Building PetaByte Warehouses with Unmodified PostgreSQL. Emmanuel Cecchet Member of Research Staff May 21 st, 2009

Transcription

1 Building PetaByte Warehouses with Unmodified PostgreSQL Emmanuel Cecchet Member of Research Staff May 21 st, 2009 Topics Introduction to frontline data warehouses PetaByte warehouse design with PostgreSQL Aster contributions to PostgreSQL Q&A 2 PGCon 2009, Ottawa 2009 Aster Data Systems

2 Enterprise Data Warehouse Under Stress Enterprise Data Warehouse 3 PGCon 2009, Ottawa 2009 Aster Data Systems Offloading The Enterprise Data Warehouse Frontline Data Warehouse Enterprise Data Warehouse Archival Data Warehouse Petabytes source data 24 x7 Availability TB/Day Load capacity In-database transforms Rapid data access Petabytes detailed data Low cost/tb Flexible compression On-line access Aging out to off-line 4 PGCon 2009, Ottawa 2009 Aster Data Systems

3 Requirements for Frontline Data Warehouses! "! #$ % & ' & ($ ) * +, $ + -. / PGCon 2009, Ottawa 2009 Aster Data Systems Who is Aster Data Systems? Aster ncluster is a software-only RDBMS for large-scale frontline data warehousing High performance High availability High value analytics Low cost Always Parallel MPP architecture Always On on-line operations In-Database MapReduce Petabytes on commodity HW 6 PGCon 2009, Ottawa 2009 Aster Data Systems

4 MySpace Frontline Data Warehouse 0 -./ %% 0 1 * 2%3 1!+', # D<E FGH IJKLMKNJNO PJQR ST UVWXYZ[!"#! $%&' ()*+, \] ^_ `a^ bc _c dc e] f ghhi jklmno pqhh rs tuvutwrx 7 PGCon 2009, Ottawa 2009 Aster Data Systems Topics Introduction to frontline data warehouses PetaByte warehouse design with PostgreSQL Aster contributions to PostgreSQL Q&A 8 PGCon 2009, Ottawa 2009 Aster Data Systems

5 Petabyte Datawarehouse Design PostgreSQL as a building block Do not hack PostgreSQL to serve the distributed database Build on top of mainline PostgreSQL Use standard Postgres APIs Service Oriented Architecture Hierarchical query planning and optimization Shared nothing replication treating Postgres and Linux as a service Compression at the OS level transparently to PostgreSQL In-database Map-Reduce out-of-process 9 PGCon 2009, Ottawa 2009 Aster Data Systems Aster ncluster Database Queries/Answers Queen Server Group Queries Worker Server Group Data Loader/Exporter Server Group Aster ncluster Database Queen Node QKQ O O OL JKM N QLMJKO O Q QKR J QKRMK IJJ RMKQL O N M O Worker Node! " # $ % % &' Loader Node ( ) $ ) *+, - '. / % % 01 &' 10 PGCon 2009, Ottawa 2009 Aster Data Systems

6 Query Processing: How It Works C F5 F5 #, 32 5 C887> %& & %& & %& & %& &. A: Users send queries to the Queen (via ACT, ODBC/JDBC, BI tools, etc.) B: The Queen parses the query, creates an optimal distributed plan, sends subqueries to the Workers, and supervises the processing C: Workers execute their subqueries (locally or via distributed communication) D: The Queen aggregates Worker results and returns the final result to the user 11 PGCon 2009, Ottawa 2009 Aster Data Systems Table Compression Architecture Query RS TUVWXX?@ABCDEEF GDHIJA?@ABCDEEF GDHIJA?@ABCDEEF GDHIJA Logical Database (eg. Schema)?@ABCDEEF F K ILM?@ABCDEEF KILM K?@ABCDEEFF K ILM?@ABCDEEF KILM?@ABCDEEF F K ILM?@ABCDEEF KILM?@ABCDEEF KILM?@ABCDEEF N@O?@ABCDEEF N@O How It Works!"##$ % $#!"##$ &#'( ')' $&# *"#$ + $,,#"#+ &'#!# $#!#+$+) +!"#+ '#-#' Architecture Benefits:.)/!"#+ " 0&))#" &'1 2# 3#'$ "# -+)4 &# "+!"#+ 0##,,55"# 6)"# 5!)"$#4 7+5""#+3 0/)/8!#"," +# 5'8&'#!"#+4?@ABCDEEF GDHIJA?@ABCDEEF GDHIJA 6#"," +#9 :;#(8"(< =5#"# $+> +##$,5'' &'# $#!"#+?@ABCDEEF F K ILM?@ABCDEEF?@ABCDEEF KILM K F K ILM?@ABCDEEF?@ABCDEEF KILM F K ILM Physical Tablespaces YWZRS TUVWXX 12 PGCon 2009, Ottawa 2009 Aster Data Systems

