10th, 2010: HBase becomes Apache Top
Level Project




Web Indexing

Social Graph

Messaging (Email etc.)

HBase is written almost entirely in Java
 JVM clients are first-class citizens
HBase Master
RegionServer
JVM Clients
RegionServer
Non-JVM
Clients
Proxy
(Thrift or REST)
RegionServer
RegionServer






All data is stored in Tables
Table rows have exactly one Key, and all rows in a
table are physically ordered by key
Tables have a fixed number of Column Families
(more on this later!)
Each row can have many Columns in each column
family
Each column has a set of values, each with a
timestamp
Each row:family:column:timestamp combination
represents coordinates for a Cell




Defined by the Table
A Column Family is a group of related
columns with it’s own name
All columns must be in a column family
Each row can have a completely different set
of columns for a column family
Row:
Column Family:
Chris
Bob
James
Columns:
Friends:Bob
Friends
Friends:Chris
Friends:Bob
Friends:James

Not exactly the same as rows in a traditional
RDBMS
 Key: a byte array (usually a UTF-8 String)
 Data: Cells, qualified by column family, column, and
timestamp (not shown here)
Row Key: Column Families :
(Defined by the
Table)
Attributes
Chris
Friends
Columns:
Cells:
(Defined by the Row) (Created with Columns)
(May vary between
rows)
Attributes:Age
30
Attributes:Height
68
Friends:Bob
1 (Bob’s a cool guy)
Friends:Jane
0 (Jane and I don’t get along)





All cells are created with a timestamp
Column family defines how many versions of
a cell to keep
Updates always create a new cell
Deletes create a tombstone (more on that
later)
Queries can include an “as-of” timestamp to
return point-in-time values
HBase deletes are a form of write called a
“tombstone”
 Indicates that “beyond this point any previously
written value is dead”
 Old values can still be read using point-in-time
queries

Timestamp
Write Type
Resulting Value
Point-In-Time
Value “as of” T+1
T+0
PUT (“Foo”)
“Foo”
“Foo”
T+1
PUT (“Bar”)
“Bar”
“Bar”
T+2
DELETE
<none>
“Bar”
T+3
PUT (“Foo Too”)
“Foo Too”
“Bar”



Requirement: Store real-time stock tick data
Ticker
Timestamp
Sequence
Bid
Ask
IBM
09:15:03:001
1
179.16
179.18
MSFT
09:15:04:112
2
28.25
28.27
GOOG
09:15:04:114
3
624.94
624.99
IBM
09:15:04:155
4
179.18
179.19
Requirement: Accommodate many
simultaneous readers & writers
Requirement: Allow for reading of current
price for any ticker at any point in time
Historical Prices:
Keys
PK
Column
Ticker
Varchar
Timestamp
DateTime
Sequence_Number
Integer
Bid_Price
Decimal
Ask_Price
Decimal
Latest Prices:
Keys
PK
DataType
Column
DataType
Ticker
Varchar
Bid_Price
Decimal
Ask_Price
Decimal
Row Key
Family:Column
Prices:Bid
[Ticker].[Rev_Timestamp].[Rev_Sequence_Number]
Prices:Ask

HBase throughput will scale linearly with # of
nodes

No need to keep separate “latest price” table
 A scan starting at “ticker” will always
return the latest price row

HBase scales horizontally

Needs to split data over many RegionServers

Regions are the unit of scale





All HBase tables are broken into 1 or more
regions
Regions have a start row key and an end row
key
Each Region lives on exactly one
RegionServer
RegionServers may host many Regions
When RegionServers die, Master detects this
and assigns Regions to other RegionServers
“Users” Table
Row Keys in Region
“Aaron” – “George”
“Aaron”
-META- Table
Table
Users
Region
Region
Server
“Bob”
“Aaron” – “George”
Node01
“George” – “Matthew”
Node02
Row Keys in Region
“George” – “Matthew”
“Matthew” – “Zachary”
Node01
“George”
“Chris”
Row Keys in Region
“Matthew” – “Zachary”
“Matthew”
“Nancy”
“Zachary”
Deceptively simple
ZooKeeper
Cluster
HBase Master
JVM Clients
Backup HBase
Master
RegionServer
RegionServer
Non-JVM
Clients
Proxy
(Thrift or REST)
RegionServer
RegionServer

ZooKeeper
 Keeps track of which server is the current HBase
Master

HBase Master
 Keeps track of Region/RegionServer mapping
 Manages the -ROOT- and .META. tables
 Responsible for updating ZooKeeper when these
change

RegionServer
 Stores table regions

Clients
 Need to be smarter than RDBMS clients
 First connect to ZooKeeper to get RegionServer
for a given Table/Region
 Then connect directly to RegionServer to interact
with the data
 All connections over Hadoop RPC – non-JVM
clients use proxy (Thrift or REST (Stargate))
-ROOT- Table
info:regioninfo
.META.[region]
info:server
info:serverstartcode
Points to DataNode hosting
.META. region.…
.META. Table
info:regioninfo
[table],[region start key],[region id] info:server
info:serverstartcode
Regular User Table
… whatever …
…
Points to DataNode
hosting table region.

HBase Master is not necessarily a single point of
failure (SPOF)
 Multiple masters can be running
 Current ‘active’ Master controlled via ZooKeeper
 Make sure you have enough ZooKeeper nodes!

