SlideShare a Scribd company logo
Use Cases and New Features
@ApachePhoenix
http://phoenix.apache.org
V5
Agenda
• Phoenix Use Cases
– Argus: Time-series data with Phoenix (Tom Valine, Salesforce.com)
– Five major tips to maximize performance on a 200+ SQL HBase/Phoenix
cluster (Masayasu Suzuki, Sony)
– Phoenix & eHarmony, a perfect match (Vijay Vangapandu, eHarmony)
• What’s new in Phoenix
– ACID Transactions with Tephra (Poorna Chandra, Cask)
– Cost-based Query Optimization with Calcite (Maryann Xue, Intel)
• Q & A
–PhoenixCon tomorrow 9am-1pm @ Salesforce.com, 1 Market St, SF
Argus: Time-series data with Phoenix
Tom Valine
Salesforce.com
OpenTSDB Limitations
OpenTSDB is good, but we need more
•Tag Cardinality
– Total number of tags per metric is limited to 8
– Performance decreases drastically as tag values increase.
•UID Exhaustion
– Hard limit of 16M UIDs
•Ad hoc querying not possible
– Join to other data sources
– Joins of time series and events
– Simplification of Argus’ transform grammar
Phoenix-backed Argus TSDB Service
• 3 day hackathon
• Modeled metric as Phoenix VIEW
– Leverage ROW_TIMESTAMP optimization
• Tag values inlined in row key
– Uses SKIP_SCAN filter optimization
– Allows for secondary indexes on particular metric + tags
• Metric and tag names managed outside of data as metadata
• Eventually leverage Drillix (Phoenix + Drill)
– Cross cluster queries
– Joins to other data sources
Write Performance
Using 2 clients to write in parallel. Phoenix is using 10 writer threads per client
Read Performance
• Metrics with one tag (60 distinct values)
– OpenTSDB and Phoenix performance comparable for small aggregations
– Phoenix outperforms OpenTSDB as aggregation size increases
Disk usage
• Phoenix & OTSDB use approximately the same amount of space with FAST_DIFF
and Snappy compression
Five major tips to maximize
performance on a 200+ SQL
HBase/Phoenix cluster
Masayasu “Mas” Suzuki
Shinji Nagasaka
Takanari Tamesue
Sony Corporation
Who we are, and why we chose HBase/Phoenix
• We are DevOps members from
Sony’s News Suite team
http://socialife.sony.net/
• HBase/Phoenix was chosen
because of
a. Scalability,
b. SQL compatibility, and
c. secondary indexing support
Our use case
Performance test apparatus & results
• Test apparatus
• Test results
Specs
Number of records 1.2 billion records (1 KB each)
Number of indexes 8 orthogonal indexes
Servers
3 Zookeepers (Zookeeper 3.4.5, m3.xlarge x 3)
3 HMaster servers (hadoop 2.5.0, hbase 0.98.6, Phoenix 4.3.0, m3.xlarge x 3)
200 RegionServers
(hadoop 2.5.0, hbase 0.98.6, Phoenix 4.3.0, r3.xlarge x 199, c4.8xlarge x 1)
Clients 100 x c4.xlarge
Results
Number of queries 51,053 queries/sec
Response time (average) 46 ms
Five major tips to maximize performance
using HBase/Phoenix
Ordered by effectiveness (most effective on the very top)
– An extra RPC is issued when the client runs a SQL statement that uses a secondary index
– Using SQL hint clause can mitigate this
– From Ver. 4.7, changing “UPDATE_CACHE_FREQUENCY” may also work (we have yet to test this)
– A memory rich node should be selected for use in RegionServers so as to minimize disk access
– As an example, running major compaction and index creation simultaneously should be avoided
Details will be presented at the PhoenixCon tomorrow (May 25)
2. Use memories aggressively
1. Use SQL hint clause when using a secondary index
4. Scale-out instead of scale-up
3. Manually split Region files if possible but never over split them
5. Avoid running power intensive tasks simultaneously
Vijay Vangapandu
Principal Platform Engineer
eHarmony and Phoenix a perfect match
NEED FOR
● Handling 30+ Million events during Batch Run
● Serving low latency queries on 16+ Billion records
75th% - 800MS 95th% - 2Sec 99th% - 4Sec
eHarmony and Phoenix a perfect match
LAMBDA FOR THE SAVE
• Layered architecture provides fault tolerance
• Hbase as batch storage for write throughput with reasonable read latency
• Apache Phoenix as query layer to work with complex queries with confidence
• Redis as speed layer cache
eHarmony and Phoenix a perfect match
PERFORMANCE
Phoenix/HBase goes live
Get Matches API Response Times
Phoenix/HBase goes live
Save Match API Response Times
eHarmony and Phoenix a perfect match
• Highly Consistent and fault tolerant
• Need for store level filtering and sorting
• Apache Phoenix helped us build an abstract high performance
query layer on top of Hbase.
• Eased the development process.
• Reduced boiler plate code, which provides maintainability.
• Build complex queries with confidence.
• Secondary indexes.
• JDBC connection.
• Good community support
WHY HBASE AND PHOENIX
HBASE
APACHE PHOENIX
eHarmony and Phoenix a perfect match
JAVA ORM LIBRARY(PHO)
• Apache Phoenix helped us build PHO (Phoenix-HBase ORM)
• PHO provides ability to annotate your entity bean and provides interfaces
to build DSL like queries.
Disjunction disjunction = new Disjunction();
for (int statusFilter : statusFilters) {
disjunction.add(Restrictions.eq("status", statusFilter));
}
QueryBuilder.builderFor(FeedItemDto.class).select()
.add(Restrictions.eq("userId", userId))
.add(Restrictions.gte("spotlightEnd", spotlightEndDate))
.add(disjunction)
.setReturnFields(projection)
.addOrder(orderings)
.setMaxResults(maxResults)
.build();
eHarmony and Phoenix a perfect match
http://eharmony.github.io/
OPEN SOURCE REPOSITORY
https://github.com/eHarmony/pho
http://www.eharmony.com/about/careers/
