|
|
|
|
Benchmark Information
| Subject |
Abstract |
pSeries Power4/Power5 Performance Comparison |
The goal of this document is to compare the competitive advantage of Power5 with AIX 5.3 servers compared to Power4 running AIX 5.2. Going further than the rPerf figures, a real 3-tier environment is tested, including a database and an application server. A large set of graphs (CPU af course, but also memory and disks) are provided to analyze the results. |
| Exercising and Measuring HACMP 5.1 on AIX 5.2 and FAStT with SWIFT SAG/SNL V5. Performance of IP Aliasing, Fast Failover and new custom policy of HACMP |
High Availability Solution. The goal is to validate and measure the performance of HACMP 5.1 under AIX 5.2. How much time is needed to detect different kinds of hardware failures or network failures, and how much time is needed to recover? The tests are based on FAStT. The FAStT configuration is explained, and SAN switches implementation is shown. The new features of HACMP 5.1, IP aliasing and Fast Disk Failover are tested. We've tested the new custom resource policies. We have tested a Resource Group with two IP service addresses on two different networks. |
| FAStT700 and AIX 5.2: Performances Tests using iozone |
The customer asked for a benchmark to measure sequential write performance on a FAStT700 in a pSeries environment using iozone tools (v172) to implement a file server. The purpose of this document is to give a feedback on the performance results and some information on the FAStT / pSeries implementation. |
| Mirroring over long distances on SAN: Performance and Tests to the limit on pSeries using real fibres, DWDM, MDS9000, ESS and iozone test tool, hardware and software mirroring comparison |
The tremendous growth of business applications is quickly heightening data-storage requirements and network expectations. Employees and customers demand uninterrupted access to corporate systems and data. Enterprises must respond to this surge in demand by providing robust, highly secure, interconnected SANs and geographically dispersed data-recovery solutions. Remote data-replication solutions commonly include disk array-based technologies that replicate and synchronize an exact copy of an application's data to a remote disk array. Although data replication can take the form of synchronous or asynchronous replication, for a guaranteed exact copy of the data, synchronous replication is the predominantly used method. Synchronous replication ensures an exact copy of a data set is present at the remote site. Therefore network latency is a critical consideration when building a synchronous-replication solution.
One of our customers needed to relocate his disaster recovery centre and requested to determine how far he could put his Disaster Recovery centre. He had to choose between two sites: one at 150 km and another at 465 km. The key factors were that he wanted a synchronous mirroring between the two sites and that he wanted to fallover with HACMP, each month, the production from one site to the other.
The customer asked for a benchmark with real fibres to measure the performance degradation created by the distance from 0 km to 465 km. We used pSeries p690 ESSs mod 800, CISCO directors and ADVA DWDMs. The I/O exerciser was iozone tools. The fallover was done by HACMP.
The purpose of this document is to give a feedback on the performance results. It can be used as a reference for Disaster Recovery planning. |
|
|
|
