Predicting Cache Contention with Setvectors

نویسندگان

  • Michael Zwick
  • Florian Obermeier
  • Klaus Diepold
چکیده

In this paper, we present a new method called setvectors to predict cache contention introduced by co-scheduled applications on a multicore processor system. Additionally, we propose a new metric to compare cache contention prediction methods. Applying this metric, we demonstrate that our setvector method predicts cache contention with about the same accuracy as the most accurate stateof-the-art method. However, our method executes nearly 4000 times as fast.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Proceedings of the First Workshop on Co - Scheduling of HPC Applications ( COSH 2016 ) Prague , Czech Republic , January 19 , 2016 Co - located with HiPEAC 2016 Workshop Co - Chairs

In this paper we present a resource-centric application classification approach that monitors data flow along the path from main memory to the cores to locate spots of high resource utilization and potential resource contention. We designate three application classes, i.e. streaming applications, last-level cache sensitive applications and applications that restrict their activity either within...

متن کامل

A resource-centric Application Classification Approach

In this paper we present a resource-centric application classification approach that monitors data flow along the path from main memory to the cores to locate spots of high resource utilization and potential resource contention. We designate three application classes, i.e. streaming applications, last-level cache sensitive applications and applications that restrict their activity either within...

متن کامل

A methodology for detailed performance modeling of reduction computations on SMP machines

In this paper, we revisit the problem of performance prediction on SMP machines, motivated by the need for selecting parallelization strategy for random write reductions. Such reductions frequently arise in data mining algorithms. In our previous work, we have developed a number of techniques for parallelizing this class of reductions. Our previous work has shown that each of the three techniqu...

متن کامل

AutoCon: Regression Testing for Detecting Cache Contention Anomalies Using Autoencoder

Cache contention is an important type of performance anomaly in this multi-core and many-core era. It can cause a significant slowdown in parallel programs. However, it is hard to detect and often, not visible in the source code. As software changes over time, modifications in code can introduce cache contention anomalies. One way to detect such anomalies, is to use performance regression testi...

متن کامل

ذخیره 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 2010