7 Table Compression Enables Powerful Archival Compressed - High Compressed - Medium Compressed - Low Not Compressed Older data accessed less frequently Compress to save space and cost Oldest data is compressed the most, recent is compressed the least Compressed tables are fully available for queries (true online archival) 13 PGCon 2009, Ottawa 2009 Aster Data Systems What is MapReduce and Why Should I Care? What is MapReduce? Popularized by Google Processes data in parallel across distributed cluster Why is MapReduce significant? Empowers ordinary developers Write application logic, not debug cluster communication code Why is In-Database MapReduce significant? Unites MapReduce with SQL: power invoked from SQL Develop SQL/MR functions with common languages 14 PGCon 2009, Ottawa 2009 Aster Data Systems

8 Aster In-Database MapReduce Users, analysts, applications SQL/MR Functions In-Database MapReduce Data Store Engine Aster ncluster 15 PGCon 2009, Ottawa 2009 Aster Data Systems Slide 15 In-Database MapReduce Extensible framework (MapReduce + SQL) Flexible: Map-Reduce expressiveness, languages, polymorphism Performance: Massively parallel, computational push-down Availability: Fault isolation, resource management Out-of-process executables Does not use PL/* for custom code execution Can execute Map and Reduce functions in any language that has a runtime on Linux (e.g. Java, Python, C#, C/C++, Ruby, Perl, R, etc ) Standard PostgreSQL APIs to send/receive data to executables Fault isolation, security and resource management for arbitrary user code 16 PGCon 2009, Ottawa 2009 Aster Data Systems

9 Always On: Minimize Planned Downtime Rebalance Data Live Queries Add Capacity Data Backup Load & Export Backup & Restore 17 PGCon 2009, Ottawa 2009 Aster Data Systems Precision Scaling W V [ [S [ [ [ [. [ [ When more CPU/memory/capacity are needed, new nodes can be added for scale-out. Precision Scaling uses standard PostgreSQL APIs to migrate vworker partitions to new nodes either for load balancing (more compute power) or capacity expansion Example: Assume Workers 1/2/3 are 100% CPU-bottlenecked. Incorporation adds a new Worker4 node and migrates over vworker partitions D/H/L. As a result of loadbalancing, CPU-utilization drops to 75% per node, eliminating hotspots. 18 PGCon 2009, Ottawa 2009 Aster Data Systems

10 Replication 0 " 0" 1 Q Q LMLMJK J O MO M QL R QL QOL JK JK QKJL KJR K L KL J KJR QMN M QLMJK JL LO Q QMKOL RQLQ JOO QKR O OL JNLQ O 0 QKOQ KL LJ JOL O M J P QKR QLQ O MMK PMKN Q MMLM O LJ Q LN RQLQ QK O 19 PGCon 2009, Ottawa 2009 Aster Data Systems Fault Tolerance & Automatic Online Failover '(##) *+, -./012!"#!$!"#!%!"#!& !"#!4!"#! Replication Failover Automatic, non-disruptive, graceful performance impact Replication Restoration Delta-based (fast) and online (non-disruptive) 20 PGCon 2009, Ottawa 2009 Aster Data Systems

11 Using Commodity Hardware V [ M Q ML L K L IMO J NKM L... V [ [X NLMIJ I O NLM I O O MOL M NL R J MO M O Q MK Q Q ML NK L W X [ VW [ [ X [ [ X [ WX [ W VX QQ J QK M QKQ Q M QMQ PJ OL I 21 PGCon 2009, Ottawa 2009 Aster Data Systems Scaling On-Demand to a PetaByte Commodity Hardware More Blocks = More Power 2 TB Building Block Dell, HP, IBM, Intel x86 16 GB Memory 2.4TB of Storage 8 Disks $5k to $10k Node 7!>=<E "8 # < Massive Power Per Rack 160 Cores 640 GB RAM 48 TB SAS 22 PGCon 2009, Ottawa 2009 Aster Data Systems

12 Heterogeneous Hardware Support W V 8E <E CPU Memory Disk [ [ [ [ [ [ [ [ Heterogeneous HW support enables customers to add cheaper/faster nodes to existing systems for investment protection Mix-n-match different servers as you grow (faster CPUs, more memory, bigger disk drives, different vendors, etc) 23 PGCon 2009, Ottawa 2009 Aster Data Systems Topics Introduction to frontline data warehouses PetaByte warehouse design with PostgreSQL Aster contributions to PostgreSQL Q&A 24 PGCon 2009, Ottawa 2009 Aster Data Systems

