Machine Learning for Result Estimation of Timing, Resource Usage, and Operation Delay in High Level Synthesis

In high level synthesis, acquiring of accurate result estimation becomes difficult in the earlier stage due to complex optimizations in the physical synthesis. Hence, there occurs a trade-off between efficiency, which involves evaluating in the HLS stage itself and accuracy, which points to waiting for post synthesis results by HLS tools. The factor of accuracy can be optimised using Machine Learning tools by learning from real benchmarks.

One set of parameters is Estimation of Timing, Resource Usage, and Operation Delay : The main methodology is to train an ML model that takes HLS reports as input and outputs a more accurate implementation report without conducting the time-consuming post-implementation.

The workflow can be divided broadly in two steps :

Data Processing :

Like every ML model, HLS estimation also requires training and testing data. The HLS and implementation reports are usually collected across individual designs by running each design through the complete C-to-bitstream flow, for various clock periods and targeting different FPGA devices. After that, one can extract features from the HLS reports as inputs and features from implementation reports as outputs. The data being handled here is of a huge quantity. In order to overcome the effect of collinearity and to reduce the dimension of the data and retain only the most significant parts, application of feature selection techniques is done. This removes the unimportant features from the data.

Selected features after dimensionality reduction


      Training the Estimation Models     

      After we have the refined dataset, regression models are trained to estimate post-implementation resource usages and clock periods. Frequently used metrics to report the estimation error include relative absolute error (RAE) and relative root mean squared error (RMSE). RAE and RMSE are favoured to be as low as possible. 

𝑦’  = vector of values predicted by the model,

𝑦 = vector of actual ground truth values in the testing set


y(bar) = mean value of y

N = number of samples

yi’ = predicted value of sample

yi = actual value of sample

The results are in terms of maximum clock frequency, throughput, and throughput-to-area ratio for the RTL code generated by the HLS tool.

Comments

  1. Aaditya Yawalkar10 June 2021 at 22:16

    Very Insightful!

    ReplyDelete
  2. Unknown10 June 2021 at 22:22

    Apart from estimation of the mentioned parameters, where else can it be used?

    ReplyDelete
    Replies
    1. Harshada Belgi10 June 2021 at 22:27

      It can also be used for cross-platform performance prediction.

      Delete

Post a Comment

Popular posts from this blog

FAST STATIC IR DROP PREDICTION USING MACHINE LEARNING

By Anmol Salvi INTRODUCTION: Logic Synthesis , High Level Synthesis , IR drop estimation and Lithography hotspot detection all fall under one broad category in a way which is Electronic Design Automation and are all related to each other to a certain extent.IR drop constraint is one of the fundamental requirements that is enforced in almost every chip design. However, it’s evaluation ends up taking a large amount of time and also the mitigation techniques for fixing any violations may require a number of iterations. Thus , fast and more importantly accurate IR drop prediction becomes essential for reducing the overall design turnaround time. Very recently, machine learning (ML) techniques have been studied very actively so as to allow for fast IR drop estimation owing to their potential promise and recorded success in a lot of fields. IR drop which is also called as voltage drop, is defined as the deviation of a power supply level from its specification which occurs when current begi...
Read more

Machine Learning for Analog Layout

By Ishan Aphale In back-end analog/mixed-signal (AMS) design flow, well generation persists as a fundamental challenge for layout compactness, routing complexity, circuit performance and robustness. The immaturity of AMS layout automation tools comes to a large extent from the difficulty in comprehending and incorporating designer expertise. To mimic the behavior of experienced designers in well generation, we can use a generative adversarial network (GAN) guided well generation framework with a post-refinement stage leveraging the previous high-quality manually-crafted layouts. Guiding regions for wells are first created by a trained GAN model, after which the well generation results are legalized through post-refinement to satisfy design rules. The network learns and mimics designers’ behavior from manual layouts. Experiments show that generated wells have comparable post-layout circuit performance with manual designs on the op-amp circuit. The power of machine learning algorithms ha...
Read more

High Level Synthesis using Linear Regression

Image
High Level Synthesis (HLS) works at the algorithmic level in hardware design systems. HLS tools provide automatic conversion from C/C++/SystemC-based specification to hardware description languages like Verilog or VHDL. It has been inferred that these HLS tools are improving the productivity in customised hardware design systems. In certain cases of HLS, the long synthesis time for each design brings about a restriction on design space exploration (DSE). DSE is the process of finding some design solution which optimally meets the design requirements from a space of tentative design points. This exploration is quite complex considering the various levels of abstraction and the entire process from selection of parameter values to choosing an algorithm and then optimising it. Machine Learning for High Level Synthesis: Machine Learning techniques are applied to improve the performance of HLS tools with 3 major advantages: 1. Fast and accurate result estimation 2. Refining the conventional ...
Read more