benchmarkingblog

Benchmarking and Systems Performance

Archive for the ‘Oracle’ Category

Oracle, I Knew You Were Trouble When You Walked In (to SPECjbb2013)

with one comment

I’m always on the lookout for anything that helps me eat well. My newest interest these days is a type of rye crisp cracker that looks like cardboard, tastes a bit better, but has absolutely no fat. (They are really good if you put a huge slice of brie on them.)

I also love reading about nutrition. I love Bloomberg’s ban on big soda. I love the concept of using color to focus on a well-rounded and healthy diet. Does lime jello count in the green group?

Pomegranate juice, which has many claims on being good for you, was one of those things. Deep red, and also great for an amazing sauce reduction for chicken. Well, I recently read that the Federal Trade Commission just barred claims of pomegranate juice helping heart disease and other ailments until truly proven — based on deceptive advertising claims.

Which reminded me of something else. Oracle recently claimed a “world record” in the SPECjbb2013 Java benchmark. What you need to know is that this benchmark does not have 100 other results to compare to. It doesn’t have 20. It doesn’t even have 10. This “world record result” is based on just two other published results. Both by Intel. With older JDK. And fewer and older processors.

Like when your black lab starts to sit on your friend’s chihuahua. Or a claim that your juice cures cancer.

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

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).

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

February 25, 2013 at 10:41 am

Posted in Oracle, SPECjbb

Tagged with ,

It Really Does Take Two to Tango, Oracle

leave a comment »

10x better, 95% reduction, 60x increase.

I saw these claims recently in press releases from Oracle.

Sounds rather nice. Until you start thinking about what a comparison really is.

A comparison involves the consideration of two things. You must have something to compare to something else. In mathematics, an inequality is a relation that holds between two values when they are different. Not just empty promises.

When you throw around “comparisons” about complex issues of application deployment, you really should mention what data you are comparing to. Rather than leaving it in thin air.

Are you comparing the performance of your newest processor with one from 10 years ago ? Are you comparing the deployment of your system with a competitor’s truck ?

That’s why it makes sense to look at real data. Like here. And here. And here. And here.

Remember that 60 times a very small number still comes out to be a very small number. And of course, as we know from the multiplicative property of zero, 60 times 0 is still 0.

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

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

December 10, 2012 at 11:30 am

Posted in Cloud, Exalogic, Oracle

Tagged with

Lackluster Larry

with 2 comments

As we await that big wave called OpenWorld to reach our shores, let’s take a look at what we have actually seen from Oracle in the last few days.

  • Questions on Oracle’s latest earnings and disappointing performance in hardware
  • A new ad on the front page of the Wall Street Journal that attempts to make a groundbreaking statement about the very old news that there are public clouds and there are private clouds.
  • A press release on a new TPC-C OLTP benchmark result.The problem is that it’s actually a Cisco benchmark result that just happens to use some Oracle software. And the problem is that the Cisco system in the comparison is 2 years newer, needs 2x the number of cores, and uses over 60% more storage than the IBM result cited. Even so, the older IBM system is 1.49x times the performance per core of the Cisco system. (1)

Based on this news plus what I’ve heard may be coming that huge OpenWorld wave may turn out to be merely a minor ripple in our IT landscape.

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

(1)Cisco UCS C240 M3, 1,609,186.39 tpmC, $0.47/tpmC, (2 processors/16 cores/32 threads) available 9/27/12 vs. IBM Power 780 Server Model 9179-MHB with IBM DB2 9.5, 1,200,011.00 tpmC, $0.69/tpmC, available 10/13/10 (2 processors/8 cores/32 threads). Results current as of 9/28/12. Source: http://www.tpc.org.
TPC-C ,TPC-H, and TPC-E are trademarks of the Transaction Performance Processing Council (TPPC).
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

September 28, 2012 at 11:42 am

Posted in Cisco, Cloud, Oracle, TPC-C

Tagged with , ,

Notes from Oracle, Chapter 3 in the Series

leave a comment »

Once again, I received a marketing email from Oracle this morning. And once again, hoping that it contained exciting new data on the T5, I was sadly disappointed.

  • I can not fathom why they continue to address the email to a bizarre truncation of my first name. I continue to question my confidence in a company that specializes in data.
  • The note stated “According to our records, you have older SPARC systems.” I should really check in my garage. Again, ironic for a company that wants clients to use them for data and record keeping.
  • The email links to a new video. From April.
  • The video claims a humongous performance improvement with SPARC T4. When you watch the video, it’s actually merely a comparison from Oracle’s very own baby M3000s consolidated to T4’s. So I would definitely expect even more than humongous.

