Memory Allocation Tracing with VampirTrace
نویسندگان
چکیده
The paper presents methods for instrumentation and measurement of applications’ memory allocation behavior over time. It provides some background about possible performance problems related to memory allocation as well as to memory allocator libraries. Then, different methods for data acquisition and representation are discussed. Finally, memory allocation tracing integrated in VampirTrace is demonstrated with a real-world HPC example application from aerodynamical simulation and optimization.
منابع مشابه
Lightweight Memory Tracing
Memory tracing (executing additional code for every memory access of a program) is a powerful technique with many applications, e.g., debugging, taint checking, or tracking dataflow. Current approaches are limited: software-only memory tracing incurs high performance overhead (e.g., for Libdft up to 10x) because every single memory access of the application is checked by additional code that is...
متن کاملWMTools - Assessing Parallel Application Memory Utilisation at Scale
The divergence between processor and memory performance has been a well discussed aspect of computer architecture literature for some years. The recent use of multi-core processor designs has, however, brought new problems to the design of memory architectures as more cores are added to each successive generation of processor, equivalent improvement in memory capacity and memory sub-systems mus...
متن کاملTracing ischemic memory by metabolic pathways: BMIPP and beyond
Myocardial ischemia (MI) resulting in infarction is an important cause of mortality and morbidity worldwide. Acute ischaemia rapidly impairs myocardial contractile function. Myocardial dysfunction persisting for several hours after transient non-lethal ischaemia, eventually resulting in full functional recovery is termed as myocardial stunning. Hibernation is now thought to be...
متن کاملRay Tracing with the Single Slab Hierarchy
Bounding volume hierarchies have become a very popular way to speed up ray tracing. In this paper we present a novel traversal and approximation scheme for bounding volume hierarchies, which is comparable in speed, has a very compact traversal algorithm and uses only 25% of the memory, compared to a standard bounding volume hierarchy.
متن کاملParallel Progressive Ray-tracing
A dynamic task allocation algorithm for ray-tracing by progressive refinement on a distributed-memory parallel computer is described. Parallelization of progressive ray-tracing is difficult because of the inherent sequential nature of the sample location generation process, which is optimized (and different) for any given image. We report on experimental results obtained from our implementation...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2007