|
Archived Performance Articles
|
23 Mar 2013
|
Joerg Moellenkamp on Performance Analysis [30245]
In a Word (or Three): 'Never Assume Anything'
Readers with the time (and the interest) might wish to have a go at Joerg Moellenkamp's treatise on performance analysis, which parses the "methodology" of that task in a more semantic than methods-oriented fashion. He begins by taking exception to Brendan Gregg's post "The USE method addresses shortcomings in other commonly used methodologies," exception that bears more on philology than performance analysis. There is no disputing Moellenkamp's advice in the end, however: The more you know, the better you will be at performance analysis.
(Get More Information . .)
|
|
22 Mar 2013
|
'Performance Tuning for Oracle Business Process Management Suite 11g' [30258]
White Paper Based on Experiences with Some of World's Largest BPM Deployments
"Performance Tuning for Oracle Business Process Management Suite 11g (BPM)" an Oracle white paper, should be read in conjunction with both the product documentation and Oracle Fusion Middleware Performance and Tuning Guide. The white paper itself presents a set of "best practices" collected from real world experiences with some of the largest and most critical BPM implementations worldwide. There is no globally relevant set of performance tuning recommendations for BPM because the applications built to run on top of BPM vary widely. These recommendations must therefore be tested against each of them to determine which ones are appropriate in a given environment.
(Get More Information . .)
|
|
08 Mar 2013
|
Oracle Produces World Record SPECjbb2013 Result with Oracle Solaris and Oracle JDK [30073]
Solaris Registers Accelerated Performance on both Composite max-jOPS and Composite critical-jOPS metrics
Oracle Solaris and Oracle JDK delivered a world record result on the SPECjbb2013 benchmark (Composite metric), a benchmark designed to showcase Java server performance, Brian posts. SPECjbb2013 is the replacement for SPECjbb2005. The major findings of this benchmark are:
- Oracle Solaris is 1.8x faster on the SPECjbb2013-Composite max-jOPS metric than the Red Hat Enterprise Linux result.
- Oracle Solaris is 2.2x faster on the SPECjbb2013-Composite critical-jOPS metric than the Red Hat Enterprise Linux result.
- The combination of Oracle Solaris 11.1 and Oracle JDK 7 update 15 delivered a result of 37,007 SPECjbb2013-Composite max-jOPS and 13,812 SPECjbb2013-Composite critical-jOPS on the SPECjbb2013 benchmark.
More details are available.
(Get More Information . .)
|
|
15 Feb 2013
|
Examining Oracle Solaris 10 1/13 Secure Copy Performance for High Latency Networks [29787]
Findings Reveal Significantly Improved Performance
With Oracle Solaris 10 1/13, the performance of secure copy or scp is significantly improved for high latency networks Brian posts. Oracle Solaris 10 1/13 enabling a TCP receive window size up to 1 MB has up to 8 times faster transfer times over the latency range 50 - 200 msec compared to the previous Oracle Solaris 10 8/11, he writes, adding that the default TCP receive window size of 48 KB delivered similar performance in both Oracle Solaris 10 1/13 and Oracle Solaris 10 8/11. Settings above 1 MB for the TCP receive window size delivered similar performance to the 1 MB results.
(Get More Information . .)
|
|
14 Feb 2013
|
Oracle VM for SPARC: Best Practices - Top Ten Tuning Tips [29782]
Jeff Savit Offers Both Tips and Cautions
While there are relatively few "tuning knobs" to adjust with Oracle VM Server for SPARC, there are nevertheless several best practices that can enhance or ensure performance. Jeff Savit's post includes 10 of these, along with several cautions about the use of best practices themselves. He advises users to be aware that best practices do not apply in every situation and that they undergo changes as technology changes. Also, once service level objectives have been met, it's best to quit tuning and move on to something else. Finally, the application of any best practice calls for a healthy measure of judgment.
(Get More Information . .)
|
|
06 Feb 2013
|
Siebel 8.1.1.4 Benchmark on SPARC T4 [29688]
Extends T-Server String of Siebel Benchmarks
A recent Siebel 8.1.1.4 benchmark on SPARC T4 as reported by Giri Mandalika ran Siebel application server instances on three SPARC T4-2/Solaris 10 8/11 systems with the Oracle database server 11gR2 configured on a single SPARC T4-1/Solaris 11 11/11 system. Several iPlanet web server 7 U9 instances with the Siebel Web Plug-in (SWE) installed ran on one SPARC T4-1/Solaris 10 8/11 system. Siebel database was hosted on a single Sun Storage F5100 flash array. Load balancing ensured near uniform load across all web and application server instances. All three Siebel application server systems consumed ~78% CPU on average with sub-second response times for the Siebel modules tested.
(Get More Information . .)
|
|
|
|
|