Master is not needed for client connectivity
 Clients connect directly to ZooKeeper to find Regions
 Everything Master does can be put off until one is
elected
ZooKeeper Quorum
ZooKeeper
Node
ZooKeeper
Node
HBase
Master
(Current)
ZooKeeper
Node
HBase
Master
(Standby)
HBase
Master
(Standby)

HBase tolerates RegionServer failure when
running on HDFS
 Data is replicated by HDFS (dfs.replication setting)
 Lots of issues around fsync, failure before data is
flushed - some probably still not fixed
 Thus, data can still be lost if node fails after a write

HDFS NameNode is still SPOF, even for HBase

Similar to log in many RDBMS

All operations by default written to log before considered
‘committed’ (can be overridden for ‘disposable fast writes’)

Log can be replayed when region is moved to another
RegionServer

One WAL per RegionServer
WAL
Flushed periodically
(10s by default)
HFile
Writes
MemStore
Flushed when
MemStore gets too big
HFile
RegionServer
Log
StoreFile
HFile
HDFS Client
Block
Block
Store
StoreFile
HFile
Block
Block
Store
StoreFile
HFile
Block
MemStore
MemStore
Store
MemStore
Region
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
HDFS DataNode
HDFS DataNode
HDFS DataNode
HDFS DataNode

A RegionServer is not guaranteed to be on
the same physical node as it’s data

Compaction causes RegionServer to write
preferentially to local node
 But this is a function of HDFS Client, not HBase


All data is in memory initially (memstore)
HBase is a write-only system
 Modifications and deletes are just writes with
later timestamps
 Function of HDFS being append-only


Eventually old writes need to be discarded
2 Types of Compactions:
 Minor
 Major

All HBase edits are initially stored in memory
(memstore)

Flushes occur when memstore reaches a
certain size
 By default 67,108,864 bytes
 Controlled by hbase.hregion.memstore.flush.size
configuration property

Each flush creates a new HFile

Triggered when a certain number of HFiles are created for
a given Region Store (+ some other conditions)
 By default 3 HFiles
 Controlled by hbase.hstore.compactionThreshold configuration
property

Compacts most recent HFiles into one
 By default, uses RegionServer-local HDFS node

Does not eliminate deletes
 Only touches most recent HFiles

NOTE: All column families are compacted at once (this
might change in the future)

Triggered every 24 hours (with random
offset) or manually
 Large HBase installations usually leave this for
manual operators

Re-writes all HFiles into one

Processes deletes
 Eliminates tombstones
 Erases earlier entries


HBase does not have transactions
However:
 Row-level modifications are atomic: All
modifications to a row will succeed or fail as a unit
 Gets are consistent for a given point in time
▪ But Scans may return 2 rows from different points in
time
 All data read has been ‘durably stored’
▪ Does NOT mean flushed to disk- can still be lost!

DO: Design your schema for linear range scans on your
most common queries.
 Scans are the most efficient way to query a lot of rows
quickly

DON’T: Use more than 2 or 3 column families.
 Some operations (flushing and compacting) operate
on the whole row

DO: Be aware of the relative cardinality of column
families
 Wildly differing cardinality leads to sparsity and bad
scanning results.

DO: Be mindful of the size of your row and column
keys
 They are used in indexes and queries, can be quite
large!

DON’T: Use monotonically increasing row keys
 Can lead to hotspots on writes

DO: Store timestamp keys in reverse
 Rows in a table need to be read in order, usually
you want most recent

DO: Query single rows using exact-match on key
(Gets) or Scans for multiple rows
 Scans allow efficient I/O vs. multiple gets

DON’T: Use regex-based or non-prefix column filters
 Very inefficient

DO: Tune the scan cache and batch size parameters
 Drastically improves performance when returning lots of
rows

Deceptively simple
HBase Master
JVM Clients
RegionServer
RegionServer
Non-JVM
Clients
Proxy
(Thrift or REST)
RegionServer
RegionServer
ZooKeeper Quorum
ZooKeeper
Node
ZooKeeper
Node
HBase
Master
(Current)
ZooKeeper
Node
HBase
Master
(Standby)
HBase
Master
(Standby)
RegionServer
HFile
Block
StoreFile
HFile
Block
Block
Block
Store
StoreFile
HFile
Block
MemStore
Log
StoreFile
HDFS Client
Store
MemStore
Store
MemStore
Region
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
Block
HDFS DataNode
HDFS DataNode
HDFS DataNode
HDFS DataNode




Requirement: Store an arbitrary set of
preferences for all users
Requirement: Each user may choose to store
a different set of preferences
Requirement: Preferences may be of
different data types (Strings, Integers, etc)
Requirement: Developers will add new
preference options all the time, so we
shouldn’t need to modify the database
structure when adding them

One possible RDBMS solution:
 Key/Value table
 All values as strings
 Flexible, but wastes space
Keys:
PK
Column:
Data Type:
UserID
Int
PreferenceName
Varchar
PreferenceValue
Varchar


Store all preferences in the Preferences
column family
Preference name as column name,
preference value as (serialized) byte array:
 HBase client library provides methods for
serializing many common data types
Row Key:
Family:
Chris
Preferences
Joe
Preferences
Column:
Value:
Age
30
Hometown
“Mineola, NY”
Birthdate
11/13/1987