Ultra-FAST Cryptography on the Sun UltraSPARC T2

Tuesday Oct 09, 2007

The UltraSPARC T2 processor has very low-overhead cryptography that basically allows one to add security at 'zero-cost'. A single Sun UltraSPARC T2 processor achieves up to 37,000 RSA 1024-bit signs/s and up to 38.9 Gbit/s of AES-128 throughput.

The comparisons below demonstrate the performance a single 1.4 GHz UltraSPARC T2 on RSA1024 (sign private key) and AES128-CBC operations

• The UltraSPARC T2 delivers over 4.1 times greater RSA1024 performance and 4.6 times greater AES128 performance than the 2-way quad-core 3 GHz Xeon.
• The UltraSPARC T2 delivers over 9.3 times greater RSA1024 performance and 10 times greater AES128 performance than the 2-way dual-core 2.6 GHz Opteron.
• The UltraSPARC T2 also delivers over 3 times greater RSA1024 performance and 15.6 times greater AES128 performance than a system using the Cavium Nitrox PX crypto acclerator card.
• The UltraSPARC T2 delivers over 30.8 times greater RSA1024 performance than the 2-way IBM p510 1.5 GHz Power5 .

To achieve these great results, the UltraSPARC T2 processor, has an on-chip cryptographic accelerator (SPU) that consists of a Cipher/hash unit and an enhanced modular arithmetic (MAU). This is an evolution of the previous generation UltraSPARC T1 that only contained modular arithmetic units.

Sun's UltraSPARC T2 processor introduces support for common bulk ciphers, secure hash operations and both prime and binary field Elliptic Cryptography. The UltraSPARC T2 processor supports RC4, DES, 3DES, AES-128, AES-192, AES-256, MD5, SHA-1, SHA-256.

Competitive Landscape

RSA/AES Cryptography Benchmark Performance as of 8/07/07 as measured by Sun on the following platforms.

System	Processor GHz	Chips total- cores	Operating System	1024bit RSA (K signs/s)	AES128 (Gbit/s)	notes
Sun SPARC Enterprise T5220	UltraSPARC T2 1.4 GHz	1 chip 8 core	Solaris 10	37.0 K	38.9 Gb/s	actual
Accelerator card	Sun SCA6000			13.0 K	1.0 Gb/s	actual
Sun Fire T2000	UltraSPARC T1 1.2 GHz	1 chip 8 core	Solaris 10	12.9 K		actual
Accelerator card	Cavium Nitrox PX			12.0 K	2.5 Gb/s	data- sheet
Sun FireT1000	UltraSPARC T1 1 GHz	1 chip 8 core	Solaris 10	10.8 K		actual
	quad-core Xeon 3 GHz	2 chip 8 core		9.0 K	8.4 Gb/s	actual
Sun Fire V490*	US IV+ 1.5 GHz	4 chip 8 core	Solaris 10	8.0 K		actual
IBM p690	Power4 1.3 GHz	16 chip 32 core	AIX 5.1	6.1 K		actual
Fujitsu PP850	SPARC64 V 1.9 GHz	16 chip 16 core	Solaris 10	6.0 K		actual
	Opteron 2.6 GHz	2 chip 4 core		4.0 K	3.9 Gb/s	actual
Sun Fire V40z	Opteron sc 2.6 GHz	4 chip 4 core	Solaris 10	3.3 K		actual
Dell PE 1850	Xeon 3.6 GHz	2 chip 2 core	Linux RHEL4 U1	1.9 K		actual
Dell PE 2850	Xeon 3.6 GHz	2 chip 2 core	Linux SLES 9	1.9 K		actual
IBM p510	Power5 1.5 GHz	1 chip 2 core	AIX 5.3	1.2 K		actual

* Used a Sun Crypto Accelerator (SCA) 4000 in the Sun Fire V490 testing.

Benchmark Description

The RSA/AES-128 Cryptography benchmark was developed by Sun to measure maximum throughput of RSA private key (sign) operations and AES-128 operations that a system can perform. On multi-chip and/or multi-core systems, multiple processes are used to achieve the maximum throughput. Two microbenchmark programs are used, pk11rsaperf/pk11aesperf on Solaris and OpenSSL speed test on non-Solaris systems. Though each microbenchmark uses different crypto APIs, they both measure the raw throughput of the same crypto operations.

