Saturday, May 24, 2008

Instrumentierung – die Chance für Linux?

My latest article for the German publication Linux Technical Review appears in Volume 8 on "Performance und Tuning" and discusses a possible roadmap for future Linux instrumentation and performance management capabilities. The Abstract reads:
  • German: Linux könnte seine Position im Servermarkt ausbauen, wenn es dem Vorbild der Mainframes folgte und deren raffiniertes Performance-Management übernähme.
  • English: Linux could be in a position to expand its presence in the server market by looking to mainframe computer performance management as a role model and adapting its instrumentation accordingly.

Topics discussed include a comparison of time-share scheduling (TSS) with fair-share scheduling (FSS) and the Linux Completely fair scheduler (CFS), how to achieve a more uniform interface to performance and capacity planning measurements, and the kind of advanced system management capabilities available on IBM System Z for both their mainframes and clusters.

1 comment:

harry said...

point 1 I wonder at the omission of cache optimisation. When one assumes that all linux boxes contain at least two (virtual) processors and probably 4*2, then the optimisation of the use af cache and also the minimisation of cache refresh seems relevant. In experiments on an 8 way processor in 1998 we found that maintaining affinity affected the performance visibly. (this was in a transaction oriented server).
point 2: Also we found in the same period that in data intensive operations the major safeguard for maintaining correct responsetimes while running background batches was to have all I/O running under the priority of the originating process.
point3: I wonder if there are more recent experiments , especially when two dispatching streams in the same physical processor are allocated by preference to subtasks from the mother task.

kind regards