benchmarkingblog

Benchmarking and Systems Performance

War and Peace with HP’s New Gen9

leave a comment »

I love my book club. We vote on what books to read and we always choose interesting and enlightening fiction and non-fiction. We always serve great food and drink. And, best of all, you don’t even have to read the book.

So I was really excited last month to hear about the book we would be reading. My good friend said she liked the book and the cover showed a woman in an olive grove. What could be better? Until I actually read the book and realized that there was nothing really that thrilling behind the cover.

And that’s exactly how I felt this week reading about HP’s newest announcement on their x86 HP ProLiant Gen9 servers. I mean, for goodness sake, we are at Gen9 — I’m sure we’ve come so far and there must be some thrilling new features and performance behind them.

But alas, once again, my excitement was thwarted by what I actually read.

There were 3 claims for these new systems:

  • Better TCO, optimization, and automation — but only comparing this new Gen9 to G6. I should hope there would be improvement.
  • Faster provisioning — The footnote says “Based on anonymous customer results” — hmmm. Plus the software isn’t even available yet for Gen9.
  • Improved performance — all it says is “equivalent controllers” in a “controlled environment.” — really hard to say what that means.

When I see claims like these with no real data all I can think is what real data looks like.

And it reminds me that the olive grove book is no Pride and Prejudice.

For next month we are reading Gabriel Garcia Marquez – so at least I will know to expect magic realism.

************************************************

The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.

technorati tags: , , ,,,,,,,,

Written by benchmarkingblog

August 29, 2014 at 2:27 pm

Posted in Uncategorized

POWER8 (and No. 23) is King

leave a comment »

So it was just announced that LeBron James will be back wearing No. 23 for his return to the Cleveland Cavaliers next season. Cleveland, the old “mistake by the lake” and my current hometown, has had a pretty good last few weeks. With two major coups — a US political national convention and (against all odds) LeBron coming home, there couldn’t be a better place to be.

POWER8 systems have had an awesome month as well. As we know, on top of enterprise RAS and security, performance of POWER8 is king. Let’s take a look at the two latest leadership benchmark results that have been published in two key workloads — Java and virtualization.

  • SPECjbb2013 is THE Java server benchmark. It is based on a world-wide supermarket company with an IT infrastructure that handles a mix of point-of-sale requests, online purchases and data-mining operations. Two metrics are important – a pure throughput metric and a metric that measures critical throughput under service-level agreements (SLAs) specifying response times. So very real world. And the IBM Power S824 (POWER8) system achieved over 95% better performance per core than  Oracle SPARC and over 2x better than x86 Cisco UCS. (1)

 

  • The SPECvirt_sc2013 virtualization benchmark reflects the heavy volume and sudden peaks of traffic faced by datacenter servers used for virtualized server consolidation. It provides four workloads based on real-world traffic faced by a web server, a Java application server, an email server and a batch server. Again, here are the workloads we use every day.And the IBM Power S824 (POWER8) system achieved the #1 result per core and is over 2x the performance per core of the HP DL560 x86 system. (2)

 

So we are talking a winning technology with POWER8 — and a winning city (LeBron, please take us there!)

************************************************

The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.

(1)SPECjbb2013: IBM Power S824 (POWER8, 3.52 GHz, 4 chips/24 cores/192 threads), 167958 SPECjbb2013-MultiJVM max-jOPS, 27041 SPECjbb2013-MultiJVM critical-jOPS vs. Oracle SPARC T5-2 (2 chips/32 cores/256 threads) 114492 SPECjbb2013-MultiJVM max-jOPS, 43963 SPECjbb2013-MultiJVM critical-jOPS vs. Cisco UCS C460 (4 chips/60 cores/120 threads) 201117 SPECjbb2013-MultiJVM max-jOPS, 52784 SPECjbb2013-MultiJVM critical-jOPS . Source: http://www.spec.org
(2)SPECvirt_sc2013: IBM Power S824 (POWER8, 3.52 GHz, 24-core), PowerVM Enterprise Edition 2.2.3, 1370@79 SPECvirt_sc2013@VMs vs. HP DL560 (32-core, Intel Xeon E5-4650), 908@50 SPECvirt_sc2013VMs. Source: http://www.spec.org

All results current as of July 28, 2014.