• pk11rsaperf & pk11aesperf is part of a set of cryptographic microbenchmark programs internally developed by the Crypto Product Group of NSN. pk11aesperf measures the performance of AES-128-CBC processing, as performed by Solaris Cryptographic Framework via PKCS#11 API. Different key sizes, data sizes and varying numbers of concurrent threads can be tested. The metric is aggregate operations per second, for pk11rsaperf and Gb/s for pk11aesperf (for large object sizes).

• OpenSSL speed test, the standard microbenchmark included in the open-source OpenSSL package, measures raw cryptographic algorithm performance as implemented in the OpenSSL library - libcrypto.so via its own proprietary crypto APIs. For RSA the metric is operations per second, while for AES-128-CBC, the metric is Gb/s.

Disclosure Statement:

RSA/DSA Cryptography Benchmark Performance as of 08/07/07 as measured by Sun on the following platforms: Sun SPARC Enterprise T5220 37K RSA1024 signs/s, 38.9 AES128 Gb/s; Sun SCA6000 (actual) 13K RSA1024 signs/s, 1 AES128 Gb/s; Cavium Nitrox PX (datasheet) 12K RSA1024 signs/s, 2.5 AES128 Gb/s; 2-chip quad-core Xeon 3GHz 9K RSA1024 signs/s, 8.4 AES128 Gb/s; 2-chip dual-core Opteron 2.6GHz 4K RSA1024 signs/s, 3.9 AES128 Gb/s; Sun Fire T2000 1.2 GHz (8 cores, 1 chip) Solaris 10, 12,850 RSA1024 signs/s; Sun Fire T1000 1GHz (8 cores, 1 chip) Solaris 10, 10,764 RSA1024 signs/s; IBM p690 1.3 GHz (32 cores, 16 chips) AIX 5.1, 6,131 RSA1024 signs/s; Fujitsu PRIMEPOWER850 1.9 GHz (16 cores, 16 chips) Solaris 10, 6,038 RSA1024 signs/s; Dell PowerEdge 1850 3.6 GHz (2 cores, 2 chips) RHEL4 U1, 1,926 RSA1024 signs/s; Dell PowerEdge 2850 3.6 GHz (2 cores, 2 chips) SLES 9, 1,900 RSA1024 signs/s; IBM p5 510 1.5 GHz (2 cores, 1 chip, SMT) AIX 5.3, 1,200 RSA1024 signs/s.

Results Summary

Results			37.0 K RSA1024 signs/s
			38.9 Gb/s AES128
Reference Date:			August 7, 2007
Systems:			Sun SPARC Enterprise T5120/T5220
Total Number Processors:			1 chip / 8 cores/chip (8 threads/core)
Processor/GHz of Server:			Sun UltraSPARC T2 1.4 GHz
Operating System:			Solaris 10

David Patterson(UC Berkeley) on UltraSPARC T2

Thursday Sep 20, 2007

In a video, Prof. David Patterson opines on UltraSPARC T2 and how Sun's CMT has some very fresh ideas to move the industry forward on practical computing. He talks about the Old-fashioned and out-dated concepts of "peak" or "clock speed" and the need to look at delivered performance. here, here!!!

He shows that the UltraSPARC T2 out of box is almost 1.5x to 2x faster than Clovertown(quad-core) & Opteron and three to four times the watt/performance advantage. In addition, he says the UltraSPARC T2 is the easiest to program and auto-tune.

He did conceded that if you look at the archaic (he used the word "old-fashioned") 20th century metrics of peak and clock that the UltraSPARC T2 is 2x to 7x slower -- but he (like I) focus on delivered performance.

David Patterson is a Professor in Computer Science at Univ of California Berkeley. David and John Hennessy (Stanford University) wrote the textbook "Computer Architecture: A Quantitative Approach Fourth Edition"

AFTERNOTE #1

To respond the the comment below (comments are now closed). I'm sure the professor will give us more details and comparison of floating-point performance on important applications between the UltraSPARC T2 and the various X64 architectures, he's very complete and thoughtful.

In terms of other comparisons. There are cpu benchmarks (int & fp) comparisons that were done at UltraSPARC T2 launch, best chip in several comparisons. There will probably be more even results before long on commercial benchmarks.

AFTERNOTE #2

A nice summary of the video is at: blogs.sun.com/geetha/entry/berkeley_view

Sun UltraSPARC T2 & IBM Power6 comparison blogged about

Wednesday Aug 29, 2007

There is more preliminary UltraSPARC T2 performance is blogged about at: http://blogs.sun.com/jmeyer/entry/power6_goes_thud_part_v

Where John states:

And IBM knows that next quarter, Sun will be introducing systems based on the new UltraSPARC T2, the world's first true system-on-a-chip and the world's fastest microprocessor. Preliminary estimates on one popular benchmark show that a single rack of UltraSPARC T2-based systems will outperform four racks of 4.7GHz POWER6-based p5 570s (more on that as we get closer to system announcement). No kidding.