*Please Join us for more details at PhoenixCon tomorrow (May 25)
ACID Transactions
+
Poorna Chandra
Cask
Why Transactions?
• All or none semantics simplifies life of
developer
– Ensures every client has a consistent view of data
– Protects against concurrent updates
– No need to reason about what state data is left in
if write fails
– Guaranteed consistency between data and index
Apache Tephra
• Transactions on HBase
– Across regions, tables and RPC calls
• ACID semantics
• Tephra Powers
– CDAP (Cask Data Application Platform)
– Apache Phoenix (4.7 onwards)
Apache Tephra Architecture
Zookeeper
Tx Manager
(standby)
HBase
Master 1 RS 1
RS 2 RS 4
RS 3
Client 1
Client 2
Client N
Tx Manager
(active)
Master 2
Tephra Components
• TransactionAware client
• Coordinates transaction lifecycle with manager
• Communicates directly with HBase for reads and writes
• Transaction Manager
• Assigns transaction IDs
• Maintains state on in-progress, committed and invalid transactions
• Transaction Processor coprocessor
• Applies server-side filtering for reads
• Cleans up data from failed transactions, and no longer visible versions
Snapshot Isolation
• Multi-version concurrency control
– Cell version (timestamp) = transaction ID
– Reads exclude other uncommitted transactions (for
isolation)
• Optimistic Concurrency Control
– Avoids cost of locking rows and tables
– Good if conflicts are rare: short transaction, disjoint
partitioning of work
Single client using 10 threads in parallel with 5K batch size
No performance penalty for non-transactional tables
Performance
Future Work
• Partitioned Transaction Manager
• Automatic pruning of invalid transaction list
• Read-only transactions
• Performance optimizations
• Conflict detection
• Appends to transaction edit log
+
Cost-based Query Optimization
Maryann Xue
Intel
Integration model
Calcite Parser & Validator
Calcite Query Optimizer
Phoenix Query Plan Generator
Phoenix Runtime
Phoenix Tables over HBase
JDBC Client
SQL + Phoenix
specific
grammar Built-in rules
+ Phoenix
specific rules
Cost-based query optimizer
with Apache Calcite
• Base all query optimization decisions on cost
– Filter push down; range scan vs. skip scan
– Hash aggregate vs. stream aggregate vs. partial stream aggregate
– Sort optimized out; sort/limit push through; fwd/rev/unordered scan
– Hash join vs. merge join; join ordering
– Use of data table vs. index table
– All above (any many others) COMBINED
• Query optimizations are modeled as pluggable rules
Beyond Phoenix 4.8
with Apache Calcite
• Get the missing SQL support
– WITH, UNNEST, Scalar subquery, etc.
• Materialized views
– To allow other forms of indices (maybe defined as external), e.g., a filter
view, a join view, or an aggregate view.
• Interop with other Calcite adaptors
– Already used by Drill, Hive, Kylin, Samza, etc.
– Supports any JDBC source
– Initial version of Drill-Phoenix integration already working
Query Example - no cost-based optimizer
select empid, e.name,
d.deptno, d.name,
location
from emps e, depts d
using deptno
order by e.deptno
Phoenix
Compiler
scan ‘depts’
send ‘depts’ over to RS
& build hash-cache
scan ‘emps’ hash-join ‘depts’
sort joined table on ‘e.deptno’
Query Example - with cost-based optimizer
(sort optimization combined with join algorithm decision)
LogicalSort
key: deptno
LogicalJoin
inner,
e.deptno = d.deptno
LogicalProject
empid, e.name, d.deptno,
d.name, location
LogicalTableScan
emps LogicalTableScan
depts
PhoenixTableScan
depts
PhoenixMergeJoin
inner,
e.deptno = d.deptno
PhoenixClientProject
empid, e.name, d.deptno,
d.name, location
Optimizer
Optimization rules
+
Phoenix operator
conversion rules
PhoenixTableScan
emps
PhoenixServerProjec
t
empid, name, deptno
PhoenixServerProject
deptno, name, location
select empid, e.name, d.deptno,
d.name, location
from emps e, depts d using deptno
order by e.deptno
PhoenixServerSort
key: deptno
empid
empid
deptno
deptno
deptno
e.deptno;
d.deptno;
e.deptno;
d.deptno;
Query Example - with cost-based optimizer
(sort optimization combined with join algorithm decision)
Phoenix
Implementor
PhoenixTableScan
depts
PhoenixMergeJoin
inner,
e.deptno = d.deptno
PhoenixClientProject
empid, e.name, d.deptno,
d.name, location
PhoenixTableScan
emps
PhoenixServerProjec
t
empid, name, deptno
PhoenixServerProject
deptno, name, location
PhoenixServerSort
key: deptno
empid
empid
deptno
deptno
deptno
e.deptno;
d.deptno;
e.deptno;
d.deptno;
scan ‘emps’
merge-join ‘emps’ and ‘depts’
sort by ‘deptno’
scan ‘depts’
Query Example - Comparison
Query plan w/o cost-based
optimizer
Query plan w/ cost-based optimizer
scan ‘emps’, ‘depts’ first ‘depts’, then ‘emps’ 2 tables in parallel
hash-cache send & build proportional to size of ‘depts’;
might cause exception if too large
none
hash-cache look-up 1 look-up per ‘emps’ row none
sorting sort ‘emps’ join ‘depts’ sort ‘emps’ only
optimization approach Local, serial optimization processes Cost-based, rule-driven, integrated
performance
(single node, 2M * 2K rows)
19.46 s 13.92 s
Drillix: Interoperability with Drill
select deptno, sum(salary) from emps group by deptno
Drill Final Aggregation
deptno, sum(salary)
Phoenix Table Scan
emps
Phoenix Tables over HBase
Drill Shuffle
Phoenix Partial Aggregation
deptno, sum(salary)
Stage 1:
Local Partial aggregation
Stage 3:
Final aggregation
Stage 2:
Shuffle partial results
Thank you! Questions?
Join us tomorrow for PhoenixCon
Salesforce.com, 1 Market St, SF 9am-1pm
(some companies using Phoenix)

