Taming hardware event samples for FDO compilation

Chen, Dehao; Vachharajani, Neil; Hundt, Robert; Liao, Shih-Wei; Ramasamy, Vinodha; Yuan, Paul; Chen, Wenguang; Zheng, Weimin

doi:10.1145/1772954.1772963

Cited by 59 publications

(32 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, while Feedback Directed Optimization (FDO) can provide substantial performance gains, the extra step involved in the programmers workflow has stopped this promising technique from being widely adopted [13]. By eliminating any extra steps, we believe that SiblingRivalry can bring autotuning to the mainstream program optimization.…”

Section: Discussionmentioning

confidence: 99%

Siblingrivalry

Ansel

Pacula

Wong

et al. 2012

Proceedings of the 2012 International Conference on Compilers, Architectures and Synthesis for Embedded Systems

View full text Add to dashboard Cite

Modern high performance libraries, such as ATLAS and FFTW, and programming languages, such as PetaBricks, have shown that autotuning computer programs can lead to significant speedups.However, autotuning can be burdensome to the deployment of a program, since the tuning process can take a long time and should be rerun whenever the program, microarchitecture, execution environment, or tool chain changes. Failure to re-autotune programs often leads to widespread use of sub-optimal algorithms. With the growth of cloud computing, where computations can run in environments with unknown load and migrate between different (possibly unknown) microarchitectures, the need for online autotuning has become increasingly important.We present SiblingRivalry, a new model for alwayson online autotuning that allows parallel programs to continuously adapt and optimize themselves to their environment. In our system, requests are processed by dividing the available cores in half, and processing two identical requests in parallel on each half. Half of the cores are devoted to a known safe program configuration, while the other half are used for an experimental program configuration chosen by our self-adapting evolutionary algorithm.When the faster configuration completes, its results are returned, and the slower configuration is terminated. Over time, this constant experimentation allows programs to adapt to changing dynamic environments and often outperform the original algorithm that uses the entire system.

show abstract

Section: Discussionmentioning

confidence: 99%

Siblingrivalry

Ansel

Pacula

Wong

et al. 2012

Proceedings of the 2012 International Conference on Compilers, Architectures and Synthesis for Embedded Systems

View full text Add to dashboard Cite

show abstract

“…This speedup is due to a series of components in the framework. Initially, using traditional PMU-based sampling to collect frequency profile, as described in [9], approximately 4.35% speedup can be achieved. If we use LBR to collect the profile, the speedup increases to 7.38%.…”

Section: Over/under Sampling Due To Optimizationmentioning

confidence: 99%

Taming Hardware Event Samples for Precise and Versatile Feedback Directed Optimizations

Chen

Vachharajani

Hundt

et al. 2013

IEEE Trans. Comput.

Self Cite

View full text Add to dashboard Cite

Abstract-Feedback-directed optimization (FDO) is effective in improving application runtime performance, but has not been widely adopted due to the tedious dual-compilation model, the difficulties in generating representative training data sets, and the high runtime overhead of profile collection. The use of hardware-event sampling overcomes these drawbacks by providing a lightweight approach to collect execution profiles in the production environment, which naturally consumes representative input. Yet, hardware event samples are typically not precise at the instruction or basic-block granularity. These inaccuracies lead to missed performance when compared to instrumentation-based FDO. In this paper, we use Performance Monitoring Unit (PMU) based sampling to collect the instruction frequency profiles. By collecting profiles using multiple events, and applying heuristics to predict the accuracy, we improve the accuracy of the profile. We also show how emerging techniques can be used to further improve the accuracy of the sample-based profile. Additionally, these emerging techniques are used to collect value profiles, as well as to assist a lightweight inter-procedural optimizer. All these profiles are represented in a portable form, thus they can be used across different platforms. We demonstrate that sampling-based FDO can achieve an average of 92% of the performance gains obtained using instrumentation-based exact profiles for both SPEC CINT2000 and CINT2006 benchmarks. The overhead of collection is only 0.93% on average, while compiler-based instrumentation incurs 2.0%-351.5% overhead (and 10x overhead on an industrial web search application).

show abstract

“…Although Chen [10] presented how hardware event samples can be used for FDO of general programs and GWP [23] profiles have provided performance insights for datacenter applications, not much has been studied about other applications of FDO on datacenter applications. As instant profiling enables various types of profiles to be collected continuously from datacenters, we expect this information to contribute to improving the performance of cloud computing.…”

Section: Future Workmentioning

confidence: 99%

Instant profiling: Instrumentation sampling for profiling datacenter applications

Cho

Moseley

Hank

et al. 2013

Proceedings of the 2013 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)

View full text Add to dashboard Cite

Profile-guided optimization possesses huge potential to save costs for datacenters. Hardware performance monitoring units enable profiling with negligible overhead and they have been proven to be effective to help programmers find code regions to optimize by monitoring datacenter applications continuously on live traffic. However, these hardware features are inflexible and often buggy, limiting the types of data that can be gathered. Instrumentation-based profiling can complement or replace hardware functionality by providing more flexible and targeted information gathering. Unfortunately, the overhead of existing instrumentation mechanisms prevents their use in production runs. In order to be used in datacenters, we need a profiling mechanism to impose overheads of less than a few percent, in terms of both throughput and latency, while still generating meaningful profile data. This paper presents instant profiling, an instrumentation sampling technique using dynamic binary translation. Instead of instrumenting the entire execution, instant profiling periodically interleaves native execution and instrumented execution according to configurable profiling duration and frequency parameters. It further reduces the latency degradation of initial profiling phases by pre-populating a software code cache. We evaluate the performance and effectiveness of this new profiling technique on the SPEC CINT2006 benchmark suite and two datacenter application benchmarks. We show that it is well-suited for deployment Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. CGO '13 23-27 February 2013, Shenzhen China. 978-1-4673-5525-4/13/$31.00 c 2013 to datacenters by incurring less than 6% slowdown and 3% computational overhead on average.

show abstract

Taming hardware event samples for FDO compilation

Cited by 59 publications

References 19 publications

Siblingrivalry

Siblingrivalry

Taming Hardware Event Samples for Precise and Versatile Feedback Directed Optimizations

Instant profiling: Instrumentation sampling for profiling datacenter applications

Contact Info

Product

Resources

About