I haven't seen this internal info yet, but I'll try to dig it up. Looking at other tests, I believe this one.

...John also talks more about the lagging IBM POWER6 rollout.

Oracle & UltraSPARC T1 - Commercial databases and CMT are no problem

Thursday Aug 23, 2007

In the last posting we showed Oracle Database with SAP-SD benchmarks all running on a Sun Fire T2000. As Sun has been saying since Day one of CMT. Major databases are perfectly matched for UltraSPARC T1. By the way Sun has also used Open source databases on benchmarks as well.

We have lots of customers deploying RDBMS on UltraSPARC T1 and planning on UltraSPARC T2 servers. It really works well even though competitors and doubters want to try to say it is special purpose, sorry it isn't.

Here is an opinion:

"Now Sun's T2 is out and it's pretty much the world beater they promised - 30% faster on SPEC throughput than IBM's 4.7 Ghz Dual core Power6 and, more significantly, one third the cost and somewhere between two and three times the throughput of the Itanium. ... anyone still buying HP-UX and Itanium after Rock comes out will be doing it because they hate Sun and are quietly hoping for a miracle, just as DEC's partisans (and HP's own MPE customer base) did before them." -- zdnet's Paul Murphy

Source: "A Dumb prediction: IBM will Buy HP's Unix Customers," By Paul Murphy, zdnet, 08/17/07, http://blogs.zdnet.com/Murphy/?p=941

World Record Single-chip SAP-SD 2-Tier ECC 6.0 Sun Fire T2000

Thursday Aug 23, 2007

The SPARC Enterprise Model T2000 | Sun Fire Model T2000 is the performance leader in Two-Tier SAP-SD Standard Application Benchmarks on single processor systems as of August 22nd, 2007. This result used the Oracle database on the UltraSPARC T1. Again as Sun has always maintained the UltraSPARC T1 is good at database-tier, application tier, and web tier!

• Sun Fire Model T2000 supported 1100 SD Benchmark Users, 5530 SAPS, using Oracle 10g is the fastest single-processor systems.
• Sun Fire Model T2000 beats a 2-chip dual-core Itanium2-based HP Integrity rx2660.
• Sun Fire Model T2000 beats a 2-chip dual-core Opteron-based HP ProLiant DL365.
• Sun Fire Model T2000 beats a 2-chip dual-core Xeon-based Fujitsu BFi20 S2 (Unicode).
• The Fujitsu BX620 S4 that uses two-chip 3GHz Xeon Quad-cores is only 1.8x faster than a single chip Sun Fire Model T2000 using UltraSPARC T1.
• The IBM p570 that uses two-chip 4.7GHz POWER6 is only 1.8x faster than a single chip Sun Fire Model T2000 using UltraSPARC T1.
• The just-announced UltraSPARC T2 has twice the thread count of the UltraSPARC T1.

SAP-SD 2-Tier Performance, Benchmark Users (bigger is better)

Sys	Users	# / GHz / Type	Mem	OS	DB	LI/Hr	SAPS	BM rev	Date
IBM p570	2035	two 4.7 POWER6+ DC	32 GB	AIX 5L 5.3	Oracle 10g	203,670	10,180	6.0	5/21/07
Fujitsu BX620 S4	1940	two 3.0 Xeon QC	32 GB	Windows Srvr 2003 EE	SQL Server 2005	194,000	9,700	6.0	8/13/07
Sun Fire T2000	1100	one 1.4 US T1	64 GB	Solaris 10	Oracle 10g	110,670	5,530	6.0	8/22/07
HP Integrity rx2660	1090	two 1.6 Itan2 DC	32 GB	HP-UX 11iV3	DB2 9	109,670	5,480	6.0	3/20/07
HP ProLiant DL365	1083	two 2.8 Opt DC	32 GB	Windows Srvr 2003 EE	SQL Srvr 2005	108,670	5,430	6.0	2/9/07
Fujitsu BFi20 S2 Unicode	1020	two 3 Xeon 5160 DC	16 GB	Solaris 10	Oracle 10g	102,330	5,120	6.0	5/4/07
IBM p550	1000	four 1.9 POWER5+ DC	32 GB	SuSE Linux ES9	DB2 UDB 8.2.2	100,330	5,020	5.0	10/04/05
Sun Fire T2000	950	one 1.2 US T1	32 GB	Solaris 10	MaxDB 7.5	95,670	4,780	5.0	11/17/05
IBM x3250	850	one 2.13 Xeon	8 GB	Windows SrVr 2003 EE	DB2 9	88,000	4,400	6.0	5/11/07

Complete benchmark results may be found at the SAP benchmark website http://www.sap.com/benchmark.

