Difference between revisions of "NAMDPerformance"

From Point
Jump to: navigation, search
Line 3: Line 3:
 
NAMD is written in [http://charm.cs.uiuc.edu/ charm++] and thus has some unique attributes when profiled by TAU. For example the charm++ scheduler, which assigns task to processors and helps in load balancing the program, has a notion of Idling while waiting for tasks to complete. Thus TAU creates a event to capture time spent when the scheduler is in its idle state (Idle) as well as a event (Main) to account for the communication latencies. You can see how NAMD performs on different hardware with these charts:
 
NAMD is written in [http://charm.cs.uiuc.edu/ charm++] and thus has some unique attributes when profiled by TAU. For example the charm++ scheduler, which assigns task to processors and helps in load balancing the program, has a notion of Idling while waiting for tasks to complete. Thus TAU creates a event to capture time spent when the scheduler is in its idle state (Idle) as well as a event (Main) to account for the communication latencies. You can see how NAMD performs on different hardware with these charts:
  
 +
[Image:intrepid-ranger-breakdown.png]
  
Where on Intrepid (BlueGene P) Idle time (red) increases as NAMD scales on Ranger (Sun x86 cluster) Main time increses (blue). This shows how Rangers relatively slower communications layer results in larger latencies as NAMD scales.
+
Where on Intrepid (BlueGene P) Idle time (red) increases as NAMD scales on Ranger (Sun x86 cluster) Main time increases (blue). This shows how Rangers relatively slower communications layer results in larger latencies as NAMD scales compared to how NAMD scales on Intrepid.

Revision as of 23:47, 9 June 2009

NAMD Performance Study

NAMD is written in charm++ and thus has some unique attributes when profiled by TAU. For example the charm++ scheduler, which assigns task to processors and helps in load balancing the program, has a notion of Idling while waiting for tasks to complete. Thus TAU creates a event to capture time spent when the scheduler is in its idle state (Idle) as well as a event (Main) to account for the communication latencies. You can see how NAMD performs on different hardware with these charts:

[Image:intrepid-ranger-breakdown.png]

Where on Intrepid (BlueGene P) Idle time (red) increases as NAMD scales on Ranger (Sun x86 cluster) Main time increases (blue). This shows how Rangers relatively slower communications layer results in larger latencies as NAMD scales compared to how NAMD scales on Intrepid.