Exploration of Compiler Optimization Sequences Using a Hybrid Approach

Xavier, Tiago; Silva, Anderson Faustino da

doi:10.4149/cai_2018_1_165

Cited by 13 publications

(17 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, it is not necessary that a good sequence is also an optimal one [33]. Similar to previous works [3,15,11], this paper reports the performance speedup relative to LLVM's highest optimization level of -O3.…”

Section: Phase Ordering Problemsupporting

confidence: 60%

“…Several techniques [14,15,11,33] have been proposed for search space reduction in the given scenario, but the Mitigates the Compiler Phase-ordering (MiCOMP) has been selected in this paper, due to its systematic and reproducible approach for reducing the optimization space of those compilers, which exhibit the phase ordering problem. It works by clustering the LLVM's -O3 optimization passes into different groups (sub-sequences).…”

Section: Micompmentioning

confidence: 99%

“…(cBench) programs [13,5,16] are used, as described in Table 3. The evaluation is done in terms of percentage experiment reduction, percentage redundancy fraction, speedup, and percentage time improvement metrics represented by Equations ( 8), ( 9), (10), and (11).…”

Section: Benchmark and Performance Metricsmentioning

confidence: 99%

“…Several techniques [14,15,3,10,11] have been proposed for reducing the search space in LLVM. However, these techniques consider the identical codes as separate optimization, hence increasing the overall search space due to redundant testings.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

REDUCER: Elimination of Repetitive Codes for Accelerated Iterative Compilation

Ahmed

Ismail

2021

cai

View full text Add to dashboard Cite

Low Level Virtual Machine (LLVM) is a widely adopted open source compiler providing numerous optimization opportunities. The discovery of the best optimization sequence in this large space is done via iterative compilation, which incurs substantial overheads, especially for big data applications operating on high volume and variety datasets. The large search space is mostly comprised of identical codes generated via different optimizations. However, no mechanism is implemented inside the LLVM compiler to suppress the redundant testings. In this regard, this paper proposes REDUCER for eliminating the identical code executions by performing Intermediate Representation (IR) level comparisons. REDUCER has been tested using the well-accepted MiCOMP technique in LLVM 3.8 and 9.0 compiler, with embedded (cBench) and big data workloads. In comparison to MiCOMP 19.5 k experiments, REDUCER lowers the experiment count up to 327, i.e. 98 %, and on average to 4 375, i.e. 77 %, for cBench (LLVM-3.8). Similarly, for LLVM-9.0 the reductions are up to 1 931, i.e. 90 %, and on average 5 863, i.e. 69.9 %. Due to the significant experiment reduction, for embedded workloads, the iterative compilation is up to 58.6× and on average 4.1× faster with REDUCER (LLVM-3.8) than Mi-COMP, whereas, with REDUCER (LLVM-9.0) the compilation is up to 8.5× and on average 2.9× faster. Moreover, REDUCER is found to be scalable and efficient for big data workloads where the iterative compilation is reduced to few days, as code is compared one time only for a single application tested on multiple datasets.

show abstract

Section: Phase Ordering Problemsupporting

confidence: 60%

Section: Micompmentioning

confidence: 99%

Section: Benchmark and Performance Metricsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

REDUCER: Elimination of Repetitive Codes for Accelerated Iterative Compilation

Ahmed

Ismail

2021

cai

View full text Add to dashboard Cite

show abstract

“…Storing String, and Objects in ROM requires to enable ROM Global inherit and disabling Hstring Array index feature 11,14 Disabling reference counting (garbage collection) requires to disable the use of double linked heap.…”

Section: Dependencies Among Js Interpreter Featuresmentioning

confidence: 99%

MoMIT: Porting a JavaScript Interpreter on a Quarter Coin

Morales

Saborido

Guéhéneuc

2021

IIEEE Trans. Software Eng.

View full text Add to dashboard Cite

The Internet of Things (IoT) is a network of physical, heterogeneous, connected devices providing services through private networks and the Internet. It connects a range of new devices to the Internet so they can communicate with Web servers and other devices around the world. Today's standard platform for communicating Web pages and Web apps is JavaScript (JS) and extending the same standard platform to connect IoT devices seems more than appropriate. However, porting JS applications to the large variety of IoT devices, specifically on System-on-a-Chip (SoCs) devices (e.g., Arduino Uno, Particle Photon), is challenging because these devices are constrained in terms of memory and storage capacity. Running JS applications adds an overhead of resources to deploy a code interpreter on the devices. Also, running JS applications may not be possible "as is" on some device missing some hardware/software capabilities. To address this problem, we propose MoMIT a multiobjective optimization approach to miniaturize JS applications to run on IoT constrained devices. To validate MoMIT, we miniaturize a JS interpreter to execute a testbed comprised of 23 applications and measure their performances before and after applying the miniaturization process. We implement MoMIT using three different search algorithms and found that it can reduce code size, memory usage, and CPU time by median values of 31%, 56%, and 36% respectively. Finally, MoMIT ported the miniaturized JS interpreters up to to 2 SoCs additional devices, in comparison of using default JS interpreter features.

show abstract

Toward a novel engine for compiler optimization space exploration of big data workloads

Ahmed

Ismail

2021

Softw Pract Exp

View full text Add to dashboard Cite

Recently, big data specific technologies have been emerging, including domain-specific languages, software frameworks, databases, third-party libraries, and so forth. These techniques are successful in concealing the low-level details by producing high-level code, which is passed through the conventional compilation cycle for generating hardware operable code. Several optimization opportunities exist in the compiler which can assist in meeting the processing deadlines of big data workloads, through optimized machine code. However, the existing iterative compilation techniques are not enough for the exploration of big data applications. In this regard, a novel engine has been presented for exploiting the compiler optimization space of big data workloads. The engine is comprised of training and testing phases. During the training stage, the big data application is optimized with Mitigates the Compiler Phase-ordering (MiCOMP) and genetic algorithm (GA) optimization sequences, which are executed with train datasets. In the testing stage, the test datasets are executed only for the best 300 optimization sequences discovered at the training stage. The proposed engine has been tested with graph mining, machine learning, and text search categories of big data applications using a wide range of real-world and synthetic datasets. Overall, the engine is 56.8×, 47×, and 9.8× faster than Iterative Optimization for the Data Center (IODC), MiCOMP, and GA respectively in exploiting the compiler search space for big data workloads. Further, the integration of best-10 and best-3 techniques with the engine brings a speedup of 5.9× and 7.8×. The compiler level exploitation of general-purpose machines incurs no extra overhead, no heavy computing, and no personnel cost. Also, the overall performance of big data specialized software solutions can be enhanced by compiling their high-level code with suitable compiler optimizations.

show abstract

Exploration of Compiler Optimization Sequences Using a Hybrid Approach

Cited by 13 publications

References 15 publications

REDUCER: Elimination of Repetitive Codes for Accelerated Iterative Compilation

REDUCER: Elimination of Repetitive Codes for Accelerated Iterative Compilation

MoMIT: Porting a JavaScript Interpreter on a Quarter Coin

Toward a novel engine for compiler optimization space exploration of big data workloads

Contact Info

Product

Resources

About