Benchmark Description

The SAP Standard Application SD (Sales and Distribution) Benchmark is a two-tier ERP business test that is indicative of full business workloads of complete order processing and invoice processing, and demonstrates the ability to run both the application and database software on a single system. The SAP Standard Application SD Benchmark represents the critical tasks performed in real-world ERP business environments.

SAP is one of the premier world-wide ERP application providers, and maintains a suite of benchmark tests to demonstrate the performance of competetive systems on the various SAP products.

SAP has specified that the Benchmark Users metric is the only metric to be used for public comparisons. However, Benchmark Users can be traded off with response time in performance tuning, and so comparing Line Items per Hour or SAPS may be a different way to compare the actual power of systems.

Funny that Sun compares against current IBM results, IBM bloggers decide to do funny comparisons on a different SAP benchmark, but compared their latest system to a 16-month old result on a US-IV system that is 2 processor GHz upgrades behind. I guess that is one way to win...

Disclosure Statement:

Two-tier SAP Standard Sales and Distribution (SD) standard SAP ERP 2004/2005 application benchmark: SPARC Enterprise Model T2000 | Sun Fire T2000 (1-way, 1 proc, 8 cores, 32 threads) 1 x 1.4 GHz UltraSPARC T1, 64GB memory, 1100 SD Benchmark users, 1.91 sec avg response time, Cert#2007051, Oracle 10g, Solaris 10; Sun Fire T2000 (1-way, 1 proc, 8 cores, 32 threads) 1 x 1.2 GHz UltraSPARC T1, 32GB memory, 950 SD Benchmark users, 1.91 sec avg response time, Cert#2005047., MaxDB 7.5 database, Solaris 10; Fujitsu Siemens Computers PRIMERGY MOdel BX620 S4 (2-way, 2 procs, 8 cores, 8 threads), 2 x 3.0 GHz Quad-Core Intel Xeon, 32 GB memory, 1940 SD Benchmark users, 1.99 sec avg response time, Cert#2007049, SQL Server 2005, Windows Server 2003 Enterprise Edition; HP ProLiant DL365 (2-way, 2 procs, 4 cores, 4 threads) 2 x 2.8 GHz Opteron, 32GB memory, 1083 SAP SD Benchmark users, 1.98 sec avg response time, Cert#2007006, SQL Server 2005, Windows Server 2003 Enterprise Edition; HP Integrity rx2660 (2-way, 2 procs, 4 cores, 8 threads) 2 x 1.6 GHz Itanium, 32GB memory, 1090 SAP SD Benchmark users, 1.93 sec avg response time, Cert#2007016, DB2 9, HP-UX 11iV3; IBM System p 570 (2-way, 2 procs, 4 cores, 8 threads) 2 x 4.7 GHz POWER6+, 32GB memory, 2035 SD Benchmark users, 1.99s avg resp time, Cert#2007037, Oracle 10g, AIX 5L Version 5.3; Fujitsu Siemens Computers PRIMERGY Model BFi20 S2 (2-way, 2 procs, 4 cores, 4 threads) 2 x 3GHz Intel Xeon 5160 dual-core, 16GB memory,(Unicode) 1020 SAP SD Benchmark users, 1.94 sec avg response time, Cert#2007031, Oracle 10g, Solaris 10; IBM System x3250 (1-way, 1 proc, 4 cores, 4 threads) 1 x 2.13 GHz Xeon, 8GB memory, 850 SD Benchmark users, 1.59s avg resp time, Cert#2007036, DB2 9, Windows Server 2003 Enterprise Edition; IBM System eServer p5 550 (4-way, 4 procs, 4 cores, 8 threads) 4 x 1.9 GHz POWER5+, 32GB memory, 1000 SD Benchmark users, 1.97s avg resp time, Cert#2005040, IBM DB2 Universal Database 8.2.2, SuSE Linux Enterprise Server 9; SAP, R/3, mySAP reg TM of SAP AG in Germany and other countries. More info http://www.sap.com/benchmark.

Certified Results
	Performance:		1100 benchmark users
	Server:		Sun Fire
	Processors:		1 1.4 GHz UltraSPARC T1
	Memory:		64 GB
	Operating system:		Solaris 10
	Database S/W:		Oracle 10g
	SAP S/W:		SAP ECC 6.0
	SAP Certification:		2007051
	Storage:		Sun StorEdge 6020

UltraSPARC T2, and Old UltraSPARC T1 world records & new Xeon's

Thursday Aug 09, 2007

Postscript:

Just added a new Oracle/SAP-SD benchmark a couple of weeks after this posting, please see:
http://blogs.sun.com/bmseer/entry/world_record_single_chip_sap

Be careful when comparing performance results, as an example look at a comment in yesterday's "Can I use 64 threads in a chip?" posting. At least this comment pointed out that you can use 4-8 threads in 2 chip Intel-based systems, but it was really trying to be a stab at UltraSPARC Performance. Here was the comment:

SAP-SD benchmarks:

9330 SAPS 2 processors Quad-Core Intel Xeon Processor X5355 2.66 GHz
www.sap.com/solutions/benchmark/pdf/cert4207.pdf

5120 SAPS 2 processors / 4 cores / 4 threads, Intel Xeon 5160 3 GHz
www.sap.com/solutions/benchmark/pdf/cert3107.pdf

This is performance !?
4780 SAPS Sun Fire Model T2000, 1 processor / 8 cores / 32 threads
www.sap.com/solutions/benchmark/pdf/cert4705.pdf

Posted by Triffids on August 09, 2007 at 03:26 AM PDT #

One really needs to look at the complete data on those .pdf's to make a fair comparison (also in the disclosure statement below).

First: The T2000 SAP-SD used a 1.2GHz UltraSPARC T1, Sun now ships faster 1.4GHz UltraSPARC T1, and has announced 1.4GHz UltraSPARC T2. The 1.4GHz T2 has double the threads of that 1.4GHz (double the computational power).

Second: The T2000 SAP-SD result was submitted in Dec 2005, at that time it was near the performance of the expensive 4-way POWER5 IBM p550.

Third: The 2-chip Dual-core Xeon SAP-SD result above was submitted 18 months after the T2000 SAP-SD result.

Fourth: Different versions of the benchmark. The 2-chip Dual-core Xeon was run with ECC 6.0 (not SAP 5.0). The a newer version of the benchmark takes more computational work to produce the same results. Dual-core SAP-SD result was also run with Solaris 10 on Xeon, how cool is that!

Fifth: The 2-chip quad-core Xeon SAP-SD result above was submitted 19 months after the T2000 SAP-SD result.

Sixth: The Sun result used open-source MySQL MaxDB database, how cool is that! The Xeon results used Oracle or MicroSoft SQL Server.

postscript:
Sun latter used Oracle, others suggested US T1 has some sort of silly database limitation - NOT TRUE!

You'll see more results soon.

Triffids, as a reminder if you work for a partner company of SAP you must put the following disclosures when you post results. If you are not they you don't need to put this in, but as you can see the data in it would have allowed you to make a better comparison of systems. Don't worry I'm not asking you to identify yourself at all.

Disclosure Statement:

Two-tier SAP ECC 5.0 Standard Sales and Distribution (SD) benchmark Sun Fire T2000 (1-way, 1 proc, 8 cores, 32 threads) 1x 1.2 GHz UltraSPARC T1, 32 GB mem, 950 SD benchmark users, 1.91 sec avg response time, Cert#2005047., MaxDB 7.5 database, Solaris 10; Two-tier SAP ECC 5.0 Standard Sales and Distribution (SD) benchmark IBM System eServer p5 550 (4-way, 4 procs, 4 cores, 8 threads) 4x 1.9 GHz POWER5+, 32GB mem, 1,000 SD benchmark users, 1.97s avg resp time, Cert#2005040, IBM DB2 Universal Database 8.2.2, SuSE Linux Enterprise Server 9; Two-tier SAP ECC 6.0 Standard Sales and Distribution (SD) benchmark Fujitsu Siemens Computers PRIMERGY Model BFi20 S2 (2 procs, 4 cores, 4 threads) 2x Intel Xeon 5160, 3.0 GHz, 16GB mem, 1,020 SD benchmark users, 1.94s avg resp time, Cert#2007031, Oracle 10g, Solaris 10; Two-tier SAP ECC 6.0 Standard Sales and Distribution (SD) benchmark Fujitsu Siemens Computers PRIMERGY Model TX300 S3 (2 procs, 8 cores, 8 threads) 4x Quad-Core Intel Xeon Processor X5355 2.66 GHz, 32GB mem, 1865 SD benchmark users, 1.99s avg resp time, Cert#2007025, SQL Server 2005, Windows Server 2003 Enterprise Edition; SAP, R/3, mySAP reg TM of SAP AG in Germany and other countries. More info www.sap.com/benchmark.

I edited in:
2 processors into Quad-Core Intel Xeon Processor X5355 2.66 GHz

...and..

32 threads to the Sun Fire T2000, 1 processor / 8 cores ...in order to make the comparisons more consistent.

[8] Comments

• >First
  Intel already ships X5355 2.66 GHz and h...

  
  

  
  

  
  

  
  
  

• It strikes me that benchmarks for computers are im...

  
  

  
  

• Triffid, Glad you agree that these Xeon results we...
  
  
  
  

  
  

  
  
  
  

  
  
  
  
  

• >ok TWO quad-cores 19 months latter is twice th...

  
  
  
  

  
  
  
  

  
  

• We'll see...

  but customers do see the value of U...
  
  
  

• >reducing storage power and cooling costs by 60...
  

• I'm still waiting for Oracle or DB2 to publis...
  

• Post a comment

Like this post? | | | |

Posted at 09:16AM Aug 09, 2007 in Benchmarks
Tags: cmt opensource performance sap-sd solaris sparc t1 t2 t5120 t5250 ultrasparc xeon

Can I use 64 threads in a chip?

Wednesday Aug 08, 2007

Can someone really use 64-threads in a chip? The answer is simple, when you look out into your datacenter do you see racks of servers or just a single naked core sitting alone in the back corner?

If you see racks of server you are running lots and lots of threads. Think of it his way, if you have a bunch of dual-core single-socket 1RU servers filling a rack you have around 80 threads in a rack, or 2-socket you have 160, or quad-core 2-socket you have 320 threads.

Now how would you judge performance of a single rack (with 80-320 threads)? Would you run one copy of "gzip" or "tar" and compare that to your laptop and say that rack is slow, of course not., You'd run a whole bunch of them.

So when you are performance testing an UltraSPARC T1 or UltraSPARC T2 server throw lots of work at it and it will have no problem. There is massive parallelism in every datacenter with racks of servers. Perfect for UltraSPARC T1/T2. Every datacenter with web-tiers, application-tiers, and database behind those tiers runs tons of threads. And remember the UltraSPARC T1 and introduction and even last week continues to set leading performance records at every tier.

Intelligence test Would you judge performance of an UltraSPARC T2 by running a single "gzip" or "tar"?

"Estimated" what does that mean for Sun's UltraSPARC T2

Wednesday Aug 08, 2007

Why does Sun designate yesterday's performance results as "estimates", why that word? Did some Sun marketeer just throw a dart and just pick a big number. No. All UltraSPARC T2 SPEC CPU and SPEC OMP metrics quoted are from full “reportable” runs, but are nevertheless designated as “estimates” because they use pre-production systems. Sun customer systems, to be announced later, are expected to perform similarly. SPEC rules do allow comparing these preliminary scores and published result.

Is Sun the only vendor to use this clause? No. Intel and AMD have made a long history of using preliminary numbers at chip announcements to get the word out about their performance. Sun is just following their lead, and trumping their performance

Ok, back to why the word "estimates?" The SPEC CPU committee voted to use that specific word for preliminary scores. Members include IBM, Intel, AMD, HP, .... And every employee of a member company must follow the rules.

By license agreement, SPEC members and customers agree to run and report results as specified in each benchmark suite's documentation. from SPEC FAQ

Postings on Sun's UltraSPARC T2 performance:
http://blogs.sun.com/bmseer/entry/performance_of_the_new_sun
http://blogs.sun.com/bmseer/entry/ultrasparc_t2_more_floating_point
http://blogs.sun.com/sprack/entry/ultrasparc_t2_world_class_crypto
OpenSPARC T2:
http://blogs.sun.com/d/entry/ultrasparc_t2_documentation_available
Ubunu (aready booted on UltraSPARC T2):
Ubuntu & Canonical & UltraSPARC T1 (May06).

As a Sun employee I try my best to follow every rule when talking about results in public, but I'm an engineer so sometimes it is hard to follow all the legalese so I try to correct things as soon as I see an error. And I do my best to remind other Sun bloggers to put in the proper disclosure statement for SPEC & TPC benchmark results. Though quite honestly I wish SPEC & TPC would streamline the rules, make them more consistent, and minimize the lengthy disclosure statements.