More Related Content

What's hot

Apache phoenix: Past, Present and Future of SQL over HBAse
Apache phoenix: Past, Present and Future of SQL over HBAseApache phoenix: Past, Present and Future of SQL over HBAse
Apache phoenix: Past, Present and Future of SQL over HBAse
enissoz
 
Apache Phoenix and Apache HBase: An Enterprise Grade Data Warehouse
Apache Phoenix and Apache HBase: An Enterprise Grade Data WarehouseApache Phoenix and Apache HBase: An Enterprise Grade Data Warehouse
Apache Phoenix and Apache HBase: An Enterprise Grade Data Warehouse
Josh Elser
 
Apache Phoenix: Transforming HBase into a SQL Database
Apache Phoenix: Transforming HBase into a SQL DatabaseApache Phoenix: Transforming HBase into a SQL Database
Apache Phoenix: Transforming HBase into a SQL Database
DataWorks Summit
 
Apache Phoenix Query Server
Apache Phoenix Query ServerApache Phoenix Query Server
Apache Phoenix Query Server
Josh Elser
 
De-Mystifying the Apache Phoenix QueryServer
De-Mystifying the Apache Phoenix QueryServerDe-Mystifying the Apache Phoenix QueryServer
De-Mystifying the Apache Phoenix QueryServer
Josh Elser
 
HBaseCon 2013: Apache HBase Replication
HBaseCon 2013: Apache HBase ReplicationHBaseCon 2013: Apache HBase Replication
HBaseCon 2013: Apache HBase Replication
Cloudera, Inc.
 
HBase state of the union
HBase   state of the unionHBase   state of the union
HBase state of the union
enissoz
 
HBaseCon 2012 | HBase Filtering - Lars George, Cloudera
HBaseCon 2012 | HBase Filtering - Lars George, ClouderaHBaseCon 2012 | HBase Filtering - Lars George, Cloudera
HBaseCon 2012 | HBase Filtering - Lars George, Cloudera
Cloudera, Inc.
 
April 2014 HUG : Apache Phoenix
April 2014 HUG : Apache PhoenixApril 2014 HUG : Apache Phoenix
April 2014 HUG : Apache Phoenix
Yahoo Developer Network
 
Apache HBase Internals you hoped you Never Needed to Understand
Apache HBase Internals you hoped you Never Needed to UnderstandApache HBase Internals you hoped you Never Needed to Understand
Apache HBase Internals you hoped you Never Needed to Understand
Josh Elser
 
Meet HBase 2.0 and Phoenix 5.0
Meet HBase 2.0 and Phoenix 5.0Meet HBase 2.0 and Phoenix 5.0
Meet HBase 2.0 and Phoenix 5.0
DataWorks Summit
 
Apache Phoenix with Actor Model (Akka.io) for real-time Big Data Programming...
Apache Phoenix with Actor Model (Akka.io)  for real-time Big Data Programming...Apache Phoenix with Actor Model (Akka.io)  for real-time Big Data Programming...
Apache Phoenix with Actor Model (Akka.io) for real-time Big Data Programming...
Trieu Nguyen
 
Apache Hive 2.0: SQL, Speed, Scale
Apache Hive 2.0: SQL, Speed, ScaleApache Hive 2.0: SQL, Speed, Scale
Apache Hive 2.0: SQL, Speed, Scale
DataWorks Summit/Hadoop Summit
 
HBaseConEast2016: HBase and Spark, State of the Art
HBaseConEast2016: HBase and Spark, State of the ArtHBaseConEast2016: HBase and Spark, State of the Art
HBaseConEast2016: HBase and Spark, State of the Art
Michael Stack
 
Apache Phoenix + Apache HBase
Apache Phoenix + Apache HBaseApache Phoenix + Apache HBase
Apache Phoenix + Apache HBase
DataWorks Summit/Hadoop Summit
 
HBase and Impala Notes - Munich HUG - 20131017
HBase and Impala Notes - Munich HUG - 20131017HBase and Impala Notes - Munich HUG - 20131017
HBase and Impala Notes - Munich HUG - 20131017
larsgeorge
 
HBase Read High Availability Using Timeline Consistent Region Replicas
HBase  Read High Availability Using Timeline Consistent Region ReplicasHBase  Read High Availability Using Timeline Consistent Region Replicas
HBase Read High Availability Using Timeline Consistent Region Replicas
enissoz
 
SQL AlwaysON for SharePoint HA/DR on Azure Global Azure Bootcamp 2017 Eisenac...
SQL AlwaysON for SharePoint HA/DR on Azure Global Azure Bootcamp 2017 Eisenac...SQL AlwaysON for SharePoint HA/DR on Azure Global Azure Bootcamp 2017 Eisenac...
SQL AlwaysON for SharePoint HA/DR on Azure Global Azure Bootcamp 2017 Eisenac...
Lars Platzdasch
 
Apache phoenix
Apache phoenixApache phoenix
Apache phoenix
Osama Hussein
 
Cloudera Impala: A Modern SQL Engine for Hadoop
Cloudera Impala: A Modern SQL Engine for HadoopCloudera Impala: A Modern SQL Engine for Hadoop
Cloudera Impala: A Modern SQL Engine for Hadoop
Cloudera, Inc.
 

What's hot (20)

Apache phoenix: Past, Present and Future of SQL over HBAse
Apache phoenix: Past, Present and Future of SQL over HBAseApache phoenix: Past, Present and Future of SQL over HBAse
Apache phoenix: Past, Present and Future of SQL over HBAse
 
Apache Phoenix and Apache HBase: An Enterprise Grade Data Warehouse
Apache Phoenix and Apache HBase: An Enterprise Grade Data WarehouseApache Phoenix and Apache HBase: An Enterprise Grade Data Warehouse
Apache Phoenix and Apache HBase: An Enterprise Grade Data Warehouse
 
Apache Phoenix: Transforming HBase into a SQL Database
Apache Phoenix: Transforming HBase into a SQL DatabaseApache Phoenix: Transforming HBase into a SQL Database
Apache Phoenix: Transforming HBase into a SQL Database
 
Apache Phoenix Query Server
Apache Phoenix Query ServerApache Phoenix Query Server
Apache Phoenix Query Server
 
De-Mystifying the Apache Phoenix QueryServer
De-Mystifying the Apache Phoenix QueryServerDe-Mystifying the Apache Phoenix QueryServer
De-Mystifying the Apache Phoenix QueryServer
 
HBaseCon 2013: Apache HBase Replication
HBaseCon 2013: Apache HBase ReplicationHBaseCon 2013: Apache HBase Replication
HBaseCon 2013: Apache HBase Replication
 