SPEC, SPECint, SPECfp, SPECjbb, SPECweb, SPECjAppServer, SPECjEnterprise, SPECjvm, SPECvirt, SPECompM, SPECompL, SPECsfs, SPECpower, SPEC MPI and SPECpower_ssj are trademarks of the Standard Performance Evaluation Corporation (SPEC).

technorati tags: , , ,,,,,,,,,,,

,,,,

Written by benchmarkingblog

July 28, 2014 at 3:41 pm

Posted in POWER8, SPECjbb, SPECvirt

Tagged with , ,

Go Fish: Why POWER8 May Just Be Perfect for Your Cloud

leave a comment »

Last weekend I spent cleaning out our fish pond. 5 years of sediment and dirt at the bottom, a monumental chore. But the most exciting part involved catching the goldfish and placing them in a temporary home (aka the kiddie pool the dog uses).

I thought it would be easy — but it was truly amazing how fast these fish swam. Eventually you were able to catch those beautiful slippery goldfish in your hands. And finally feel that you had a handle on the job.

I think that’s how many of us were thinking about Cloud in the past. Getting our hands around this slippery beast — needing to see real products and applications.

Of course we all know how awesome Power Systems are for performance and other non-functional requirements such as RAS and security. Well let’s see this extension of leadership in the cloud arena.

Power Systems servers are indeed architected to achieve maximum performance and efficiency for both the system and its virtual machines. Intelligent workload based resource allocation with dynamic processor thread switching and logical memory expansion deliver optimal performance for critical Cloud services.

Power Systems now have great alternatives for:

  • Public Cloud: An Open source Linux solution for scale-out clouds services with
    1) Flexibility, agility and interoperability with open source virtualization and cloud management
    2) Accelerated insights for big data and compute intensive Cloud services
  • Private Cloud: Pre-built, pre-installed entry cloud system for Enterprise and Scale Out clouds
    1) Get up and running with a private cloud — in just 1 day
    2) 0 common vulnerability exposures — that’s right, NONE
  • Hybrid Cloud: Next Generation with OpenAPIs
    1) Open alternative to proprietary cloud stacks
    2) Cross-platform support for x86, Power and System z
    3) Single pane of glass to monitor

So just as the fish are now happy and back in their crystal clear pond, we can be happy and choose our cloud alternatives with Power Systems – whether you like an open solution, the security of PowerVM on a choice of operating systems or a combination of the best of both worlds.

************************************************

The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.

technorati tags: , , ,,,,,,,,,,

,,

Written by benchmarkingblog

June 3, 2014 at 1:38 pm

Posted in Cloud, POWER8

Tagged with ,

It’s My POWER8 Party, Oracle

with 2 comments

There always has to be someone who tries to ruin the party.

In this case, Oracle recently published a blog entry claiming to address the POWER8 announcement. But funny. It didn’t really address the POWER8 announcement.

What it does try to do is scatter uncertainty about IBM. Here is what you need to know:

  • Oracle’s blog article doesn’t address POWER8 performance because of the simple fact that IBM has a wide portfolio of #1 benchmark results, including 6 new ones for POWER8 systems. (Note that two of those ironically happen to be using Oracle software.)
  • For their own hardware performance, Oracle states only that SPARC performance is increasing “with each release” — essentially uselessly comparing themselves with themselves.
  • Did Oracle happen to mention the awesome IBM POWER8 Open Server innovation? Of course not. And I wouldn’t want to bet my business based on Oracle’s claim of a “public roadmap” — a roadmap so basic it’s like trying to party in the dark without a flashlight.
  • As we know, hardware is behind everything that gets done in the world these days and “IBM will remain a leader in high-performance and high-end systems, storage and cognitive computing, and . . . continue to invest in R&D for advanced semiconductor technology.” See the IBM Annual Report for more.
  • With the trends in Cloud, Analytics, Mobile, and Social (CAMS) applications, there will be huge growth in the future in these focus areas — for hardware, software, and services — with positive implications for all of IBM, including Power Systems.
  • In terms of investment strategies, remember what the experts say: You must look not to the past but to the future. IBM continually shifts its portfolio of businesses, in line with its strategy to invest in higher-margin fields. Analysts have confirmed that IBM is making the right strategic moves, clearly investing in the future.

 

 

So get the party started . . . again.

************************************************

The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.

technorati tags: , , ,,,,,,,,,

,,,

Written by benchmarkingblog

May 1, 2014 at 5:11 pm

Posted in announcement, POWER8

Tagged with

Awesome POWER8 Benchmarks, Awesome Dessert

with 2 comments

New frozen yogurt establishments seem to be popping up everywhere. You know, the ones with the cute name, the pink and green decor, the pink and green spoons to match.

