This portal is to open public enhancement requests against IBM System Storage products. To view all of your ideas submitted to IBM, create and manage groups of Ideas, or create an idea explicitly set to be either visible by all (public) or visible only to you and IBM (private), use the IBM Unified Ideas Portal (https://ideas.ibm.com).
We invite you to shape the future of IBM, including product roadmaps, by submitting ideas that matter to you the most. Here's how it works:
Start by searching and reviewing ideas and requests to enhance a product or service. Take a look at ideas others have posted, and add a comment, vote, or subscribe to updates on them if they matter to you. If you can't find what you are looking for,
Post an idea.
Get feedback from the IBM team and other customers to refine your idea.
Follow the idea through the IBM Ideas process.
Welcome to the IBM Ideas Portal (https://www.ibm.com/ideas) - Use this site to find out additional information and details about the IBM Ideas process and statuses.
IBM Unified Ideas Portal (https://ideas.ibm.com) - Use this site to view all of your ideas, create new ideas for any IBM product, or search for ideas across all of IBM.
ideasibm@us.ibm.com - Use this email to suggest enhancements to the Ideas process or request help from IBM for submitting your Ideas.
Due to processing by IBM, this request was reassigned to have the following updated attributes:
Brand - Servers and Systems Software
Product family - IBM Spectrum Scale
Product - Spectrum Scale (formerly known as GPFS) - Private RFEs
Component - Product functionality
For recording keeping, the previous attributes were:
Brand - Servers and Systems Software
Product family - IBM Spectrum Scale
Product - Spectrum Scale (formerly known as GPFS) - Private RFEs
Component - Technical Foundation
As noted in a previous comment, when GPFS uses the trace subsystem provided by the OS (which is the case on AIX and Windows), there's not much that can be done in GPFS about the trace overhead. On Linux, GPFS uses its own custom trace subsystem. The need to keep the trace overhead low is very well understood and appreciated. There have been successive improvements in this area in GPFS versions V3.3, V3.4, V3.5, and V4.1. We'll continue to work on further improvements, although at this point the low-hanging fruit has been picked, and the law of diminishing returns will be a major factor.
The basic reason GPFS trace collection is so resource-intensive is the level of code complexity. In order to record the events with enough detail to allow effective troubleshooting, data needs to be records at the rate of megabytes per second. It is very hard to do something like this with near-zero overhead under any circumstances. The verbosity of tracing can be controlled at a very fine level already, by varying individual trace levels. Obviously, it's possible to get the tracing overhead to a lower level by making the trace less verbose, but just as obviously this would make the trace less useful. So there's a need for a delicate balancing act, in keeping the trace verbose enough to allow troubleshooting while keeping the overhead to a reasonable level.
On AIX, GPFS uses built-in trace facility in AIX OS. So this request should be routed to the AIX team to enhance performance of current AIX trace facility.
Creating a new RFE based on Community RFE #54499 in product GPFS.