I’ve gotten some notes and tweets about how amusing these “Notes from Oracle” are. And we all could certainly use a laugh these days. So please, Mr. Oracle Business Development Manager, keep those cards and letters coming.

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

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

September 6, 2012 at 12:13 pm

Posted in M3000, Oracle, SPARC T4

Tagged with ,

Oracle’s Historical Fiction

leave a comment »

I love reading historical fiction. It’s somewhat like the real thing, but not quite. And definitely more exciting.

This morning I received another email from Oracle that reminded me of just that.

  • The note was once again addressed to “Elis”. I can not fathom why they continue to truncate my first name to only four letters. Doesn’t give me much confidence in a software company that specializes in data.
  • The note contained a quote from an unnamed “Infrastructure Architect” from an anonymous “Major Credit Card” company on Oracle/Sun benefits. I’m thinking that just maybe that architect didn’t really want to be known throughout history for that one.
  • The note stated “Our records show that you have older SPARC systems.” How funny, maybe they’re in my basement hidden behind the washing machine. Again, a data company that surely got that historical data wrong.
  • The note links to an independent report on customer perspectives. The report is from May. May 2011. A lifetime ago in the IT world.
  • The independent report is full of various quotes, all from anonymous people at anonymous companies.
  • This independent report is written by a company called ORC. Hmmm, have I seen those letters somewhere before?

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

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 30, 2012 at 11:55 am

Posted in Oracle, SPARC T4

Tagged with ,

You’ve Got Mail, Weird Mail from Oracle

leave a comment »

This morning I received an email from Oracle. Not surprising since I signed up to get some of their communications a few years ago. What surprised me is what the email contained.

  • First of all, the email was addressed to “Elis”. Now I have gone by many nicknames in my past, but never “Elis”. I can only guess that for some reason Oracle’s programs can not handle the 9 characters in my first name and truncate to 4. Doesn’t give me much confidence in a software company that specializes in data.
  • The second part of the email to catch my eye was the picture displayed in the center. Not a cool logo, not a detailed shot of the hardware, but a picture of Larry talking.
  • Finally, and here is the clincher, the email focused on an announcement of a new system. At first I got all excited, what could this one possibly be. And then I read the details. The announcement cited in the letter dated today was actually from back in September 2011, almost 1 full year ago. And I have to ask, is that the best you can do Oracle, is there not anything else more recent than almost a year ago worth having me look at?

 

After I read the note, I decided not to delete it because I enjoyed it so much. And I had to ask myself — Was this even a true communication from Oracle or someone spamming a little fun here?

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

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

July 11, 2012 at 2:30 pm

Posted in announcement, Oracle, SPARC T4

Tagged with

Oracle, The Dates They are A-Changin´

with one comment

One of the great things about using industry standard benchmarks for performance evaluation is that you can trust the data. It is run according to benchmark kits developed by a committee. It is audited. It produces data and the results don’t change.

Except when they do.

One of the very useful pieces of data that is part of a TPC benchmark result, in addition to performance and price performance, is the availability date. This is the date for which all components, hardware and software are available for purchase. There are TPC rules around this: “The Committed delivery date for general availability (availability date) of products used in the price calculations must be reported. The Availability Date must be reported on the first page of the Executive Summary and with a precision of one day. When the priced system includes products with different availability dates, the reported availability date for the priced system must be the date at which all Components are committed to be Generally Available. Each Component used in the Priced Configuration is considered to be Available on the Availability Date unless an earlier date is specified.”

As a consumer of industry standard benchmarks, we trust that the vendor has made extensive and detailed plans to meet this date and that we can use this date in our own systems planning.

It is not very common to change this date after the result is published. It is not very common to change this date on the exact date that the system is supposed to become available. Oracle just did both — yesterday they delayed the system availability date on a TPC-C Sun Fire X4800 result.(1) If you take a look at the Full Disclosure Report you can see that the components that aren’t ready are no small extraneous part of the configuration — they’re huge DIMMs, memory modules.

I’m going to take a wild guess that these components that are not yet available are very much needed to produce this benchmark result.

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

(1)Oracle Sun Fire X4800 M2 server (8 chips/80 cores/160 threads) – 5,055,888 tpmC, US$.89/tpmC, available 7/10/12.
Source: http://www.tpc.org. Result current as of 6/27/12.
TPC-C ,TPC-H, and TPC-E are trademarks of the Transaction Performance Processing Council (TPPC).
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 27, 2012 at 11:34 am

Posted in Oracle, TPC-C

Tagged with , ,

%d bloggers like this: