Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

PTAT

PTAT As the memory access footprints of applications in areas like data analytics increase, the latency overhead of translation lookaside buffer (TLB) misses increases. Thus, the efficiency of TLB becomes increasingly critical for overall system performance. Analyzing TLB miss traces is useful for hardware architecture design and software application optimization. Utilizing cycle-accurate simulators or instrumentation tools is very time-consuming and/or inaccurate for tracing and profiling TLB misses. In this article, we propose an efficient and precise tool to collect and profile last-level TLB misses. This tool utilizes a novel software method called Page Table Access Tracing (PTAT), storing last-level page table entries of certain workload processes into a reserved uncached memory region. Therefore, each last-level TLB miss incurred by user process corresponds to one uncached page table access to main memory, which can be captured and recorded by a hardware memory bus monitor. The detected information is then dumped into offline storage. In this manner, full TLB miss traces are collected and can be analyzed flexibly. Compared to previous software-based methods, this method achieves higher performance. Experiments show that, compared with a state-of-the-art kernel instrumentation method (BadgerTrap), which lacks complete dumping trace function, the speedup is still up to 3.88-fold for memory-intensive benchmarks. Due to the improved efficiency and completeness of tracing, case studies validate that more flexible profiling can be conducted, which is of great significance for TLB performance optimization. The accuracy of PTAT is verified by both dedicated sequence and performance counters. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ACM Transactions on Embedded Computing Systems (TECS) Association for Computing Machinery

PTAT

Download PDF
17 pages

Loading next page...
 
/lp/association-for-computing-machinery/ptat-yCmBUAZX0M

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Association for Computing Machinery
Copyright
Copyright © 2018 ACM
ISSN
1539-9087
eISSN
1558-3465
DOI
10.1145/3182174
Publisher site
See Article on Publisher Site

Abstract

As the memory access footprints of applications in areas like data analytics increase, the latency overhead of translation lookaside buffer (TLB) misses increases. Thus, the efficiency of TLB becomes increasingly critical for overall system performance. Analyzing TLB miss traces is useful for hardware architecture design and software application optimization. Utilizing cycle-accurate simulators or instrumentation tools is very time-consuming and/or inaccurate for tracing and profiling TLB misses. In this article, we propose an efficient and precise tool to collect and profile last-level TLB misses. This tool utilizes a novel software method called Page Table Access Tracing (PTAT), storing last-level page table entries of certain workload processes into a reserved uncached memory region. Therefore, each last-level TLB miss incurred by user process corresponds to one uncached page table access to main memory, which can be captured and recorded by a hardware memory bus monitor. The detected information is then dumped into offline storage. In this manner, full TLB miss traces are collected and can be analyzed flexibly. Compared to previous software-based methods, this method achieves higher performance. Experiments show that, compared with a state-of-the-art kernel instrumentation method (BadgerTrap), which lacks complete dumping trace function, the speedup is still up to 3.88-fold for memory-intensive benchmarks. Due to the improved efficiency and completeness of tracing, case studies validate that more flexible profiling can be conducted, which is of great significance for TLB performance optimization. The accuracy of PTAT is verified by both dedicated sequence and performance counters.

Journal

ACM Transactions on Embedded Computing Systems (TECS) – Association for Computing Machinery

Published: May 22, 2018

Keywords: TLB misses

References

  • The Graph 500 List
    Bader, D. A.; Feo, J.; Gilbert, J.; Kepner, J.; Koester, D.; Loh, E.; Madduri, K.; Mann, W.; Meuse, Theresa
  • Oprofile
    Cohen, Will; Suthikulpanit, Suravee; Deacon, Will; Allard, Gilles; Hansel, Daniel; Richter, Robert
  • HMTT: A platform independent full-system memory trace monitoring system
    Bao, Yungang; Chen, Mingyu; Ruan, Yuan; Liu, Li; Fan, Jianping; Yuan, Qingbo; Song, Bo; Xu, Jianwei
  • QEMU, a fast and portable dynamic translator
    Bellard, Fabrice
  • The GEM5 simulator
    Binkert, Nathan; Beckmann, Bradford M.; Black, Gabriel; Reinhardt, Steven K.; Saidi, Ali G.; Basu, Arkaprava; Hestness, Joel; Hower, Derek R.; Krishna, Tushar; Sardashti, Somayeh
  • A lightweight hybrid hardware/software approach for object-relative memory profiling
    Chen, Licheng; Cui, Zehan; Bao, Yungang; Chen, Mingyu; Huang, Yongbing; Tan, Guangming
  • CMD: classification-based memory deduplication through page access characteristics
    Chen, Licheng; Wei, Zhipeng; Cui, Zehan; Chen, Mingyu; Pan, Haiyang; Bao, Yungang
  • BadgerTrap: A tool to instrument x86-64 TLB misses
    Gandhi, Jayneel; Basu, Arkaprava; Hill, Mark D.; Swift, Michael M.
  • SPEC CPU2006 benchmark descriptions
    Henning, John L.
  • TopMC: An Approach for Understanding Architectural Characteristics (Technical Report)
    Huang, Yongbing
  • HMTT: A hybrid hardware/software tracing system for bridging the DRAM access trace’s semantic gap
    Huang, Yongbing; Chen, Licheng; Cui, Zehan; Ruan, Yuan; Bao, Yungang; Chen, Mingyu; Sun, Ninghui
  • HaLock: hardware-assisted lock contention detection in multithreaded applications
    Huang, Yongbing; Cui, Zehan; Chen, Licheng; Zhang, Wenli; Bao, Yungang; Chen, Mingyu
  • Intel-64 and IA-32 Architectures Software Developer’s Manual, Vol
    Intel,
  • Intel VTune Amplifier 2017
    Intel,
  • A look at several memory management units, TLB-refill mechanisms, and page table organizations
    Jacob, Bruce L.; Mudge, Trevor N.
  • Kibra 480 analyzer
    LeCroy, Teledyne
  • LPM: Concurrency-driven layered performance matching
    Liu, Yuhang; Sun, Xian-He
  • Reevaluating data stall time with the consideration of data access concurrency
    Liu, Yuhang; Sun, Xian-He
  • Evaluating the combined effect of memory capacity and concurrency for many-core chip design
    Liu, Yuhang; Sun, Xian-He
  • SIMICS: A full system simulation platform
    Magnusson, Peter S.; Christensson, Mattias; Eskilson, J.; Forsgren, Daniel; Hallberg, G.; Hogberg, J.; Larsson, F.; Moestedt, Andreas; Werner, B.
  • A comparison of counting and sampling modes of using performance monitoring hardware
    Moore, Shirley V.
  • Non-intrusive scalable memory access tracer
    Ohba, Nobuyuki; Munetoh, Seiji; Okazaki, Atsuya; Katayama, Yasunao
  • Performance, energy characterizations and architectural implications of an emerging mobile platform benchmark suite-MobileBench
    Pandiyan, Dhinakaran; Lee, Shin-Ying; Wu, Carole-Jean
  • MARSS: A full system simulator for multicore x86 CPUs
    Patel, Avadh; Afram, Furat; Chen, Shunfei; Ghose, Kanad
  • A simple synchronous distributed-memory algorithm for the HPCC randomaccess benchmark
    Plimpton, Steven J.; Brightwell, Ron; Vaughan, Courtenay; Underwood, Keith; Davis, Mike
  • Supertrace probe
    Software, Green Hills
  • Intel virtualization technology
    Uhlig, Rich; Neiger, Gil; Rodgers, Dion; Santoni, Amy L.; Martins, Fernando C. M.; Anderson, Andrew V.; Bennett, Steven M.; Kagi, Alain; Leung, Felix H.; Smith, Larry
  • TLB misses: The missing issue of adaptive radix tree? In Proceedings of the 11th International Workshop on Data Management on New Hardware
    Wong, Petrie; Feng, Ziqiang; Xu, Wenjian; Lo, Eric; Kao, Ben
  • The impulse memory controller
    Zhang, Lixin; Fang, Zhen; Parker, Mike; Mathew, Binu K.; Schaelicke, Lambert; Carter, John B.; Hsieh, Wilson C.; McKee, Sally A.