Other Projects

From SuperTech Wiki

(Difference between revisions)
Jump to: navigation, search
(Created page with "Under construction.")
Line 1: Line 1:
-
Under construction.
+
= <br>Location-Based Memory Fences =
 +
 
 +
Traditional memory fences are program-counter (PC) based.  That is, a memory fence enforces a serialization point in the program instruction stream --- it ensures that all memory references before the fence in the program order have taken effect before the execution continues onto instructions after the fence.  Such PC-based memory fences always cause the processor to stall, even when the synchronization is unnecessary during a particular execution.  We propose the concept of '''location-based memory fences''', which aim to reduce the cost of synchronization due to the latency of memory fence execution in parallel algorithms.
 +
 
 +
Unlike a PC-based memory fence, a location-based memory fence serializes the instruction stream of the executing thread only when a different thread attempts to read the memory location which is guarded by the location-based memory fence.  In this work, we describe a hardware mechanism for location-based memory fences, prove its correctness, and evaluate its potential performance benefit.  Our experimental results are based on a software simulation of the proposed location-based memory fence, which incurs higher overhead than the proposed hardware mechanism would.  Even though applications using the software prototype implementation do not scale as well compared to the traditional memory fences due to the software overhead, our experiments show that applications can benefit from using location-based memory fences.  These results suggest that a hardware
 +
support for location-based memory fences is worth considering.

Revision as of 18:02, 16 May 2011


Location-Based Memory Fences

Traditional memory fences are program-counter (PC) based. That is, a memory fence enforces a serialization point in the program instruction stream --- it ensures that all memory references before the fence in the program order have taken effect before the execution continues onto instructions after the fence. Such PC-based memory fences always cause the processor to stall, even when the synchronization is unnecessary during a particular execution. We propose the concept of location-based memory fences, which aim to reduce the cost of synchronization due to the latency of memory fence execution in parallel algorithms.

Unlike a PC-based memory fence, a location-based memory fence serializes the instruction stream of the executing thread only when a different thread attempts to read the memory location which is guarded by the location-based memory fence. In this work, we describe a hardware mechanism for location-based memory fences, prove its correctness, and evaluate its potential performance benefit. Our experimental results are based on a software simulation of the proposed location-based memory fence, which incurs higher overhead than the proposed hardware mechanism would. Even though applications using the software prototype implementation do not scale as well compared to the traditional memory fences due to the software overhead, our experiments show that applications can benefit from using location-based memory fences. These results suggest that a hardware support for location-based memory fences is worth considering.

Personal tools
Members Only
Off Topic