HBase state of the union
HBase   state of the unionHBase   state of the union
HBase state of the union
 
HBaseCon 2012 | HBase Filtering - Lars George, Cloudera
HBaseCon 2012 | HBase Filtering - Lars George, ClouderaHBaseCon 2012 | HBase Filtering - Lars George, Cloudera
HBaseCon 2012 | HBase Filtering - Lars George, Cloudera
 
April 2014 HUG : Apache Phoenix
April 2014 HUG : Apache PhoenixApril 2014 HUG : Apache Phoenix
April 2014 HUG : Apache Phoenix
 
Apache HBase Internals you hoped you Never Needed to Understand
Apache HBase Internals you hoped you Never Needed to UnderstandApache HBase Internals you hoped you Never Needed to Understand
Apache HBase Internals you hoped you Never Needed to Understand
 
Meet HBase 2.0 and Phoenix 5.0
Meet HBase 2.0 and Phoenix 5.0Meet HBase 2.0 and Phoenix 5.0
Meet HBase 2.0 and Phoenix 5.0
 
Apache Phoenix with Actor Model (Akka.io) for real-time Big Data Programming...
Apache Phoenix with Actor Model (Akka.io)  for real-time Big Data Programming...Apache Phoenix with Actor Model (Akka.io)  for real-time Big Data Programming...
Apache Phoenix with Actor Model (Akka.io) for real-time Big Data Programming...
 
Apache Hive 2.0: SQL, Speed, Scale
Apache Hive 2.0: SQL, Speed, ScaleApache Hive 2.0: SQL, Speed, Scale
Apache Hive 2.0: SQL, Speed, Scale
 
HBaseConEast2016: HBase and Spark, State of the Art
HBaseConEast2016: HBase and Spark, State of the ArtHBaseConEast2016: HBase and Spark, State of the Art
HBaseConEast2016: HBase and Spark, State of the Art
 
Apache Phoenix + Apache HBase
Apache Phoenix + Apache HBaseApache Phoenix + Apache HBase
Apache Phoenix + Apache HBase
 
HBase and Impala Notes - Munich HUG - 20131017
HBase and Impala Notes - Munich HUG - 20131017HBase and Impala Notes - Munich HUG - 20131017
HBase and Impala Notes - Munich HUG - 20131017
 
HBase Read High Availability Using Timeline Consistent Region Replicas
HBase  Read High Availability Using Timeline Consistent Region ReplicasHBase  Read High Availability Using Timeline Consistent Region Replicas
HBase Read High Availability Using Timeline Consistent Region Replicas
 
SQL AlwaysON for SharePoint HA/DR on Azure Global Azure Bootcamp 2017 Eisenac...
SQL AlwaysON for SharePoint HA/DR on Azure Global Azure Bootcamp 2017 Eisenac...SQL AlwaysON for SharePoint HA/DR on Azure Global Azure Bootcamp 2017 Eisenac...
SQL AlwaysON for SharePoint HA/DR on Azure Global Azure Bootcamp 2017 Eisenac...
 
Apache phoenix
Apache phoenixApache phoenix
Apache phoenix
 
Cloudera Impala: A Modern SQL Engine for Hadoop
Cloudera Impala: A Modern SQL Engine for HadoopCloudera Impala: A Modern SQL Engine for Hadoop
Cloudera Impala: A Modern SQL Engine for Hadoop
 

Viewers also liked

Design Fair Poster - Sustainable Control Systems
Design Fair Poster - Sustainable Control SystemsDesign Fair Poster - Sustainable Control Systems
Design Fair Poster - Sustainable Control Systems
Jordan Robinson
 
Sehenswürdigkeiten in Triest - Deutsch / Italienisch
Sehenswürdigkeiten in Triest - Deutsch / ItalienischSehenswürdigkeiten in Triest - Deutsch / Italienisch
Sehenswürdigkeiten in Triest - Deutsch / Italienisch
Sabinavekk
 
ein besonderer Meilenstein in Triest
ein besonderer Meilenstein in Triestein besonderer Meilenstein in Triest
ein besonderer Meilenstein in Triest
SabinaVG
 
Barbara Donaldson_CV
Barbara Donaldson_CVBarbara Donaldson_CV
Barbara Donaldson_CV
Barbara Donaldson
 
CA Project and Portfolio Management v14.x: MS Project Integration: Tips, Tric...
CA Project and Portfolio Management v14.x: MS Project Integration: Tips, Tric...CA Project and Portfolio Management v14.x: MS Project Integration: Tips, Tric...
CA Project and Portfolio Management v14.x: MS Project Integration: Tips, Tric...
CA Technologies
 
Lintech ball screw_rs062series_specsheet
Lintech ball screw_rs062series_specsheetLintech ball screw_rs062series_specsheet
Lintech ball screw_rs062series_specsheet
Electromate
 
Advanced motion controls mc1xaz02
Advanced motion controls mc1xaz02Advanced motion controls mc1xaz02
Advanced motion controls mc1xaz02
Electromate
 
Resume
ResumeResume
Oil and Politics: The Venezuelan Experience- Francisco Monaldi
Oil and Politics: The Venezuelan Experience- Francisco MonaldiOil and Politics: The Venezuelan Experience- Francisco Monaldi
Oil and Politics: The Venezuelan Experience- Francisco Monaldi
Economic Research Forum
 
Food security
Food securityFood security
Food security
Juliet Abisha
 
Pathology of Urinary Tract Infectionws
Pathology of Urinary Tract InfectionwsPathology of Urinary Tract Infectionws
Pathology of Urinary Tract Infectionws
Shashidhar Venkatesh Murthy
 
Apache Hadoop and HBase
Apache Hadoop and HBaseApache Hadoop and HBase
Apache Hadoop and HBase
Cloudera, Inc.
 
Apache HBase in the Enterprise Data Hub at Cerner
Apache HBase in the Enterprise Data Hub at CernerApache HBase in the Enterprise Data Hub at Cerner
Apache HBase in the Enterprise Data Hub at Cerner
HBaseCon
 
Ibm innovate-2014 #2478-hammel_pub
Ibm innovate-2014 #2478-hammel_pubIbm innovate-2014 #2478-hammel_pub
Ibm innovate-2014 #2478-hammel_pub
Christof Hammel
 
Veränderungsmanagement
VeränderungsmanagementVeränderungsmanagement
Veränderungsmanagement
Christof Hammel
 

Viewers also liked (15)

Design Fair Poster - Sustainable Control Systems
Design Fair Poster - Sustainable Control SystemsDesign Fair Poster - Sustainable Control Systems
Design Fair Poster - Sustainable Control Systems
 
Sehenswürdigkeiten in Triest - Deutsch / Italienisch
Sehenswürdigkeiten in Triest - Deutsch / ItalienischSehenswürdigkeiten in Triest - Deutsch / Italienisch
Sehenswürdigkeiten in Triest - Deutsch / Italienisch
 
ein besonderer Meilenstein in Triest
ein besonderer Meilenstein in Triestein besonderer Meilenstein in Triest
ein besonderer Meilenstein in Triest
 
Barbara Donaldson_CV
Barbara Donaldson_CVBarbara Donaldson_CV
Barbara Donaldson_CV
 
CA Project and Portfolio Management v14.x: MS Project Integration: Tips, Tric...
CA Project and Portfolio Management v14.x: MS Project Integration: Tips, Tric...CA Project and Portfolio Management v14.x: MS Project Integration: Tips, Tric...
CA Project and Portfolio Management v14.x: MS Project Integration: Tips, Tric...
 
Lintech ball screw_rs062series_specsheet
Lintech ball screw_rs062series_specsheetLintech ball screw_rs062series_specsheet
Lintech ball screw_rs062series_specsheet
 
Advanced motion controls mc1xaz02
Advanced motion controls mc1xaz02Advanced motion controls mc1xaz02
Advanced motion controls mc1xaz02
 
Resume
ResumeResume
Resume
 
Oil and Politics: The Venezuelan Experience- Francisco Monaldi
Oil and Politics: The Venezuelan Experience- Francisco MonaldiOil and Politics: The Venezuelan Experience- Francisco Monaldi
Oil and Politics: The Venezuelan Experience- Francisco Monaldi
 
Food security
Food securityFood security
Food security
 
Pathology of Urinary Tract Infectionws
Pathology of Urinary Tract InfectionwsPathology of Urinary Tract Infectionws
Pathology of Urinary Tract Infectionws
 
Apache Hadoop and HBase
Apache Hadoop and HBaseApache Hadoop and HBase
Apache Hadoop and HBase
 
Apache HBase in the Enterprise Data Hub at Cerner
Apache HBase in the Enterprise Data Hub at CernerApache HBase in the Enterprise Data Hub at Cerner
Apache HBase in the Enterprise Data Hub at Cerner
 
Ibm innovate-2014 #2478-hammel_pub
Ibm innovate-2014 #2478-hammel_pubIbm innovate-2014 #2478-hammel_pub
Ibm innovate-2014 #2478-hammel_pub
 
Veränderungsmanagement
VeränderungsmanagementVeränderungsmanagement
Veränderungsmanagement
 

Similar to HBaseCon2016-final

HBaseCon 2015: Apache Phoenix - The Evolution of a Relational Database Layer ...
HBaseCon 2015: Apache Phoenix - The Evolution of a Relational Database Layer ...HBaseCon 2015: Apache Phoenix - The Evolution of a Relational Database Layer ...
HBaseCon 2015: Apache Phoenix - The Evolution of a Relational Database Layer ...
HBaseCon
 
hbaseconasia2019 Phoenix Improvements and Practices on Cloud HBase at Alibaba
hbaseconasia2019 Phoenix Improvements and Practices on Cloud HBase at Alibabahbaseconasia2019 Phoenix Improvements and Practices on Cloud HBase at Alibaba
hbaseconasia2019 Phoenix Improvements and Practices on Cloud HBase at Alibaba
Michael Stack
 
Introduction to Impala
Introduction to ImpalaIntroduction to Impala
Introduction to Impala
markgrover
 
messaging.pptx
messaging.pptxmessaging.pptx
messaging.pptx
NParakh1
 
Introducing Amazon EMR Release 5.0 - August 2016 Monthly Webinar Series
Introducing Amazon EMR Release 5.0 - August 2016 Monthly Webinar SeriesIntroducing Amazon EMR Release 5.0 - August 2016 Monthly Webinar Series
Introducing Amazon EMR Release 5.0 - August 2016 Monthly Webinar Series
Amazon Web Services
 
Stream processing on mobile networks
Stream processing on mobile networksStream processing on mobile networks
Stream processing on mobile networks
pbelko82
 
Large-scale Web Apps @ Pinterest
Large-scale Web Apps @ PinterestLarge-scale Web Apps @ Pinterest
Large-scale Web Apps @ Pinterest
HBaseCon
 
SQL and Machine Learning on Hadoop
SQL and Machine Learning on HadoopSQL and Machine Learning on Hadoop
SQL and Machine Learning on Hadoop
Mukund Babbar
 
HBaseConAsia2018 Track2-4: HTAP DB-System: AsparaDB HBase, Phoenix, and Spark
HBaseConAsia2018 Track2-4: HTAP DB-System: AsparaDB HBase, Phoenix, and SparkHBaseConAsia2018 Track2-4: HTAP DB-System: AsparaDB HBase, Phoenix, and Spark
HBaseConAsia2018 Track2-4: HTAP DB-System: AsparaDB HBase, Phoenix, and Spark
Michael Stack
 
AI&BigData Lab 2016. Сарапин Виктор: Размер имеет значение: анализ по требова...
AI&BigData Lab 2016. Сарапин Виктор: Размер имеет значение: анализ по требова...AI&BigData Lab 2016. Сарапин Виктор: Размер имеет значение: анализ по требова...
AI&BigData Lab 2016. Сарапин Виктор: Размер имеет значение: анализ по требова...
GeeksLab Odessa
 
SQL On Hadoop
SQL On HadoopSQL On Hadoop
SQL On Hadoop
Muhammad Ali
 
Architectural Evolution Starting from Hadoop
Architectural Evolution Starting from HadoopArchitectural Evolution Starting from Hadoop
Architectural Evolution Starting from Hadoop
SpagoWorld
 
Etu Solution Day 2014 Track-D: 掌握Impala和Spark
Etu Solution Day 2014 Track-D: 掌握Impala和SparkEtu Solution Day 2014 Track-D: 掌握Impala和Spark
Etu Solution Day 2014 Track-D: 掌握Impala和Spark
James Chen
 
Performance Optimizations in Apache Impala
Performance Optimizations in Apache ImpalaPerformance Optimizations in Apache Impala
Performance Optimizations in Apache Impala
Cloudera, Inc.
 
Big Data Streams Architectures. Why? What? How?
Big Data Streams Architectures. Why? What? How?Big Data Streams Architectures. Why? What? How?
Big Data Streams Architectures. Why? What? How?
Anton Nazaruk
 
Kinesis and Spark Streaming - Advanced AWS Meetup - August 2014
Kinesis and Spark Streaming - Advanced AWS Meetup - August 2014Kinesis and Spark Streaming - Advanced AWS Meetup - August 2014
Kinesis and Spark Streaming - Advanced AWS Meetup - August 2014
Chris Fregly
 
Meet HBase 2.0 and Phoenix-5.0
Meet HBase 2.0 and Phoenix-5.0Meet HBase 2.0 and Phoenix-5.0
Meet HBase 2.0 and Phoenix-5.0
DataWorks Summit
 
Streaming Solutions for Real time problems
Streaming Solutions for Real time problemsStreaming Solutions for Real time problems
Streaming Solutions for Real time problems
Abhishek Gupta
 
Apache conbigdata2015 christiantzolov-federated sql on hadoop and beyond- lev...
Apache conbigdata2015 christiantzolov-federated sql on hadoop and beyond- lev...Apache conbigdata2015 christiantzolov-federated sql on hadoop and beyond- lev...
Apache conbigdata2015 christiantzolov-federated sql on hadoop and beyond- lev...
Christian Tzolov
 
Real time fraud detection at 1+M scale on hadoop stack
Real time fraud detection at 1+M scale on hadoop stackReal time fraud detection at 1+M scale on hadoop stack
Real time fraud detection at 1+M scale on hadoop stack
DataWorks Summit/Hadoop Summit
 

Similar to HBaseCon2016-final (20)

HBaseCon 2015: Apache Phoenix - The Evolution of a Relational Database Layer ...
HBaseCon 2015: Apache Phoenix - The Evolution of a Relational Database Layer ...HBaseCon 2015: Apache Phoenix - The Evolution of a Relational Database Layer ...
HBaseCon 2015: Apache Phoenix - The Evolution of a Relational Database Layer ...
 
hbaseconasia2019 Phoenix Improvements and Practices on Cloud HBase at Alibaba
hbaseconasia2019 Phoenix Improvements and Practices on Cloud HBase at Alibabahbaseconasia2019 Phoenix Improvements and Practices on Cloud HBase at Alibaba
hbaseconasia2019 Phoenix Improvements and Practices on Cloud HBase at Alibaba
 
Introduction to Impala
Introduction to ImpalaIntroduction to Impala
Introduction to Impala
 
messaging.pptx
messaging.pptxmessaging.pptx
messaging.pptx
 
Introducing Amazon EMR Release 5.0 - August 2016 Monthly Webinar Series
Introducing Amazon EMR Release 5.0 - August 2016 Monthly Webinar SeriesIntroducing Amazon EMR Release 5.0 - August 2016 Monthly Webinar Series
Introducing Amazon EMR Release 5.0 - August 2016 Monthly Webinar Series
 
Stream processing on mobile networks
Stream processing on mobile networksStream processing on mobile networks
Stream processing on mobile networks
 
Large-scale Web Apps @ Pinterest
Large-scale Web Apps @ PinterestLarge-scale Web Apps @ Pinterest
Large-scale Web Apps @ Pinterest
 
SQL and Machine Learning on Hadoop
SQL and Machine Learning on HadoopSQL and Machine Learning on Hadoop
SQL and Machine Learning on Hadoop
 
HBaseConAsia2018 Track2-4: HTAP DB-System: AsparaDB HBase, Phoenix, and Spark
HBaseConAsia2018 Track2-4: HTAP DB-System: AsparaDB HBase, Phoenix, and SparkHBaseConAsia2018 Track2-4: HTAP DB-System: AsparaDB HBase, Phoenix, and Spark
HBaseConAsia2018 Track2-4: HTAP DB-System: AsparaDB HBase, Phoenix, and Spark
 
AI&BigData Lab 2016. Сарапин Виктор: Размер имеет значение: анализ по требова...
AI&BigData Lab 2016. Сарапин Виктор: Размер имеет значение: анализ по требова...AI&BigData Lab 2016. Сарапин Виктор: Размер имеет значение: анализ по требова...
AI&BigData Lab 2016. Сарапин Виктор: Размер имеет значение: анализ по требова...
 
SQL On Hadoop
SQL On HadoopSQL On Hadoop
SQL On Hadoop
 
Architectural Evolution Starting from Hadoop
Architectural Evolution Starting from HadoopArchitectural Evolution Starting from Hadoop
Architectural Evolution Starting from Hadoop
 
Etu Solution Day 2014 Track-D: 掌握Impala和Spark
Etu Solution Day 2014 Track-D: 掌握Impala和SparkEtu Solution Day 2014 Track-D: 掌握Impala和Spark
Etu Solution Day 2014 Track-D: 掌握Impala和Spark
 
Performance Optimizations in Apache Impala
Performance Optimizations in Apache ImpalaPerformance Optimizations in Apache Impala
Performance Optimizations in Apache Impala
 
Big Data Streams Architectures. Why? What? How?
Big Data Streams Architectures. Why? What? How?Big Data Streams Architectures. Why? What? How?
Big Data Streams Architectures. Why? What? How?
 
