Sun World Record for Largest Data Warehouse - One Petabyte!

Thursday May 15, 2008

Sun SPARC Enterprise M9000 server, Sybase IQ, and BMMsoft Server managed one Petabyte of raw data. That was over 6 trillion rows of transactional data and more than 185 million content-searchable documents, emails, reports, spreadsheets and other multimedia objects! This even set a new Guinness World Record™.

This is twice the size of the largest commercial data warehouse known to date. The largest known database is Walmart which is said to have half a terabyte of data using the Teradata DB.

Like this post? | | | |

Posted at 07:01AM May 15, 2008 by BM Seer in Misc
Tags: m9000 performance tpc-h

Single system world record TPC-H Sun SPARC Enterprise M9000 Sun and StorageTek 2540

Thursday May 08, 2008

The Sun SPARC Enterprise M9000 configured with SPARC VI processors, Sun StorEdge 2540 Arrays, and running Solaris 10 combined with Oracle 11g achieved World Record TPC-H performance of 118,573.3 QphH@1000GB for non-clustered systems.

The TPC-H result demonstrates that the Sun SPARC Enterprise M9000 can handle the increasingly large databases required of DSS systems. Oracle delivered 13 GB/sec during the benchmark. editorial note: IBM has never proven delivered IO rates? Why does IBM only resort to quoting un-obtainable peaks?

• Why no single-system 4.7GHz or 5.0GHz Power6 on TPC-H? IBM has cluster results, this allows IBM to avoid comparisons to a single system.
• The Sun SPARC Enterprise M9000 outperformed the next best competitor non-clustered system, the HP Integrity Superdome by 69%.
• The Sun SPARC Enterprise M9000 outperformed the next best competitor non-cluster system, the HP Integrity Superdome by 18% on price/performance.
• The Sun SPARC Enterprise M9000 outperformed the clustered IBM xSeries 346 by 122%.
• The Sun SPARC Enterprise M9000 outperformed the clustered IBM xSeries 346 by 29% on price/performance.
• Sun StorageTek 2540 Array disk configuration - the 20x ST2540 configuration in this benchmark delivered sustained rates of 13.7 GB/sec and showed linear scaling from 1 to 20 arrays.
• This result demonstrates the effectiveness of Solaris 10 running Oracle 11g.

QphH = the Composite Metric (bigger is better), $/QphH = the Price/Performance metric (smaller is better)

System	Metric QphH	3 Year Total Sys $	$/QphH	QppH	QthH	CPUs	Storage Amount
Sun SPARC Enterprise M9000	118,573.3	$2,772,675	$23.38	114,725.4	122,550.2	32	34.8 TB
HP Superdome	69,999.0	$2,008,168	$28.69	90,909.1	53,898.5	32	39.7 TB
HP Superdome	68,100.6	$4,008,065	$59.00	83,041.7	55,847.7	64	40.6 TB

System	CPU	Cluster	CPU MHz	CPU	Operating System	Database	RDBMS+HW Avail date
Sun SPARC Enterprise M9000	32	N	2400	SPARC64 VI	Solaris 10	Oracle 11g	09/10/2008
HP Superdome	32	N	1600	Itanium2	Windows Server 2003	SQL Server 2005	06/18/2007
HP Superdome	64	N	1600	Itanium2	HP-UX 11.i V2	Oracle 10gR2	01/18/2006

Benchmark Description

The TPC-H benchmark is a performance benchmark established by the Transaction Processing Council (TPC) to demonstrate Data Warehousing/Decision Support Systems (DSS). TPC-H measurements are produced for customers to evaluate the performance of various DSS systems. These queries and updates are executed against a standard database under controlled conditions. Performance projections and comparisons between different TPC-H Database sizes (100GB, 300GB, 1000GB, 3000GB and 10000GB) are not allowed by the TPC.

TPC-H is a data warehousing-oriented, non-industry-specific benchmark that consists of a large number of complex queries typical of decision support applications. It also includes some insert and delete activity that is intended to simulate loading and purging data from a warehouse. TPC-H measures the combined performance of a particular database manager on a specific computer system.

The main performance metric reported by TPC-H is called the TPC-H Composite Query-per-Hour Performance Metric (QphH@SF, where SF is the number of GB of raw data, referred to as the scale factor). QphH@SF is intended to summarize the ability of the system to process queries in both single and multi user modes. The benchmark requires reporting of price/performance, which is the ratio of QphH to total HW/SW cost plus 3 years maintenance. A secondary metric is the storage efficiency, which is the ratio of total configured disk space in GB to the scale factor.

Disclosure Statement:

Sun SPARC Enterprise M9000 118,573.3 QphH@1000GB, $23.38/QphH@1000GB, avail 09/10/08, HP Integrity Superdome 69,999.0 QphH@1000GB, $28.69/QphH@1000GB avail 06/18/07, HP Integrity Superdome 68,100.6 QphH@1000GB, $59.00/QphH@1000GB avail 01/18/06, IBM xSeries 346 QphH@1000GB, $32.80/QphH@1000GB, avail 02/14/05, TPC-H, QphH, $/QphH tm of Transaction Processing Performance Council (TPC). More info www.tpc.org.