Of course because Sun is in the lead and because I made some suggestions, I'm sure this entry will be fully scrutinized by every competitor. If I made errors let me know in the comments and I will correct them.

Disclosure Statement

SPEC, SPECint, SPECfp, and SPEComp registered trademarks of Standard Performance Evaluation Corporation. Results from www.spec.org as of August 6, 2007. Actually this one is short because I didn't put any specific results in this posting, the ones at the links have the more extensive disclosures because they show scores & results.

Solaris and Sun Studio compiler important to UltraSPARC T2 announcements & benchamrks

Tuesday Aug 07, 2007

Beyond UltraSPARC T2 what other technologies matter? There are two more keys to Sun providing such effective performance in the new single-chip Sun UltraSPARC T2 64-thread processor, that is Solaris (and now of course OpenSolaris) and Sun Studio compilers. Here is a nice slide of the history of hardware history of SPARC, I borrowed this on from an entry in "On the Record"

An important thing to remember that besides Sun's long history with SPARC, we've also lead the way in parallelism. Over 15 years ago, Solaris supported 64-way SPARC systems and provided near-linear scaling. For those of you old enough to remember, at that time IBM, SGI, HP, and everyone else thought there was no way Sun could produce effective 64-way systems. They were wrong and now our competitors have finally all have introduced systems with lots of processors and/or threads.

Solaris and Sun Studio compilers have a LONG history and lots of experience with industrial-strength applications with lots of threads.

Solaris and Sun Studio compilers were great at scaling to 64-way systems 15 years ago, with a lot more experience and hard work we are even better at scaling and will scale to lots more threads right now. Many thanks to all of those compiler & OS engineers!

Postings on Sun's UltraSPARC T2 performance:
http://blogs.sun.com/bmseer/entry/performance_of_the_new_sun
http://blogs.sun.com/bmseer/entry/ultrasparc_t2_more_floating_point
http://blogs.sun.com/sprack/entry/ultrasparc_t2_world_class_crypto
OpenSPARC T2:
http://blogs.sun.com/d/entry/ultrasparc_t2_documentation_available

...I've focused on Solaris, but there are options, for example Ubuntu. Ubuntu has already booted on the UltraSPARC T2.

As as a reminder Ubuntu and Canonical proved it on an UltraSPARC T1 almost 14 months ago, see this article on that work.

UltraSPARC T2: more floating-point performance

Tuesday Aug 07, 2007

More about floating-point on the Sun UltraSPARC T2 in this posting, In the previous posting SPECfp_2006 scores and the UltraSPARC T2 design being open-sourced were discussed.

In the UltraSPARC T2 there are eight floating-point units that are well suited for scientific applications. Based upon preliminary runs the Sun UltraSPARC T2 processor at 1.4 GHz beats all single chip scores showing 14230(est)/15081(est) SPECompMbase2001/SPECompMpeak2001.

How do these preliminary runs (we must use the term "estimated" by SPEC rules) compare to SPECompMbase2001/SPECompMpeak2001 scores?

• These Sun UltraSPARC T2 1.4GHz processor scores beat the best single-chip IBM p520 POWER5+ 1.9GHz processor published result by 85%.
• ...Sun is waiting for POWER6 4.7GHz results, maybe UltraSPARC T2 results will scare IBM from ever publishing a single-chip result?

Benchmark description:

The SpecOMP benchmark is a test of the performance of 9 High Performance computing applications. It is used to compare the performance of shared memory servers. All C/C++ and FORTRAN applications in this suite use the OpenMP programming model that provides a portable, scalable model for developing parallel applications for platforms ranging from the desktop to the supercomputer.

The OpenMP Application Program Interface (API) supports multi-platform shared-memory parallel programming in C/C++ and Fortran on all architectures, from the largest Unix servers to the small Windows NT platforms.

Disclosure statement:

All UltraSPARC T2 SPEC CPU metrics quoted are from full “reportable” runs, but are nevertheless designated as “estimates” because they use preproduction systems. SPEC, and SPEComp registered trademarks of Standard Performance Evaluation Corporation. Sun UltraSPARC T2 1.4GHz (1 chip, 8 cores, 64 threads) 14230 (est)/ 15081 (est) SPECompMbase2001/SPECompMpeak2001. Competitive results from www.spec.org as of August 6, 2007. IBM p520 1.9GHz (1 chip, 2 cores, 4 threads) published 8141/8174 SPECompMbase2001/SPECompMpeak2001.