Kinesis and Spark Streaming - Advanced AWS Meetup - August 2014
Kinesis and Spark Streaming - Advanced AWS Meetup - August 2014Kinesis and Spark Streaming - Advanced AWS Meetup - August 2014
Kinesis and Spark Streaming - Advanced AWS Meetup - August 2014
 
Meet HBase 2.0 and Phoenix-5.0
Meet HBase 2.0 and Phoenix-5.0Meet HBase 2.0 and Phoenix-5.0
Meet HBase 2.0 and Phoenix-5.0
 
Streaming Solutions for Real time problems
Streaming Solutions for Real time problemsStreaming Solutions for Real time problems
Streaming Solutions for Real time problems
 
Apache conbigdata2015 christiantzolov-federated sql on hadoop and beyond- lev...
Apache conbigdata2015 christiantzolov-federated sql on hadoop and beyond- lev...Apache conbigdata2015 christiantzolov-federated sql on hadoop and beyond- lev...
Apache conbigdata2015 christiantzolov-federated sql on hadoop and beyond- lev...
 
Real time fraud detection at 1+M scale on hadoop stack
Real time fraud detection at 1+M scale on hadoop stackReal time fraud detection at 1+M scale on hadoop stack
Real time fraud detection at 1+M scale on hadoop stack
 