A key differentiator in this new wave of stores is the do-it-yourself aspect. But even more extraordinary is the mind-boggling array of toppings. Dozens, in some cases hundreds. I especially love the portfolio of berries that are offered — but my favorite happens to be the small pieces of chocolate that look like rocks.

These stores have pretty much bloomed everywhere these days — whether rural, suburban, or urban area. I had seen them first in Manhattan a couple of years ago; but I knew they had become a true game changer when I located one in, of all places, suburban Poughkeepsie.

IBM today has just formally announced new POWER8 systems, servers that allow data centers to manage staggering data requirements with unprecedented speed, all built on an open server platform. This game-changing infrastructure represents IBM’s singular commitment to providing higher-value, open technologies for the latest types of applications, including cloud, big data and analytics, and mobile and social computing.

Of course, performance is a key factor in this groundbreaking technology. Some of us may have heard about these new systems earlier; but today is the day if you are really into performance — and this dessert is the best part. IBM has just added 6 new #1 benchmarks to the already huge portfolio of existing record benchmarks. Let’s take a look at these for the new IBM Power S824:

 

What’s especially interesting about these 6 is that they represent a wide portfolio of excellence and value in a real world environment — from specific applications that you run everyday, like sales, payroll, and order management, to Java and even technical computing. And these are varied workloads (just like all those berries) from various vendors, including Oracle, that have been shown via popular and well-accepted third party benchmarks to surpass all other systems, including x86.

 

Benchmarks, pick your favorite #1.

Mine is still the chocolate that looks like rocks.

************************************************

The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.

(1)IBM Power System S824 on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 4 processors / 24 cores / 192 threads, POWER8; 3.52GHz, 512 GB memory, 21,212 SD benchmark users, running AIX® 7.1 and DB2® 10.5, dialog response: 0.98 seconds, line items/hour: 2,317,330, dialog steps/hour: 6.952,000 SAPS: 115,870 database response time (dialog/update): 0.011 sec / 0.019sec, CPU utilization: 99%, Certification #2014016. Source: http://www.sap.com/benchmark.
(2)The 12-core IBM Power S824 (3.52 GHz) achieved the best 12-core extra-large Oracle E-business 12.1.3 benchmark Payroll batch result (1,090,909 checks per hour). Source: http://www.oracle.com/us/solutions/performance-scalability/index.html
(3)The 6-core IBM Power S824 (4.1 GHz) database server achieved the best overall Siebel CRM 8.1.1.4 result (50,000 users).
Source: http://www.oracle.com/us/solutions/benchmark/white-papers/siebel-167484.html
(4)The 24-core IBM Power S824 (3.52 GHz) db running DB2 10.5 / 24-core IBM Power S824 (3.52 GHz) app running WebSphere 8.5 is the best 24-core SPECjEnterprise2010 configuration (22,543 Enterprise jAppServer Operations Per Second (EjOPS)) . Source: http://www.spec.org
(5)The 24-core IBM Power S824 (3.5 GHz, POWER8) is the best 24-core system (1370 SPECfp_rate2006 result, 24 cores, 4 chips, 6 cores/chip, 8 threads/core). http://www.spec.org
(6)The 24-core IBM Power S824 (3.5 GHz, POWER8) is the best 24-core system (1750 SPECint_rate2006 result, 24 cores, 4 chips, 6 cores/chip, 8 threads/core). http://www.spec.org

All results current as of April 28, 2014.

SAP and all SAP logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries. Other names may be trademarks of their respective owners.

Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.

SPEC, SPECint, SPECfp, SPECjbb, SPECweb, SPECjAppServer, SPECjEnterprise, SPECjvm, SPECvirt, SPECompM, SPECompL, SPECsfs, SPECpower, SPEC MPI and SPECpower_ssj are trademarks of the Standard Performance Evaluation Corporation (SPEC).

technorati tags: , , ,,,,,,,,,,

,,

Written by benchmarkingblog

April 28, 2014 at 7:24 am

Oracle’s SPARC Enhancements: Construction or Wind ?

leave a comment »

Two nights ago I spent a lovely 6 hours in the airport. Flight cancelled, next plane delayed for incoming aircraft, no runways to be had in one of the largest airports in the country. Announcement 1: There was only one runway because the others were under construction. Announcement 2: There was only one runway that could be used because the wind patterns were strange.

All you want is to get home to your couch and your dog. At the same time it would be great to get the real story on what is happening. Just because you want to know, you want it to make sense.

And that’s exactly how I was feeling again as I read one of Oracle’s recent press releases on the Fujitsu SPARC M10 “enhancements.” The claim was for “15 world records.” I decided to take a look at each one just to know — was it the construction or the wind ?