A 128-core (32-node 4-core) IBM Power 570 cluster (4.7 GHz, 64 chips, 256 threads) with DB2 is the best overall system at 10TB (343,551 QphH@10000GB, 32.89$/QphH, configuration available 04/15/08, Results as of 5/07/08). Note: Do not divide this result by 32 to guess at single node performance, do not compare $/perf between different GB tests, these are not permitted by TPC rules!

Results Summary

Audited Results
	Database Size:	1000 GB (Scale Factor 1000)
	TPC-H Composite:	118,573.3 QphH@1000GB
	Price/performance:	$23.38/QphH@1000GB
	Available:	09/10/2008 for Oracle (Sun HW/SW available 05/02/2008)
Number of Systems:		One Sun SPARC Enterprise M9000
Total Number Processors:		32
Processor/MHz of Server:		SPARC VI 2400 MHz / 6MB L2 Cache
Storage:		34.8 Terabytes of disk
Database:		Oracle 11g
Operating System:		Solaris 10 Update 4
Total 3 year Cost:		$2,772,675
Other Performance Metrics
	TPC-H Power:	114,725.4
	TPC-H Throughput:	122,550.2
	Database Load Time:	1:35:27

Like this post? | | | |

Posted at 06:17AM May 08, 2008 by BM Seer in Benchmark Issues
Tags: m9000 tpc-h

Sun's Delivered Memory Bandwidth Leadership: Stream Benchmark

Wednesday Apr 18, 2007

Sun has faster delivered memory bandwidth than the best that IBM or HP can do. The Sun SPARC Enterprise M9000 beat IBM p5 595 by 10% on Stream TRIAD benchmark. The Sun SPARC Enterprise M9000 beat the HP Integrity Superdome by 33% on Stream TRIAD benchmark. The Sun SPARC Enterprise M9000, running with 2.4GHz SPARC64 VI processors, delivered a Stream TRIAD benchmark result of 227.1GB/s.

Don't let the core count confuse you, IBM cores cost over twice Sun's cores. Look at the other benchmark results posted to see that IBM costs more, is slower, and has fewer cores - but it is the best IBM that offers.

Be careful to compare measured/delivered bandwidth, other vendors sometimes try to confuse with peaks.

Stream Performance Chart - GB/s (1 MB=10^9 B, *not* 2^x B, bigger is better)

System	GHz	cores	COPY	SCALE	ADD	TRIAD
Sun SE M9000	2.4	128	224.4	223.1	224.2	227.1
IBM p5 595	2.3	64	186.1	179.6	200.4	206.2
HP Integrity SuperDome	1.6	128	154.5	153.0	169.5	170.8
HP Integrity SuperDome	1.6	64	116.1	114.6	127.9	128.7
Sun SE M9000	2.4	64	114.9	114.6	130.0	134.4
IBM p5-575	2.2	8	77.9	81.2	96.7	100.5
Sun SE M8000	2.4	32	60.3	60.2	69.3	69.6
Sun SE M5000	2.15	16	24.8	24.8	25.2	25.3
Sun SE M4000	2.15	8	12.6	12.5	12.7	12.7

Benchmark Description

The STREAM benchmark is a simple synthetic benchmark program that measures sustainable memory bandwidth (in MB/s) for simple vector kernels. All memory accesses are sequential, so a picture of how fast regular data can be moved through the system is portrayed. Properly run, the benchmark displays the characteristics of the memory system of the machine and not the advantages of running from the systems memory caches.

STREAM counts how many bytes that were read plus how many bytes that were written. For the simple "Copy" kernel, this is exactly twice the number obtained from the "bcopy" convention. STREAM does this because three of the four kernels do arithmetic, so it makes sense to count both the data read into the CPU and the data written back from the CPU. The "Copy" kernel does no arithmetic, but for consistency, counts bytes the same way as the other three.

The sequential nature of the memory references is the benchmark's biggest weakness. The benchmark does not expose limitations in a system's interconnect to move data from anywhere in the system to anywhere.

Disclosure Statement:

Stream is a publically available benchmark and can be found at http://www.cs.virginia.edu/stream. Results as of 4/13/07.

System Configuration

Systems under test:

• Sun SPARC Enterprise M9000
• 64 x 2.4GHz SPARC64 VI processors
• 1TB memory
• Solaris 10
• Sun Studio 12

Like this post? | | | |

Posted at 02:52PM Apr 18, 2007 by BM Seer in Benchmarks
Tags: bandwidth m8000 m9000 memory stream

World Record SPECompL2001 on Sun SPARC Enterprise M9000

Wednesday Apr 18, 2007

The Sun SPARC Enterprise M9000 (2.4GHz SPARC64 VI) set a World Record on the SPECompL2001 benchmark with a score of 1230446. The Sun SPARC Enterprise M9000 (2.4GHz SPARC64 VI) set a World Record on the SPECompLbase2001 benchmark with a score of 1148235.