HBaseCon2016-final

  • 1. Use Cases and New Features @ApachePhoenix http://phoenix.apache.org V5
  • 2. Agenda • Phoenix Use Cases – Argus: Time-series data with Phoenix (Tom Valine, Salesforce.com) – Five major tips to maximize performance on a 200+ SQL HBase/Phoenix cluster (Masayasu Suzuki, Sony) – Phoenix & eHarmony, a perfect match (Vijay Vangapandu, eHarmony) • What’s new in Phoenix – ACID Transactions with Tephra (Poorna Chandra, Cask) – Cost-based Query Optimization with Calcite (Maryann Xue, Intel) • Q & A –PhoenixCon tomorrow 9am-1pm @ Salesforce.com, 1 Market St, SF
  • 3. Argus: Time-series data with Phoenix Tom Valine Salesforce.com
  • 4. OpenTSDB Limitations OpenTSDB is good, but we need more •Tag Cardinality – Total number of tags per metric is limited to 8 – Performance decreases drastically as tag values increase. •UID Exhaustion – Hard limit of 16M UIDs •Ad hoc querying not possible – Join to other data sources – Joins of time series and events – Simplification of Argus’ transform grammar
  • 5. Phoenix-backed Argus TSDB Service • 3 day hackathon • Modeled metric as Phoenix VIEW – Leverage ROW_TIMESTAMP optimization • Tag values inlined in row key – Uses SKIP_SCAN filter optimization – Allows for secondary indexes on particular metric + tags • Metric and tag names managed outside of data as metadata • Eventually leverage Drillix (Phoenix + Drill) – Cross cluster queries – Joins to other data sources
  • 6. Write Performance Using 2 clients to write in parallel. Phoenix is using 10 writer threads per client
  • 7. Read Performance • Metrics with one tag (60 distinct values) – OpenTSDB and Phoenix performance comparable for small aggregations – Phoenix outperforms OpenTSDB as aggregation size increases
  • 8. Disk usage • Phoenix & OTSDB use approximately the same amount of space with FAST_DIFF and Snappy compression
  • 9. Five major tips to maximize performance on a 200+ SQL HBase/Phoenix cluster Masayasu “Mas” Suzuki Shinji Nagasaka Takanari Tamesue Sony Corporation
  • 10. Who we are, and why we chose HBase/Phoenix • We are DevOps members from Sony’s News Suite team http://socialife.sony.net/ • HBase/Phoenix was chosen because of a. Scalability, b. SQL compatibility, and c. secondary indexing support
  • 12. Performance test apparatus & results • Test apparatus • Test results Specs Number of records 1.2 billion records (1 KB each) Number of indexes 8 orthogonal indexes Servers 3 Zookeepers (Zookeeper 3.4.5, m3.xlarge x 3) 3 HMaster servers (hadoop 2.5.0, hbase 0.98.6, Phoenix 4.3.0, m3.xlarge x 3) 200 RegionServers (hadoop 2.5.0, hbase 0.98.6, Phoenix 4.3.0, r3.xlarge x 199, c4.8xlarge x 1) Clients 100 x c4.xlarge Results Number of queries 51,053 queries/sec Response time (average) 46 ms
  • 13. Five major tips to maximize performance using HBase/Phoenix Ordered by effectiveness (most effective on the very top) – An extra RPC is issued when the client runs a SQL statement that uses a secondary index – Using SQL hint clause can mitigate this – From Ver. 4.7, changing “UPDATE_CACHE_FREQUENCY” may also work (we have yet to test this) – A memory rich node should be selected for use in RegionServers so as to minimize disk access – As an example, running major compaction and index creation simultaneously should be avoided Details will be presented at the PhoenixCon tomorrow (May 25) 2. Use memories aggressively 1. Use SQL hint clause when using a secondary index 4. Scale-out instead of scale-up 3. Manually split Region files if possible but never over split them 5. Avoid running power intensive tasks simultaneously
  • 15. eHarmony and Phoenix a perfect match NEED FOR ● Handling 30+ Million events during Batch Run ● Serving low latency queries on 16+ Billion records 75th% - 800MS 95th% - 2Sec 99th% - 4Sec
  • 16. eHarmony and Phoenix a perfect match LAMBDA FOR THE SAVE • Layered architecture provides fault tolerance • Hbase as batch storage for write throughput with reasonable read latency • Apache Phoenix as query layer to work with complex queries with confidence • Redis as speed layer cache
  • 17. eHarmony and Phoenix a perfect match PERFORMANCE Phoenix/HBase goes live Get Matches API Response Times Phoenix/HBase goes live Save Match API Response Times
  • 18. eHarmony and Phoenix a perfect match • Highly Consistent and fault tolerant • Need for store level filtering and sorting • Apache Phoenix helped us build an abstract high performance query layer on top of Hbase. • Eased the development process. • Reduced boiler plate code, which provides maintainability. • Build complex queries with confidence. • Secondary indexes. • JDBC connection. • Good community support WHY HBASE AND PHOENIX HBASE APACHE PHOENIX
  • 19. eHarmony and Phoenix a perfect match JAVA ORM LIBRARY(PHO) • Apache Phoenix helped us build PHO (Phoenix-HBase ORM) • PHO provides ability to annotate your entity bean and provides interfaces to build DSL like queries. Disjunction disjunction = new Disjunction(); for (int statusFilter : statusFilters) { disjunction.add(Restrictions.eq("status", statusFilter)); } QueryBuilder.builderFor(FeedItemDto.class).select() .add(Restrictions.eq("userId", userId)) .add(Restrictions.gte("spotlightEnd", spotlightEndDate)) .add(disjunction) .setReturnFields(projection) .addOrder(orderings) .setMaxResults(maxResults) .build();
  • 20. eHarmony and Phoenix a perfect match http://eharmony.github.io/ OPEN SOURCE REPOSITORY https://github.com/eHarmony/pho http://www.eharmony.com/about/careers/ *Please Join us for more details at PhoenixCon tomorrow (May 25)
  • 22. Why Transactions? • All or none semantics simplifies life of developer – Ensures every client has a consistent view of data – Protects against concurrent updates – No need to reason about what state data is left in if write fails – Guaranteed consistency between data and index
  • 23. Apache Tephra • Transactions on HBase – Across regions, tables and RPC calls • ACID semantics • Tephra Powers – CDAP (Cask Data Application Platform) – Apache Phoenix (4.7 onwards)
  • 24. Apache Tephra Architecture Zookeeper Tx Manager (standby) HBase Master 1 RS 1 RS 2 RS 4 RS 3 Client 1 Client 2 Client N Tx Manager (active) Master 2
  • 25. Tephra Components • TransactionAware client • Coordinates transaction lifecycle with manager • Communicates directly with HBase for reads and writes • Transaction Manager • Assigns transaction IDs • Maintains state on in-progress, committed and invalid transactions • Transaction Processor coprocessor • Applies server-side filtering for reads • Cleans up data from failed transactions, and no longer visible versions
  • 26. Snapshot Isolation • Multi-version concurrency control – Cell version (timestamp) = transaction ID – Reads exclude other uncommitted transactions (for isolation) • Optimistic Concurrency Control – Avoids cost of locking rows and tables – Good if conflicts are rare: short transaction, disjoint partitioning of work
  • 27. Single client using 10 threads in parallel with 5K batch size No performance penalty for non-transactional tables Performance
  • 28. Future Work • Partitioned Transaction Manager • Automatic pruning of invalid transaction list • Read-only transactions • Performance optimizations • Conflict detection • Appends to transaction edit log
  • 30. Integration model Calcite Parser & Validator Calcite Query Optimizer Phoenix Query Plan Generator Phoenix Runtime Phoenix Tables over HBase JDBC Client SQL + Phoenix specific grammar Built-in rules + Phoenix specific rules
  • 31. Cost-based query optimizer with Apache Calcite • Base all query optimization decisions on cost – Filter push down; range scan vs. skip scan – Hash aggregate vs. stream aggregate vs. partial stream aggregate – Sort optimized out; sort/limit push through; fwd/rev/unordered scan – Hash join vs. merge join; join ordering – Use of data table vs. index table – All above (any many others) COMBINED • Query optimizations are modeled as pluggable rules
  • 32. Beyond Phoenix 4.8 with Apache Calcite • Get the missing SQL support – WITH, UNNEST, Scalar subquery, etc. • Materialized views – To allow other forms of indices (maybe defined as external), e.g., a filter view, a join view, or an aggregate view. • Interop with other Calcite adaptors – Already used by Drill, Hive, Kylin, Samza, etc. – Supports any JDBC source – Initial version of Drill-Phoenix integration already working
  • 33. Query Example - no cost-based optimizer select empid, e.name, d.deptno, d.name, location from emps e, depts d using deptno order by e.deptno Phoenix Compiler scan ‘depts’ send ‘depts’ over to RS & build hash-cache scan ‘emps’ hash-join ‘depts’ sort joined table on ‘e.deptno’
  • 34. Query Example - with cost-based optimizer (sort optimization combined with join algorithm decision) LogicalSort key: deptno LogicalJoin inner, e.deptno = d.deptno LogicalProject empid, e.name, d.deptno, d.name, location LogicalTableScan emps LogicalTableScan depts PhoenixTableScan depts PhoenixMergeJoin inner, e.deptno = d.deptno PhoenixClientProject empid, e.name, d.deptno, d.name, location Optimizer Optimization rules + Phoenix operator conversion rules PhoenixTableScan emps PhoenixServerProjec t empid, name, deptno PhoenixServerProject deptno, name, location select empid, e.name, d.deptno, d.name, location from emps e, depts d using deptno order by e.deptno PhoenixServerSort key: deptno empid empid deptno deptno deptno e.deptno; d.deptno; e.deptno; d.deptno;
  • 35. Query Example - with cost-based optimizer (sort optimization combined with join algorithm decision) Phoenix Implementor PhoenixTableScan depts PhoenixMergeJoin inner, e.deptno = d.deptno PhoenixClientProject empid, e.name, d.deptno, d.name, location PhoenixTableScan emps PhoenixServerProjec t empid, name, deptno PhoenixServerProject deptno, name, location PhoenixServerSort key: deptno empid empid deptno deptno deptno e.deptno; d.deptno; e.deptno; d.deptno; scan ‘emps’ merge-join ‘emps’ and ‘depts’ sort by ‘deptno’ scan ‘depts’
  • 36. Query Example - Comparison Query plan w/o cost-based optimizer Query plan w/ cost-based optimizer scan ‘emps’, ‘depts’ first ‘depts’, then ‘emps’ 2 tables in parallel hash-cache send & build proportional to size of ‘depts’; might cause exception if too large none hash-cache look-up 1 look-up per ‘emps’ row none sorting sort ‘emps’ join ‘depts’ sort ‘emps’ only optimization approach Local, serial optimization processes Cost-based, rule-driven, integrated performance (single node, 2M * 2K rows) 19.46 s 13.92 s
  • 37. Drillix: Interoperability with Drill select deptno, sum(salary) from emps group by deptno Drill Final Aggregation deptno, sum(salary) Phoenix Table Scan emps Phoenix Tables over HBase Drill Shuffle Phoenix Partial Aggregation deptno, sum(salary) Stage 1: Local Partial aggregation Stage 3: Final aggregation Stage 2: Shuffle partial results
  • 38. Thank you! Questions? Join us tomorrow for PhoenixCon Salesforce.com, 1 Market St, SF 9am-1pm (some companies using Phoenix)