Performance of the new Sun UltraSPARC T2

Tuesday Aug 07, 2007

Sun UltraSPARC T2 is an amazing chip and very fast! The UltraSPARC T2 features several industry firsts:

• Eight cores and 64 threads
• Integrated 10 GbE networking and I/O
• Dedicated, cryptographic and floating point units per core
• 10 cryptographic functions supported with hardware
• open-source design: www.opensparc.net

Based upon preliminary runs, the Sun UltraSPARC T2 processor at 1.4 GHz, beat all single chip scores showing 78.3 est. SPECint_rate2006. How do these preliminary runs (we must use the term "estimated" by SPEC rules) compare to SPECint_rate2006 results.

• These Sun UltraSPARC T2 1.4GHz processor scores beat the best single-chip IBM POWER6 4.7GHz processor published result by 29%.
• These Sun UltraSPARC T2 1.4GHz processor scores beat the best single-chip estimated scores of the AMD Barcelona by 23%.
• These Sun UltraSPARC T2 1.4GHz processor scores beat the best single-chip published scores of the 2.66GHz Intel X5355 (Clovertown) by 48%.

Based upon preliminary runs, the Sun UltraSPARC T2 processor at 1.4 GHz, beat all single chip scores showing 62.3 est. SPECfp_rate2006. How do these preliminary runs (we must use the term "estimated" by SPEC rules) compare to SPECfp_rate2006 results.

• These Sun UltraSPARC T2 1.4GHz processor scores beat the best published single-chip IBM POWER6 4.7GHz processor result by 7%.
• These Sun UltraSPARC T2 1.4GHz processor scores beat the best single-chip estimated scores of the AMD Barcelona by 11%.
• These Sun UltraSPARC T2 1.4GHz processor scores beat the best single-chip published scores of the 2.66GHz Intel X5355 (Clovertown) by 66%.

Performance per core doesn't matter GHz doesn't matter, what matters is numbers of cores, efficiency, and design of the chip! Competitors are saying that UltraSPARC T2 is proprietary... this makes no sense. both UltraSPARC T1 and UltraSPARC T2 are open source designs (www.opensparc.net). You do not find the latest design of Intel, AMD, or IBM as open source designs.

Disclosure Statement:

All Sun UltraSPARC T2 SPEC CPU metrics quoted are from full “reportable” runs, but are nevertheless designated as “estimates” because they use preproduction systems. SPEC, SPECint, SPECfp registered trademarks of Standard Performance Evaluation Corporation. Sun UltraSPARC T2 1.4GHz (1 chip, 8 cores, 64 threads) 78.3 est. SPECint_rate2006, 62.3 est. SPECfp_rate2006. Competitive results from www.spec.org as of August 6, 2007. IBM POWER6 4.7GHz (1 chip, 2 cores, 4 threads) 60.9. SPECint_rate2006, 58.0 SPECfp_rate2006. AMD Barcelona 2.6 GHz (1 chip, 4 cores, 4 threads) 63.9 est SPECint_rate2006, 56.3 est. SPECfp_rate2006. Barcelona estimates based upon "The Register" article stating 2.6GHz quad is 21% and 50% faster than Intel 2.66 system. Fujitsu RX300 Intel X5355 2.66 GHz (1 chip, 4 cores, 4 threads) 52.8 SPECint_rate2006, 47.5 SPECfp_rate2006.

Reminder: The Niagara 2 score was obtained from a full "reportable" SPEC run, but is designated as an "estimate" because a pre-production system was used.

...more information on the UltraSPARC T2 later today.

Lots hitting the wires: UltraSPARC T2, the next generation

Monday Aug 06, 2007

Many news sources now covering UltraSPARC T2, the new high-performance chip from Sun. This new UltraSPARC T2 chip leads in many ways. I'll cover the performance numbers tomorrow.

For now:
http://www.computerworld.com.au/index.php/id;898889798
http://www.reuters.com/article/technologyNews/idUSN0625780420070806
http://www.channelweb.co.uk/vnunet/news/2195718/sun-lifts-lid-niagara-processor
etc..

For some of my previous comments:
http://blogs.sun.com/bmseer/entry/news_trickles_out_on_niagara2

Please remember that the previous generation chip, the UltraSPARC T1, just set an application-tier world record (all details at link). How many times has the "old" chip with half as many threads set a world record weeks before the new one is announced?

A final note. I venture that this chip is going to lead for database, application tier, and of course web tier, oh and don't forget HPC, yes it is that versatile.