Don't let the core count confuse you, IBM cores cost over twice Sun's cores. Look at the other benchmark results posted to see that IBM costs more, is slower, and has fewer cores - but it is the best IBM that offers.

SPEComp2001 Performance Chart - SPECompL2001 (bigger is better, ordered by peak metric)

Result		Cores	Chips	OpenMP Threads	System
Peak	Base
1230446	1148235	128	64	128	Sun SE M9000, SPARC64 VI 2.4GHz
1056459	1005583	64	32	128	IBM p5-595, POWER5+ 2.3GHz
1005076	987139	256	128	256	SGI Altix 4700, Itanium2 1.6GHz

Benchmark Description

The SPEC OMPL2001 Benchmark Suite was released in June 2001 and tests HPC performance using OpenMP for parallelism. SPEC OMPL2001 consists of 9 programs (2 in C and 7 in Fortran) parallelized using OpenMP API.

Goals of suite:

Disclosure Statement:

SPEC, SPEComp reg tm of Standard Performance Evaluation Corporation. Results from www.spec.org as of 04/17/07. Sun results submitted to SPEC. Sun SPARC Enterprise M9000 (128 OMP threads, 128 cores, 64 chips, 2.4GHz) 1230446 SPECompL2001. SGI Altix 4700, Itanium 2 (256 OMP threads, 256 cores, 128 chips) 1005076 SPECompL2001. IBM p5-595 (128 OMP threads, 64 cores, 32 chips) 1056459 SPECompL2001. Sun SPARC Enterprise M9000 (128 cores, 64 chips, 128 OMP threads, 2.4GHz) 1148235 SPECompLbase2001.

Result
	M9000:		1230446 SPECompL2001
Reference Date:			Apr 17, 2007
System:			Sun SPARC Enterprise M9000
Total Number Processors:			64
Total Memory :			1 TB (512x2GB DIMMs)
Processor/GHz of Server:			SPARC64 VI, 2.4 GHz
Operating System:			Solaris 10
Compiler:			Sun Studio 12

Like this post? | | | |

Posted at 08:46AM Apr 18, 2007 by BM Seer in Benchmarks
Tags: m9000 sparc64 specompl2001

Sun SPARC Enterprise M9000 tops 1 TFLOP/s - twice as fast as IBM p595

Tuesday Apr 17, 2007

The Sun SPARC Enterprise M9000 outperforms the best published single system from IBM p5 595 (1.9GHz POWER5) by over 2X on the Linpack benchmark (Highly Parallel Computing). The Sun SPARC Enterprise M9000 also tops the high-end single-system Itanium 2 based system from HP (Superdome, 1.6GHz/24MB) by 38% on the Linpack.

Of the 3 vendors Sun, IBM and HP, only Sun can deliver over a TFLOP/s of performance in a single system on the Linpack HPC benchmark. (IBM, POWER5-based systems).

This benchmark also used the Sun Performance Library which as many routines important to scientific users. This library has been enhanced to take advantage of the SPARC64 VI architecture.

LINPACK HPC Performance - GFLOPS (bigger is better)

System	GFLOPS		Processors
	Total	Peak	Threads	CPUs	Type	GHz
Sun SPARC Enterprise M9000	1032.0	1228.8	128	64	SPARC64 VI	2.4
HP Superdome	745.5	819.2	128	64	Itanium 2	1.6
IBM p5 595	418.0	486.4	64	32	POWER5+	1.9

Benchmark Description

The Linpack benchmark suite measures the performance for factoring and solving a dense set of linear equations in double-precision floating-point.

The Linpack HPC benchmark allows the solution of any size matrix with a single right hand side. It was developed to allow vendors to show off their hardware. Because big problems allow for peak performance potentials, the benchmark is seen as an upper bound of potential performance of a machine. The run rules are much more flexible. The solution technique must use a pivoting scheme and the driver must follow the spirit of the Linpack 1000 or Linpack 100 benchmarks.

Disclosure Statement:

Linpack HPC, results from http://www.netlib.org/benchmark/index.html as of 04/13/07. Sun SPARC Enterprise M9000 (SPARC64 VI @2.4, 64 chips, 128 cores), 1.032 TFLOPS. IBM p5 595 (POWER5 1.9GHz, 32 chips, 64 cores) 418.0 GFLOPS. HP Superdome (Itanium 2 1.6GHz/24MB, 64 chips, 128 cores) 745.5 GFLOPS.

System Configuration

Like this post? | | | |

Posted at 01:50PM Apr 17, 2007 by BM Seer in Benchmarks
Tags: itanium2 linpack m9000 power5+ sparc sun