1. Oracle needed 2.5x more cores/memory than IBM. The IBM result was from 4 years ago.
2. Oracle needed 2x more cores/memory than IBM. The IBM result was from 4 years ago.
3. Oracle compared themselves with themselves.
4. Oracle compared themselves with themselves.
5. Oracle needed 2x more cores than SGI.
6. Oracle compared themselves with themselves.
7. Oracle needed 2x more cores than IBM.
8. Oracle compared themselves with themselves.
9. Oracle needed 4x more cores than IBM.
10. Oracle compared themselves with themselves.
11. Oracle picked on little x86.
12. Oracle compared themselves with themselves.
13. Oracle needed 16x more cores than IBM. The IBM result was from 6 years ago.
14. Oracle needed 8x more cores than IBM. The IBM result was from 6 years ago.
15. Oracle needed 8x more cores than IBM. The IBM result was from 6 years ago.

Also note that there are really only 4 different benchmarks here. And notably all but 2 of these 15 are in the Technical Computing space, using simple component type benchmarks.

So that’s the real story. The other real story is that if I had driven the 500 miles I would have been home much faster.

************************************************

The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.

technorati tags: , , ,,,,,,,,,,

,

Written by benchmarkingblog

April 11, 2014 at 2:49 pm

Posted in SPARC

Tagged with , , ,

Digging into SAP HANA on HP

with one comment

All of the snow has finally melted in my backyard. And what that means to me is that I don’t have to shovel for awhile. What that means to my dog is another story.

It’s a field day for him. Now he can get back to what he surely thinks is his real job – digging in the dirt. And what has he found? A plastic flower pot that he can chew on. 3 beat up tennis balls from the summer. A soup bone from 3 months ago. Treasures.

HP just announced this morning that they are delivering a “System with Faster Analytics Engine for SAP HANA Environments.” HP claims a 2x performance advantage over other solutions. Let’s dig into this claim and take a look at the facts:

  • The performance claim is based on the SAP BW Extended Mixed Workload Benchmark, a benchmark with only 4 results. And 3 of those results are from, you guessed it, HP.
  • The SAP BW-EML benchmark results that HP references in their footnote in this press release are from September 2013 — a lifetime ago in the benchmark world.
  • The HP system they reference in the press release is not even the system that is in the benchmark. HP’s new system is not even available.
  • Even if you did try to compare the HP and IBM results, it does not make sense. The HP and IBM results are in different categories of the benchmark, using a different number of records. The HP result used an application and database tier; the IBM result is on a central server.
  • Even if you ignored the fact that it doesn’t make sense to compare these results, if you do, the HP result used 2.5x the processing cores, 2x the memory, 4x the L1 cache, 2x the L2 cache per core. So the IBM result actually has 28% better throughput per core than the HP result.(1)

Please HP, dig more and come up with some new exciting treasures next time so we readers are not left in the dirt.

************************************************

(1) IBM Power System 7 Express Server Central Server: IBM Power System 750 Express Server, 4 processor / 32 cores / 128 threads, POWER7+, 4.06GHz, 32KB(D) + 32KB(I) L1 cache and 256KB L2 cache per core, 10MB L3 cache per core, 512GB main memory; certification #2013020 (SUSE Linux Enterprise Server 11, DB2 for I 7.1, SAP NetWeaver 7.30, 66,900 ad-hoc navigation steps/hr) vs.
SAP BW-EML Benchmark: Database tier: HP ProLiant DL580 G7, 4 processor / 40 cores / 80 threads, Intel Xeon Processor E7-4870, 2.40GHz, 64KB L1 cache and 256KB L2 cache per core, 30MB L3 cache per processor, 512GB main memory; Application tier: HP ProLiant BL680 G7, 4 processor / 40 cores / 80 threads, Intel Xeon Processor E7-4870, 2.40GHz, 64KB L1 cache and 256KB L2 cache per core, 30MB L3 cache per processor, 512GB main memory; certification #2013027 (SUSE Linux Enterprise Server 11,SAP HANA 1.0,SAP NetWeaver 7.30,129,930 ad-hoc navigation steps/hr)
http://www.sap.com. Results current as of 3/19/14.

SAP and all SAP logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries. Other names may be trademarks of their respective owners.

The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.

technorati tags: , , ,,,,,,,,,,,

Written by benchmarkingblog

March 19, 2014 at 12:19 pm

Posted in HP, SAP

Tagged with , ,

%d bloggers like this: