Contention on 2nd Level Cache May Limit the Effectiveness of Simultaneous Multithreading

نویسندگان

  • Sébastien Hily
  • André Seznec
چکیده

Simultaneous multithreading (SMT) is an interesting way of maximizing performance by enhancing processor utilization. We investigate issues involving the behavior of the memory hierarchy with SMT. First, we show that ignoring L2 cache contention leads to strongly overestimate the performance one can expect and may lead to incorrect conclusions. We then explore the impact of various memory hierarchy parameters. We show that the number of supported threads has to be setup according to the cache size, that the L1 caches have to be associative and small blocks have to be used. Then, the hardware constraints put on the design of memory hierarchies should limit the interest of SMT to a few threads. Les travaux de S ebastien Hily sont en partie nanc es par la r egion Bretagne La contention sur le second niveau de cache pourrait limiter l'eecacit e du multiiot simultan e R esum e : Le multiiot simultan e (SMT) est une voie int eressante pour augmenter les performances des microprocesseurs en am eliorant leur utilisation. Dans cette etude, nous evaluons le comportement de la hi erarchie m emoire plac ee dans une architecture supportant le multiiot simultan e. D'abord, nous montrons qu'ignorer la contention sur le cache de second niveau (ou la m emoire) am ene a largement surestimer les performances que l'on peut attendre et peut conduire a des conclusions fausses. Nous explorons ensuite l'impact de dii erents param etres de la hi erarchie m emoire. Nous montrons que le nombre de ots support e doit ^ etre en accord avec la taille des caches, que les caches de premier niveau doivent ^ etre associatifs et que les lignes de caches doivent ^ etres petites. Alors, les contraintes mat erielles impliqu ees par la mise en uvre d'une hi erarchie m e-moire devraient limiter l'int er^ et du multiiot simultan e a quelques ots.

برای دانلود 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

Standard Memory Hierarchy Does Not Fit Simultaneous Multithreading

Simultaneous multithreading (SMT) is a promizing approach in maximizing performance by enhancing processor utilization. We investigate issues involving the behavior of the memory hierarchy with SMT. First, we show that ignoring L2 cache contention leads to strongly overestimate the performance one can expect and may lead to incorrect conclusions. We then explore the impact of various memory hie...

متن کامل

Understanding the Impact of Inter-Thread Cache Interference on ILP in Modern SMT Processors

Simultaneous Multithreading (SMT) has emerged as an effective method of increasing utilization of resources in modern super-scalar processors. SMT processors increase instruction-level parallelism (ILP) and resource utilization by simultaneously executing instructions from multiple independent threads. Although simultaneously sharing resources benefits system throughput, coscheduled threads oft...

متن کامل

Predictable Fine-Grained Cache Behavior for Enhanced Simultaneous Multithreading (SMT) Scheduling

By converting thread-level parallelism to instruction level parallelism, Simultaneous Multithreaded (SMT) processors are emerging as effective ways to utilize the resources of modern superscalar architectures. However, the full potential of SMT has not yet been reached as most modern operating systems use existing single-thread or multiprocessor algorithms to schedule threads, neglecting conten...

متن کامل

Dynamic Helper Threaded Prefetching on the Sun UltraSPARC CMP Processor

Data prefetching via helper threading has been extensively investigated on Simultaneous MultiThreading (SMT) or Virtual Multi-Threading (VMT) architectures. Although reportedly large cache latency can be hidden by helper threads at runtime, most techniques rely on hardware support to reduce context switch overhead between the main thread and helper thread as well as rely on static profile feedb...

متن کامل

Improving Latency Tolerance of Multithreading through Decoupling

ÐThe increasing hardware complexity of dynamically scheduled superscalar processors may compromise the scalability of this organization to make an efficient use of future increases in transistor budget. SMT processors, designed over a superscalar core, are therefore directly concerned by this problem. This work presents and evaluates a novel processor microarchitecture which combines two paradi...

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

برای دانلود 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

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1997