13 Error Logging in COPY Separate good from malformed tuples Per tuple error information Errors to capture Type mismatch (e.g. text vs. int) Check constraints (e.g. int < 0) Malformed chars (e.g. invalid UTF-8 seq.) Missing / extra column Low-performance overhead Activated on-demand using environment variables 25 PGCon 2009, Ottawa 2009 Aster Data Systems Error Logging in COPY Detailed error context is logged along with tuple content 26 PGCon 2009, Ottawa 2009 Aster Data Systems

14 Error Logging Performance 1 million tuples COPY performance 27 PGCon 2009, Ottawa 2009 Aster Data Systems Auto-partitioning in COPY COPY into a parent table route tuples directly in the child table with matching constraints COPY y2008 FROM data.txt 28 PGCon 2009, Ottawa 2009 Aster Data Systems

15 Auto-partitioning in loading Activated on-demand set tuple_routing_in_copy = 0/1; Configurable LRU cache size set tuple_routing_cache_size = 3; COPY performance into parent table similar to direct COPY in child table if data is sorted Will leverage partitioning information in the future (WIP for 8.5) 29 PGCon 2009, Ottawa 2009 Aster Data Systems Other contributions Temporary tables and 2PC transactions [Auto-]Partitioning infrastructure Regression test suite LFI ( Fault injection at the library level or below out-of-memory conditions, network connection errors, interrupted system calls, data corruption, hardware failures, etc Lightning talk on Friday! 30 PGCon 2009, Ottawa 2009 Aster Data Systems

16 Topics Introduction to Aster and data warehousing PetaByte warehouse design with PostgreSQL Aster contributions to PostgreSQL Q&A 31 PGCon 2009, Ottawa 2009 Aster Data Systems PetaByte Warehouses with PostgreSQL Unmodified PostgreSQL Always Parallel MPP architecture Always On on-line operations In-Database MapReduce PetaByte on commodity Hardware 32 PGCon 2009, Ottawa 2009 Aster Data Systems

17 Aster Data Systems Learn more Free TDWI report on advanced analytics: asterdata.com/mapreduce Free Gartner webcast on mission-critical DW: asterdata.com/gartner Contact us 33 PGCon 2009, Ottawa 2009 Aster Data Systems Bonus slides 34 PGCon 2009, Ottawa 2009 Aster Data Systems

18 ncluster Components 5 C887> FG=G 6969 DE8EG> 1 / &' $ % & ' (!+, + "&)+&+, #%! # )%&, / ncluster 35 PGCon 2009, Ottawa 2009 Aster Data Systems Aster Loader ) 8G# 5 G7G9*69 +,-../01/ :79: ; <:=75 >75=?5@4<;7 A?99B7<7;CD EF.GHIF02 JKB9L>B7 3?4875: 9? MN?5C75:?< N?5C75 <?87:D O=975 >459L9L?<L<6P 3?4875 QLBB >454BB7B B?48 K<LRK L<9? MN?5C75:D S/0/TIH2 3L<745BU :;4B4AB7 B?48 >75=?5@4<;7 V=5?@ 3?4875: 9? N?5C75:W Scalable Partitioning +,-../01/2 A7 :;4B4AB7 4<8 <?9 L@>787 9Y7 B?48L<6 >5?;7:: EF.GHIF02 Z4;Y 3?4875 ;?<94L<: 4 [X459L9L?<75\P QYL;Y K:7: 4< 4B6?5L9Y@ 9? 4::L6< 8494 L<9? [AK;C79:\D Z4;Y AK;C79 L: 9? QL9YL< 9Y7 ;BK:975D S/0/TIH2 ]4:9P L<97BBL67<9 >459L9L?<L< B?48: _ G! 7= `G9 #76 9 +,-../01/2 ]4LB78 B?:7 8494?5 :L6<L=L;4<9BU 85?> B?48 >75=?5@4<;7D EF.GHIF02 a= 4 <?87 =4LB:P?9Y75 :9574@: ;?<9L<K7 9? B?48 Vbcd >75=?5@4<;7 YL9WD O8@L<: ;4< 74:LBU 57;?M75 B?: AU 57^B?48L<6 ekbc ] D S/0/TIH2 f494 B?:: >5?97;9L?< g >75=?5@4<;7 ;?<:L:97<;U 8K5L<6 <?87 =4LBK57D 36 PGCon 2009, Ottawa 2009 Aster Data Systems