SAP Network Performance and Availability Activities
I would like to share with the SAP Network Community some of the activities we have been doing and plan to do on the site during 2006 in the area of performance and availability improvements. This will give you a better understanding of what we have been working on and the expected changes to look forward to during the rest of the year. So far during 2006 we have implemented the following availability and performance improvements to the site: 1. Adding of front end servers (Portal, Discussion Groups, etc…) to our landscape to cope with increase of SDN Traffic. 2. New file Server Cluster. 3. Distribution of SDN’s various Data Bases across different Data Base Servers. 3. Implementation of additional caching mechanisms at differnt levels (content,navigation, permissions, etc…) 4. Additional Indexes and SQL Statements optimizations to Data Bases. 5. Implementation of Netli technology (www.netli.com) which allows us to provide better performance for users which are located in remote locations (compared to our servers which are in Germany). For example: performance of the site in regions such as Asia Pacific has been improved significantly. With the above changes the availability of the site has increased from 98% (average during 2005) to 99.2% (current average during 2006). What’s next? During November we plan to upgrade SDN to NW04 in which we expect to further improve our availability and performance with the following changes: 1. Implementation of SSL accelerator. This will allow us to better distribute the load between the various servers. 2. Support for Online Backup. There will be no more need to bring down SDN during the weekend for an offline backup. 3. Overall new landscape with additional and more powerful servers. 4. New Portal version (Upgrade from EP 6.0 to NW04) 5. New Discussion Groups Version (Upgrade from Jive 3.x to 4.x) 6. Refactoring and optimization of outdated code. 7. Backend search improvements We are trying to test as much as possible on our staging environment before the go live however it is a very big change and there might be a short period of disruption and stabilization time required for which I apologize upfront for any inconvenience this may cause. I hope you found this short blog useful and that it gave you some insight in to the various activities we are doing to further improve the site. Happy Surfing! Ran Cliff Senior Director, SAP Network Platform.
Please note that the massive work Ran and his team are doing are directly related to your feedback.
For example, we do twice-annual surveys of the SDN and BPX membership (this summer we had 4200 thorough survey responses from a very passionate community!) and we learn alot.
Aside from those distinct projects, we watch the forums constantly to look for dissatisfiers we can remove, improvements we can make, for trends that are cropping up... And of course we watch server response times, uptime, etc. thru various means (both scientific and anecdotal).
One of the things we noticed was that site performance needed attention. (When you grow a community to nearly 600,000 members and 24x7 activity across 200 countries in three years you've got to expect a bit of that, right?)
So, as a note to Ran's blog I just want to point out that all of this effort (and expense!) is a direct result of our SDN and BPX member feedback as we work hard (Ran's team at the top of the "overworked" stack ranking) to make these sites and these communities more valuable and satisfying to their members (our customers, partners, SAP colleagues, and supporters).
We hope you noticed the improvements over the past several months, and hope you see big improvements in the couple of months ahead.
Keep that feedback coming.